ELNET 2013 Proceedings of the 10th Workshop

Faculty of Electrical Engineering and Computer Science ˇ – Technical University of Ostrava VSB ISBN 978–80–248–3254–8

ELNET 2013 http://www.cs.vsb.cz/elnet/

10th Workshop Ostrava, 26th November 2013 Proceedings of papers

Organized by ˇ – Technical University of Ostrava VSB Faculty of Electrical Engineering and Computer Science

ELNET 2013 c Zdenˇek Hrad´ılek, editor

ISBN 978–80–248–3254–8

This work is subject to copyright. All rights reserved. Reproduction or publication of this material, even partial, is allowed only with the editors’ permission.

Technical editors: Peter Chovanec [email protected] Michal Kr´ atk´ y [email protected] Faculty of Electrical Engineering and Computer Science, ˇ – Technical University of Ostrava VSB

Page count: Impression: Edition: First published:

80 100 1st 2013

This proceedings was typeset by PDFLATEX.

ˇ – Technical Published by Faculty of Electrical Engineering and Computer Science, VSB University of Ostrava

Preface

ˇ The conference ELNET 2013 was held on 26th November 2013 at VSB-Technical University of Ostrava, Czech Republic. This is the tenth conference. The conception of ELNET conferences was a response to increasing interest in Energy and Power Systems and related aspects in the Czech Republic and Slovakia, in the last few years. An important point is the interdisciplinary nature of key topics of the conference: – – – – – – – –

Energy and Power Systems Distributed Power Generation Fault Diagnosis Power Breakdown Analysis Survivable Network System Analysis Energy Data Storing and Analysis Visualisation Structure and Grow of Networks

ELNET is a workshop intended for meeting of promoters of Energy and Power Systems and related aspects. It is focused on theoretical and technical foundations of information technologies, time-proven methods and development trends. It also serves as a place for discussion about new ideas. Conference provided an excellent opportunity for faculty, scholars, and practitioners to meet renowned researchers and to discuss innovative ideas, results of research, and best practices on various conference topics. I would like to cordially thank the authors and PC members for their effort, materialised in this volume. Special thanks go to the Organising Committee members for their arduous editing work. In conclusion, I would like to thank all contributing authors for their excellent research papers.

November 2013

Zdenˇek Hrad´ılek Program Committee Chair ELNET 2013

Organization

Evaluation Committee

Chair: ˇ – Technical University of Ostrava, Czech Republic) Zdenˇek Hrad´ılek (VSB Members: ˇ – Technical University of Ostrava, Czech Republic) V´ aclav Sn´ aˇsel (VSB ˇ – Technical University of Ostrava, Czech Republic) Stanislav Rusek (VSB Aleˇs Hor´ ak (Masaryk University in Brno, Czech Republic)

Program Committee

Anna Gawlak (Technical University Czestochowa, Poland) ˇ – Technical University of Ostrava, Czech Republic) Radom´ır Goˇ no (VSB Przemyslaw Janik (Technical University Wroclaw, Poland) Michal Kolcun (Technical University Koˇsice, Slovak Republic) Zbigniew Leonowicz (Technical University Wroclaw, Poland) Zbynˇek Martnek (University of West Bohemia, Czech Republic) Harald Schwarz (BTU Cottbus, Germany) Jerzy Szkutnik (Technical University Czestochowa, Poland) Petr Toman (VUT Brno, Czech Republic) Ladislav Varga (Technical University Koˇsice, Slovak Republic) ˇ Jiˇr´ı T˚ uma (CVUT Praha, Czech Republic)

Organizing Committee ˇ – Technical University of Ostrava, Czech Republic) Peter Chovanec (VSB ˇ Michal Kr´ atk´ y (VSB – Technical University of Ostrava, Czech Republic) ˇ – Technical University of Ostrava, Czech Republic) Yveta Geletiˇcov´ a (VSB

VII

Workshop Location: ˇ – Technical University of Ostrava Campus of VSB 17. listopadu 15, 708 33 Ostrava–Poruba, Czech Republic 26th November 2013 http://www.cs.vsb.cz/elnet/

VIII

Sponsor

ˇ Workshop ELNET 2013 is supported by Skupina CEZ.

http://www.cez.cz/

Table of Contents

Biogas station - operating measurement of electrical quantities . . . . . . . . . . Jiˇr´ı Janˇsa, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

1

Analysis of the results of methodology for 110 kV power lines restoration V´ıt Houdek, Tom´ aˇs Mozdˇreˇ n, Stanislav Rusek, Radom´ır Goˇ no

7

Possible reasons for the complaints to power quality at the point of renewable energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Petr Rozehnal, Jan Unger, Petr Krejˇc´ı

13

Maximum Extreme States, Annual Production of Photovoltaic Power Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Martin Smoˇcek, Zdenˇek Hrad´ılek

20

Problems of connecting of SHPs to distribution system and legislative changes to support RES after 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Martin Nov´ ak, Radom´ır Goˇ no

26

Actual Results of the Reliability Computation in 2013 . . . . . . . . . . . . . . . . . Martin Slivka, Radom´ır Goˇ no, Stanislav Rusek, Michal Kr´ atk´y, Vladimir Kral

32

Design, realization and analysis of Measuring Heat Pump Energy Balance ˇ amek, Zdenˇek Hrad´ılek Jan Sr´

39

Improved Physical Design of Outage Database . . . . . . . . . . . . . . . . . . . . . . . . Peter Chovanec, Michal Kr´ atk´y, Pavel Bedn´ aˇr

46

Automatic Consumption Optimization with regard to the Green Premium Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Miroslav Pr´ymek, Aleˇs Hor´ ak, Luk´ aˇs Prokop

55

Accumulation of Electrical Energy from Solar Power . . . . . . . . . . . . . . . . . . Jan Vacul´ık, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

60

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

68

X

Biogas station – operating measurement of electrical quantities Jiří Janša1, Zdeněk Hradílek2, Petr Moldřík3

Biogas station - operating measurement of 1

VŠB – TU Ostrava, Katedra elektroenergetiky, 17. listopadu 15, 708 33 Ostrava, electrical quantities http://fei1.vsb.cz/kat410/ tel: +420 597 329 321, email: [email protected], 2 VŠB – TU Ostrava, listopadu 15, 708 33 Ostrava, Jiˇr´ıKatedra Janˇsa, elektroenergetiky, Zdenˇek Hrad´ılek,17. Petr Moldˇr´ık http://fei1.vsb.cz/kat410/ of email: Electrical Power Engineering, tel: +420Department 597 321 235, , [email protected] ˇ – elektroenergetiky, FEECS,Katedra VSB Technical University of Ostrava, VŠB – TU Ostrava, 17. listopadu 15, 708 33 Ostrava, 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic http://fei1.vsb.cz/kat410/ [email protected], [email protected], [email protected] tel: +420 597 329 320, email: [email protected] 3

Abstract. This article deals with the evaluation of the measured operating electrical quantities of biogas station. The introduction describes the technology biogas plant. Subsequently, described the measurement of electrical quantities. In the second part evaluates the measured data with the application of mathematical and statistical methods. The greatest emphasis is placed on evaluating the progress and quality of power supply. Another important parameter is the process of total active power supplied by all cogeneration unit units to 22kV power grid. The analysis result through changes in the voltage depending on the instantaneous power supplied by biogas plants to the grid. More detailed analyzes will be carried out in further research.

1

Introduction

Biogas station in which the measurement was carried out is located in the territory of the Moravian-Silesian region. Biogas station is situated in the premises of agricultural enterprise concerned with pork farming and processes mainly maize silage and pig slurry. Primary reason was the source of slurry for use liquids for wet fermentation. One of the other reasons of this location was possibility of using waste heat for heating the adjacent pigsties, administrative building and in the summer in a newly built postharvest unit. Biogas station has an installed capacity in the amount of 1 090 kW and 1 080 kW thermal output. To the conversion of biogas into electrical energy are used four cogeneration unit. Three identical diesel engines with an output of 250 kWe and ones petrol engine with an output of 340 kWe. The electric power is fed to the overhead lines of distribution network 22kV through transformer station 0,4/22kV equipped with one transformer with maximum output of 1250 kVA and the nominal current of 1 804 A.

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 1–6, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

2

2

Jiˇr´ı Janˇsa, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

Measurement

The measurement was implemented in late spring and summer of the year 2013, namely since May 20, 2013 (8:25) until June 28, 2013 (8:25), thus for 39 days. The measurement was carried out by automatically operating a digital measuring device MDSU (monitor distribution network), which was measured at one-minute intervals and saved to memory rms phase voltage, current and power factor of each phase. The remaining electrical quantities, i.e. active, reactive and apparent power were calculated. The voltage was measured directly on bus bars in the switchgear and currents using current clamp converters MT-UNI using already installed instrument transformers current 1 500/5 A.

3

Evaluation of the measured data

During measurement was obtained in total 56 161 record lines. Inasmuch as in used software MS Excel is the maximum number of data points in a data series in twodimensional graphs of the 32 000, below are graphs of measured, respectively calculated values presented in two ways. Either waveform is divided into two separate graphs for about two halves of time 39 days (20.5. – 11.6. and 11.6. – 28.6.), which were created from minute samples measured values, or the waveform presented in a single graph, which created from calculated as the arithmetic average of the 15-minute intervals in all samples.

3.1

Electric power

The graph in Figure 1 presents the overall progress of active power all cogeneration units for the entire 39 days of the measurement. The power (P) is the sum of active power in each phase (P1 + P2 + P3). The graphs shows that over duration of the measurement was seven outages in the supply of power cogeneration units to the network. Total duration of outages is 151 minutes, i.e. 2 hours and 31 minutes, which represents only 0.3 % of the total operating time of cogeneration units. The duration of the last three outages is only on the order of minutes (8, 5 and 6 minutes). In addition to these failures occurred on the same day (20.6.) in relatively quick succession (between the hours of 12:48 to 3:21 p.m.). Further in Figure 1, except for the aforementioned failures indicated by date range for the most significant declines in the output of active power.

Biogas station - operating measurement of electrical quantities

3

Fig. 1 – Total active power supplied by all CHP (created from 15-minute average values)

Fig. 2 – Total active power supplied by all CHP between 25.5th and 28.5th (detail of Fig. 1) The figure 2 shows a detail of the total active power during selected for four day (from 25.5th to 28.5th), during which there were significant short- and medium-term declines in delivered power (about 250 to 500 kW). These decreases were due to failure of one or two cogeneration units.

4

3.2

Jiˇr´ı Janˇsa, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

Voltage

In terms of supply voltage deviation (in all three phases) are not exceeded the limits of ±10% Un claimed standard, which corresponds to values in the range 207-253 V. Due to long-term operational overvoltage changing the voltage, a step consisting in reduction of about 5 V, which was implemented supplier of cogeneration units. In the figures 3-5 are shown voltages histograms of each phase, i.e. the number (frequency) of the voltage in each class (range of values). Due to step voltage reduction of about 5 V, implemented on 18.6th, it was necessary to divide histograms of the period before change (20.5th – 17.6th) and after the change (19.6th – 28.6th). Data from the affected day (18.6th), as well as data on duration of outages are not included in the histograms. For the purposes of presentation in this article were selected only histograms from the period before the change in voltage due to increased data content.

Fig. 3 – Histogram voltage phase 1 – for the period between 20.5. and 17.6.

Fig. 4 – Histogram voltage phase 2 – for the period between 20.5. and 17.6.

Biogas station - operating measurement of electrical quantities

5

Fig. 5 – Histogram voltage phase 3 – for the period between 20.5. and 17.6. The histograms shows the distribution of voltage in all phases corresponds almost Gaussian probability distribution with more than 50% of the values lying in the range 238-240 volts for the second and third phase, and 239-241 in the first phase. After the jump the voltage dropped to about 230 to 233 V.

Fig. 6 – Current waveform in medium-conductor (created from 15-minute average values) The figure 6 shows waveform of the amplitude of the calculated current in medium conductor, which indicated to us unbalanced load phases. This asymmetry was obvious from the voltages histograms for each phase.

6

3.3

Jiˇr´ı Janˇsa, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

Power factor

In the figure 7 shows the waveform of power factor for each phase of the first half of monitored period. Without taking into account the periods of time when there was a outages in the power supply, then the power factor for the entire period of 39 days has always moved around a value of 0.99. Most of the time worked cogeneration units with an active power about 1040 kW (i.e. 96% of installed capacity) and reactive power in the range of about 180 to 25 kvar corresponding factor from 0.985 to 0.999.

Fig. 7 – Power factor waveform between 20.5. and 11.6.

4

Conclusion

In the context of this article were summarized interim results of research into the study of the impact on the quality of biogas voltage 22 kV distribution network. Over the entire period of measurement respond voltage range the indicated norm ČSN EN 50 160. From the measured data results high reliability of power to the grid (time outages is 0,3% of operating hours). As part of this research will be further and further data obtained subjected to detailed mathematical statistic examination.

Acknowledgement This work was supported by the Ministry of Education, Youth and Sports of the Czech Republic (No. SP2013/137), by the Czech Science Foundation (No. GAČR 102/09/1842) and by the project ENET (Research and Development for Innovations Operational Programme No. CZ.1.05/2.1.00/03.0069).

Analysis of the results of methodology for 110 Analysis of the results of methodology for 110 kV kV power lines restoration power lines restoration

V´ıt Houdek, Tom´ aˇs Mozdˇreˇ n, Stanislav Rusek, Radom´ır Goˇ no Vit Houdek, Tomas Mozdren, Stanislav Rusek, Radomir Gono

Department of Electrical Power Engineering, Department of Electrical Power Engineering ˇ – Technical FEECS, VSB University of Ostrava, VŠB – Technical University of Ostrava, 17.listopadu 15, 708 33 Ostrava - Poruba 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected] [email protected] Abstract. The paper deals with the restoration method for 110 kV power lines. The input data are the database of faults, the effect of failure and distributed power. The methodology is based on the principles of reliability centered maintenance (RCM). Two versions of methodology are described. The results and analysis of the first version of the method were used to modify the second version of restoration method. These modifications are described in the article, including the results.

1 Introduction To maintain distribution networks, most distributors follow the schedule of preventive maintenance which stipulates procedures and times of preventive maintenance. New maintenance methodology is based on the principle of reliability centered maintenance (RCM) which is directly dependant on the technical condition and importance of the equipment. By employing RCM principles, maintenance can be economized and the restoration of distribution network can be properly planned. Our method aims to establish the sequencing of pieces of distribution network equipment for restoration.

2 Methodology for restoration of 110 kV power lines Restoration method for 110 kV power lines is based on the principles of RCM [1]. To determine the technical condition, the 110 kV line fault database is used. The line importance is determined with respect to the power transmitted by the line and also the effect of line failure. The outcome of the method is a quantity we call priority of restoration PO. Priority of restoration is a relative value within the range of 0 and 100 %. Higher PO value means increasing urgency of restoration. Restoration priority is calculated in the relation (1), where TS is the technical condition of the line, and DV is the importance of the line. Coefficient kTSDV divides the

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 7–12, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

8

V´ıt Houdek, Tom´ aˇs Mozdˇreˇ n, Stanislav Rusek, Radom´ır Goˇ no

influence of TS and DV on final restoration priority value. If the value is more than 0.5, line importance is preferred and vice versa. (1)

PO  (100  TS )  (1  k TSDV )  DV  k TSDV

Not only accurate assessment of technical condition, but also connection of particular maintenance area with the appropriate line is important. The technical condition is relevant to the maintenance section; however, line importance is relevant to the whole line.

2.1 Technical condition Technical condition is assessed on the basis of the fault database [2]. Much data is stored about every fault in this database - the most important items are fault priority and quantity. Fault priority defines urgency of repair, while quantity defines total number and units (Table 1). There are four types of fault priority and three categories for quantity. Table 1. Categories for quantity Quantity Meters Pieces to 25 m 1 to 10 pcs from 26 to 50 m 2 from 11 to 30 pcs above 51 m 3 above 31 pcs

1 2 3

Technical condition of line must be determined for every maintenance section. For every section, the number of faults for each priority and relevant category for quantity are determined with help of the fault database. Firstly, points of technical condition (BTS) are calculated for the relevant section. (2): (2)

BTS  VPx   ( pM1x  k 2  pM2x  k3  pM3x ) ,

where VPx is weight value for priority x. Variable pM1x means the number of faults with x priority in category 1 of quantity; pM2x a pM3x are numbers for higher category of quantity. Coefficients k2 and k3 increase the final value of BTS with respect to the distributed power category. After the calculation of points of technical condition, the technical condition of a maintenance section is assessed according to results in Table 2. Table 2. Limits of BTS to determine the technical condition TS Limits of BTS BTS_1 BTS_2 BTS_3 BTS_4 Technical condition TS 95 % 80 % 70 % 60 %

>BTS_4 40 %

Analysis of the results of methodology for 110 kV power lines restoration

9

2.1 Importance of line The importance of line must be determined for the whole length of the line. The input parameters for importance are distributed power and the effect of a failure for the whole line. Distributed power is calculated according to the measurements in the distribution network. Every line is allocated appropriate category for distributed power (Table 3). Table 3. Categories of distributed power Distributed power Category

< 50 GWh 1

50 ÷ 150 GWh 2

150 ÷ 250 GWh 3

> 250 GWh 4

The effect of failure generally defines the consequences of a line failure in the distribution system. Altogether, we differentiate 7 types of effects of failures: 1 – „No effect“– no manipulation necessary, 2 – Back-up by 110 kV line of approx. the same length, 3 – Back-up by 110 kV line of approx. the same length with limited power, 4 – Back-up by a longer 110 kV line, 5 – Back-up by a longer 110 kV line - limited power, 6 – Back-up by MV line, 7 – Back-up by MV line - limited power. When both input parameters - the distributed power and the effect of failure are categorized, the line importance can be determined by means of matrix (Table 4).

Effect of failure

Table 4. Matrix of line importance MATRIX Importance of 1 line 0% 1 5% 2 15 % 3 7% 4 25 % 5 20 % 6 35 % 7

Distributed power 2

3

4

5% 10 % 25 % 12 % 35 % 25 % 45 %

10 % 15 % 30 % 17 % 40 % 30 % 55 %

20 % 25 % 45 % 27 % 50 % 40 % 70 %

3 First version of methodology For the hereby given method, we used data from one distribution district. The settings of weights and coefficients are in Table 5 and the matrix for technical condition assessment is in Table 4. The kTSDV coefficient is set on 0.5, which means that final priority is distributed evenly on both - technical condition and line importance. The results of the first version of methodology are in 6. Partial results TS and DV implied low validity of the methodology. The authors corrected their method by in-

10

V´ıt Houdek, Tom´ aˇs Mozdˇreˇ n, Stanislav Rusek, Radom´ır Goˇ no

creasing the number of categories for technical condition assessment with BTS and by altering the limits for restoration priority assessment.

Table 5. Settings for technical condition – first version Values for evaluation of BTS VP1 100 k2 1,2 VP2 40 k3 1,4 VP3 15 VP4 10 Limits of BTS 50 100 200 400 Technical condition TS 95 % 80 % 70 % 60 %

>400 40 %

Table 6. Results of the first version of the methodology Priority of Number of maintenance restoration sections from to 0% 7 1% 5% 40 6% 10% 88 11% 15% 62 16% 25% 87 26% 35% 26 36% 50% 5 >51% 0

4 Second version of methodology The second version of methodology underwent several modifications compared to the theory described in section 2: • expansion of categories for technical condition, • priority to TS, • alteration of limits for PO analysis. Table 7. Settings for technical condition – second version Limits of BTS 50

100

150

250

350

500

>500

40%

30%

Technical condition TS 95%

80%

70%

60%

50%

Analysis of the results of methodology for 110 kV power lines restoration

11

The line importance stayed the same (Table 4), and therefore a decision was made to prioritize the technical condition by decreasing the kTSDV coefficient to 0.4. Technical condition was specified more accurately by increasing categories for TS assessment on the basis of BTS from 5 to 7. The last modification is the alteration of limits for restoration priority assessment – settings in Table 7. The values of weights and coefficients are identical with the first version of (Table 5). Table 8. Results of second version of the methodology Number of maintenance sections Priority of restoration (-) (%) 0% 7 2% 5% 45 14% 15% 150 48% 25% 48 15% 40% 52 17% >40% 13 4% The results (Table 8) show the partition of maintenance sections into zones in order of priority. The first zone includes sections with restoration priority 0 to 5 %, their number represents approximately 16 % of the total. These sections do not need renovation. The second zone with restoration priority 6 to 25% represents almost 60 % of maintenance sections. These sections do not require restoration works, but optimisation of maintenance is possible. The sections in the third zone with priority higher than 40% require urgent restoration. 160

Count of mainetance sections

140 120 100 80 60 40 20 0 0%

5%

15%

25%

40%

Priority of restoration PO

Fig. 1. Graph of results – second version of methodology

>40%

12

V´ıt Houdek, Tom´ aˇs Mozdˇreˇ n, Stanislav Rusek, Radom´ır Goˇ no

5 Conclusion The first section of the paper describes the restoration method for 110 kV lines. The PO restoration priority assessment depending on the TS technical condition and DV line importance are described. The manner of acquiring the TS and DV from input data is also described. The third section of the paper describes the first version of the restoration method, determining the weights and coefficients for the calculation. Results of priority restoration are in Table 6. The results of technical condition and line importance were analysed. Both of them showed poor resolution. Selected restoration priority limits showed poor validity. The second version of the restoration priority method employs increased resolution of technical condition, which is the reason for prioritising the technical condition to line importance. The limits for the final statistical analysis of results were modified to allow distribution into three areas. A function file (EXCEL) was developed with input templates. The file is meant to serve as an example for future software to be developed for the calculation of 110 kV power line priority of restoration.

Acknowledgements This work was supported by the Czech Science Foundation (No. 102/09/1842), by the Grant of SGS VŠB - Technical University of Ostrava (No. SP2013/137) and by the project ENET (No. CZ.1.05/2.1.00/03.0069).

References 1. Moubray J.: Reliability-centered Maintenance. 1997, Industrial Press. ISBN 978-083-1131463 2. Houdek, V., Kral, V., Rusek, S., Gono, R.: Analysis of input parameters and development of restoration method for 110 kV power line. In Proceedings of the 14th International Scientific Conference Electric Power Engineering 2013. 2013, p.89-92, ISBN 978-80-248-29883 3. Houdek, V., Rusek, S., Gono, R.: Backup Alternatives For 110 kV lines . In: Advances in Electrical and Electronic Engineering. Vol. 11, No.3. ISSN 1336-1376.

Possible reasons for the complaints to power quality at the point of renewable energy Possible reasons for the complaints to power quality at Petr Rozehnal, Jan Unger, Petr Krejˇc´ı the point of renewable energy Petr Rozehnal, Jan Unger, Krejčí, Department of Electrical PowerPetr Engineering, ˇ of –Electrical Department Power Engineering, FEECS, FEECS, VSB Technical University of Ostrava, VŠB-Technical university of Ostrava, 17. listopadu 15, 708 33 Ostrava – Poruba 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic http://www.fei1.vsb.cz/kat410 {petr.rozehnal1,jan.unger,petr.krejci}@vsb.cz

[email protected], [email protected], [email protected] Abstract. Electricity is considered one of the most used energy. Power quality is an important factor for the electrical equipment. Violation of power quality parameters causes a rise of complaints on power quality. Complaints supplied electricity is a common problem distributor of electricity. The Czech Republic is evaluated parameters of power quality according to the European standard ČSN EN50160. The increase in electricity production from renewable energy sources leads to the need of new ways of examining energy systems. Evaluation of interruption of renewable energy sources is becoming very important for the assessment of their impact on the power grid, while the reliability of energy systems. Wind turbines and solar panels, due to its structure and nature of work , may be the source of failures in the power networks , and also in their work affects the quality of electrical energy . Operators of wind and solar power plants connected to the grid at risk for non-compliance with prescribed quality parameters power penalty.

1

Introduction

The power system is designed and operated to transfer power from large sources of electrical energy (power) to the point of electricity consumption (purchaser). For the transmission of electrical energy is used by transmission and distribution networks, which are connected to power, but also consumers of electricity. Renewable sources of electricity that are connected to the electricity grid are rather smaller capacities ranging from a few hundred watts to ten megawatt. Renewable sources of electricity are considered solar, wind, biomass, geothermal, hydropower. Electrical energy is degraded quality parameters are increasingly becoming the most important issue for the industry and the company providing the service. Impaired quality of supplied electricity causes in European industry increased cost over 10 billion € per year. Issues of this nature are fairly new. Do awareness on a wider scale have moved only in the mid- 90th years. Electricity is probably the most important commodity in today's era of industry and commerce. These are commodity specific nature of which cannot be stored on a mass scale and can therefore not prior to use to verify its quality. It is therefore a typical example of access ' Just in Time' when components are delivered directly into the production process at the appropriate time and in the proper place credible and proven supplier, without the need to control.

2

Major failures in the network

There several causes of failures in the net. The main two causes of failures and therefore also causes of complaints are development of harmonics and voltage fluctuations. The main cause which distribution companies have to deal with is voltage fluctuations. Voltage fluctuations causes luminous flow fluctuations of luminous sources and this disturbs human. There are several causes of voltage fluctuations. The common cause on low voltage level is

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 13–19, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

14

Petr Rozehnal, Jan Unger, Petr Krejˇc´ı

switching on of big appliances, where big current surges can be created. The common cause on high voltage level and very high voltage level are big industry companies, which have appliances with big consumption. Another problem which can occur in distribution network is presence of harmonics. Common cause of harmonics development in the network is connected appliances on customer’s sides. These appliances are fluorescent lamps, regulators, switched sources of welding machine etc. Problems which can occur due to that are failures of luminous sources, losses in distribution nets, failures in process of HDO system, disturbance in telecommunications etc.

3

Causes of complaints on quality of electric energy

There are several causes of complaints on quality of electric energy from customer’s perspective. These causes can be presence of harmonics, voltage fluctuations in the net, voltage asymmetry and interruption of electric energy supply to the customer. All those reasons can lead to big losses on customer’s side and on supplier’s side. All complaints on electric energy quality have to be verified by supplier of electric energy. Supplier decides if the reason for complaint is justified. Supplier informs customer about justification of the complaint on electric energy quality and fix the issue in case that complaint is justified. All parameters of electric energy which are compared in case of complaint are stated in CSN EN 50160.

4

Complaint to power quality

Complaint to electricity is a problem that the distribution companies face every day. Distribution companies must deal with these complaints in the shortest possible time. In the course of the complaint on the quality of electricity has to determine whether this is a valid complaint to power quality. A well-founded complaint to power quality is, if the violation of any of the parameters of power quality occured. All the 13 power quality parameters are specified in ČSN EN 50 160. The electricity consumers at low voltage levels carry out complaints about the quality of electricity. Complaint can be processed in several ways, in person, by phone call or via electronic contact. Electricity consumers often complain to power quality. Since 2004 electricity distributors are recording the number of complaints to power quality. During the period 2004 - 2012 there were 14,081 complaints to power quality. All of these complaints were categorized into the regions by year and complaint to the power quality.

Possible reasons for the complaints to power quality at the point . . .

15

Graph 1 - trend complaints on power quality Graph 1 shows the number of complaints in each region for the period of 2004 - 2012. Graph 1 shows the increasing number of complaints on the quality of electrical energy. The largest number of complaints on the quality of electricity was reported in the region of West Bohemia (ZCE), where the number of complaints on the quality of electricity is the highest one. Another regions with high number of complaints on the quality of the electrical energy are areas of northern Moravia (SME) and eastern Bohemia (VCE). Compared to these areas in Central Bohemia (STE) and Northern Bohemia (SCE) from 2010 to 2012 decreases the number of complaints on the quality of electricity.

5

Development trend of quality complaints in 2004-2011

Complaints on electric energy quality are tracked by suppliers of electric energy since 2004. There are 11794 complaints on electric energy quality since this year. Distribution companies register reported complaints, but not all entries about complaint are accurate.

Table 1. Number of power quality complaints Year of Complaint

2004

2005

2006

2007

2008

2009

2010

2011

2012

Justified complaints

0

1

146

175

155

127

46

59

0

Unjustified complaints

0

10

222

271

259

332

409

390

2276

Unspecified type complaints

462

322

974

1252

1340

1498

1741

1603

0

Number of complaints in

462

333

1342

1698

1754

1957

2196

2052

2276

There is visible amount of complaints on electric energy quality in table 1. We can see that overall amount of complaints is increased every year. This trend can be caused by more frequent connection of appliances which are causing undesirable retroactive affects in network. 709 complaints from overall amount of 14081 were verified as justified, which is 6%. Major amount of complaints on electric energy quality doesn’t state if this is justified or unjustified complaint. This situation can be caused by multiple complaints from different customers on same issue or it can be complaint without results (solution of these complaints is done during 24 hours).

6

Description of measurement

During March and April, were installed portable analyzers BK-Elcom three measuring points. Measurements were carried out simultaneously at the place where the wind turbine is connected to the distribution network, as well as in the distribution substation transformer and high voltage. This allows us to determine whether the operation of wind turbines affects the surrounding system and how. Due to interrupt measurement, weather conditions, etc., were selected period, from which all available data from all measuring points. This is the period from 20.3. to April 3, 2013. Were evaluated waveforms, long-term and short-term flicker, and harmonic distortion and compared these with allowable values in EN 50160th To determine the effects of rear-connected wind power were realized measurement performance balance beam distribution network, which is controlled wind turbine is connected. Simplified wiring diagram wind electricity is shown in Figure 1

16

Petr Rozehnal, Jan Unger, Petr Krejˇc´ı

Measuring point 1 was at the outlet of the reference beam distribution network in the substation 110/22 kV. Measuring point No. 2 was placed in a distribution transformer (DTS) between wind electricity and MV and measuring point No. 3 was in the connection of wind power plants to the distribution network.

Figure 1 - Simplified diagram of the measurement

7

Selected parameters of electric energy

As mentioned above, the operation of wind power plants can be expected to influence the parameters of the distribution system. Interaction between the distribution system and analyzed wind turbine is defined in the point of common coupling. The DSO is a priority to ensure a stable supply of electricity if possible with constant system parameters. In terms of quality of the supplied energy is to be monitored, in particular: 1. First Voltage changes. 2. Second Flicker - voltage fluctuations. 3. Total Harmonic Distortion Ad1) Voltage changes

Possible reasons for the complaints to power quality at the point . . .

25,0

17

P [MW] 8

Urms [kV] Urms_VTE Urms_RVN P_VTE

24,5

[MW]

24,0

6

23,5 23,0

4

22,5 22,0

2

21,5 21,0

0

19.3

21.3

23.3

25.3

27.3

29.3

31.3

2.4

4.4

Figure 2 - Relationship of high voltage for operation of Wind electricity 260

P [MW]

Urms [V]

255

Urms_DTS P_VTE

250

8

6

245 240

4

235 230

2

225 220 19.3

0 21.3

23.3

25.3

27.3

29.3

31.3

2.4

4.4

Figure 3 - Dependence on the low voltage operation of Wind electricity From the previous figure 2 and 3 it is obvious that voltage waveforms are identical to the low voltage and the high voltage level. At the same time, we can notice that in all measuring points of the voltage is stable and there are no variations in time or not delivered at the time the wind power plant is running at full capacity 2 MW. Generally, the period of operations in the growing and voltage and the voltage follows the change in power output. We can therefore claim that wind power has adverse retroactive effect in terms of voltage. Ad2) Flicker - voltage fluctuations Flicker is defined as the human eye perceptible variation of flux of light sources as a result of periodic dips in sub harmonic frequencies. These voltage changes are generally caused by changes in customer load or changes in generation capacity.

18

Petr Rozehnal, Jan Unger, Petr Krejˇc´ı

If we analyze the theoretical possibility of flicker that accompanies the operation of wind power, it is possible to identify two basic causes of its origin: the effect of wind gusts and wind power tube effect. Effect of wind gusts in the short-term variations of wind speed from its mean value eliminates the inherent inertia of rotating parts of the wind turbine, due to stronger gusts of more or less eliminates the power turbine control. Effect of wind power tube (mast) suppresses much worse. Tube for flowing a wind barrier that slows him. As a parameter determining the flicker is not applied directly to the voltage drop caused by the flicker, but variable called issue of flicker or a flicker severity. Distinguish between short-term (short term) flicker emission Pst, measured or computed at intervals of ten minutes long (long term) emission flicker Plt, determined the interval of two hours. Generally, the more leaves the wind turbine, the emission is less flicker. Systems with frequency converter in most cases have lower emissions than systems with asynchronous generator connected directly. Rules for the operation of the distribution system define the maximum allowed value longterm rate of flicker severity Plt and so it must not exceed 0.46. [1], [2] 1,0

Pst [-]

12

P [MW] Pst_VTE Pst_DTS Pst_RVN P_VTE

0,8

10

0,6

8

0,4

6

0,2

4

0,0

2

-0,2 19.3

0 21.3

23.3

25.3

27.3

29.3

31.3

2.4

4.4

Figure 4 - Dependence of short-term flicker severity level of the operation of wind electricity Rotor inertia is so large that the value Pst is practically negligible and short-term change in speed or direction of the wind does not affect the voltage fluctuations in the distribution network around wind electricity. 0,40

Plt [-]

0,30

10

P [MW]

Plt_VTE Plt_DTS Plt_RVN P_VTE

0,35

8

0,25 0,20

6

0,15 0,10

4

0,05 0,00

2

-0,05 -0,10 19.3

0 21.3

23.3

25.3

27.3

29.3

31.3

2.4

4.4

Figure 5 - Dependence of long-term flicker severity level of the operation of wind electricity Rules for the operation of distribution networks dictate that long-term flicker severity shall not exceed 0.46, which was fulfilled in the entire measurement period. Only the first April

Possible reasons for the complaints to power quality at the point . . .

19

increased value Plt to 0.4 and only in two stages over a period of 2 hours, which was during the peak hours of wind electricity. The MV was a period of lowest value P lt. Ad3) Total Harmonic Distortion 3,0

8 THDu [%]

P [MW]

THDu_VTE THDu_DTS THDu_RVN P_VTE

2,5 2,0

6

1,5 1,0

4

0,5 0,0

2

-0,5 -1,0 19.3

0 21.3

23.3

25.3

27.3

29.3

31.3

2.4

4.4

Figure 6 - Dependence THDU the operation of wind electricity Total harmonic distortion, or if THDU is shown in Figure 6 It has a characteristic waveform that does not matter too much on the production of wind power, but rather the switching power supply, such as television sets. In this graph, we can see some regularity at intervals. These increases are in the afternoon until the evening, when the switch on these characteristic appliances. Nowadays, manufacturers indicate the value of the total harmonic factor not exceeding 5%. It should be sufficient to avoid adversely affecting other devices connected to the network. Size THDU in this case does not exceed 2%.

Conclusion This article summarized the possible return to power quality at the point of renewable electricity. Were summarized numbers of complaints about the quality of electric power, including the legitimacy of these claims. Furthermore summarized possible causes complaints about power quality. In the second half of the article described and evaluated measuring power quality parameters at the site of the complaint, while wind power as a renewal power source. The measurement shows that all the measured power quality parameters were normal. Measurement and subsequent evaluation was carried out according to DIN EN 50160th

Acknowledgement This work was supported by the Czech Science Foundation (No. GA ČR 102/09/1842) and by the Ministry of Education, Youth and Sports of the Czech Republic (No. SP2013/137)

References ČEZ Distribuce [online]. 2011 - Příloha č. 4. Dostupné z WWW: [2] MIŠÁK, Stanislav, PROKOP Lukáš, KREJČÍ Petr, SIKORA Tadeusz: Větrné elektrárny s asynchronními generátory v sítích VN, Elektrorevue. [online]. 11.12.2008, 47 [1]

Maximum Extreme Annual Production Maximum Extreme States, States, Annual Production of of Photovoltaic Power Plant Photovoltaic Power Plant Martin cek, Zdenˇ ek Hrad´ ılek MartinSmoˇ Smoček, Zdeněk Hradílek Department of ElectricalFEI, Power Department of Electrical Power Engineering, VŠBEngineering, –Technical University of Ostrava, ˇ – Technical FEECS, VSB of Ostrava, 17. listopadu 15, 708 33,University Ostrava-Poruba, [email protected] 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic [email protected], [email protected] Abstract. In terms of electricity supply to the grid, photovoltaic power plants are characterized by its stochastic production. Produced power is strongly dependent on the instantaneous meteorological conditions. This article is a subsection of a study that aims to evaluate the impact of the photovoltaic power operation to change the keeping support services with respect to the impact of seasonal climatic conditions. Everything is done at voltage levels 22/110 kV. The study is based on real measured data both the photovoltaic power plant and the appropriately outlet power lines, where the plant is connected. This article mentions the issue of extreme conditions of photovoltaic power plants production. Further is addressed in more detail change the maximum production during the year.

1 Introduction The aim of this survey is to evaluate the influence of the photovoltaic power plant (PVP) operation with respect to the power difference of active power flowing from the substation. These changes to power flowing from the substation plant have adverse effect on supporting services provided by operators of distribution networks. The evaluation is conducted by means of several individual tasks that can be split into two stages. The first stage determines methodology for set-up of extreme thresholds for production generated by the PVP and the second stage concerns analysis of impact of these extreme threshold values on the magnitude of active power flows at the place of measurement within substation plant.

2 Network topology The data available has been obtained by measurement at the PVP operation site (0,4 kV) and the point of electric power output from substation plant (22 kV). Both measurements were conducted on synchronised basis. The electric line has a loop system, yet it is operated in radial system. When this measurement was conducted, the PVP was the only electric power source with significant supply to power line. Fig. 1 shows graphic illustration of a simplified network topology. The measurement points are marked with red square.

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 20–25, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

Maximum Extreme States, Annual Production of Photovoltaic Power Plant

21

Fig. 1 Network topology in analysed area

3 Instance of flowing power The peak output of this PVP is 1.1 MWp. The measurement was conducted on the low voltage side on continuous basis for one year; that was from 30.6.2010 till 29.6.2011. The facts considered include power changes caused by both production output from PVP as well as the nature of consumption within the specific area. On the Fig. 2 are described three curves as sample of measured data, green line produced power from PVP (PVP), blue line power flowing through the substation (SUB) and black line loads (DLD). 6 5

P (MW)

4 3 2 1 0 1

2

3

4

5

6 PVP

7 SUB

8

9

10

DLD

Fig. 2 The sample of actually measured data

11

12

13

22

Martin Smoˇcek, Zdenˇek Hrad´ılek

4 Extreme conditions It is necessary to investigate reasons of power differences at the substation separately both for the photovoltaic power plat and for the loads of consumption. It is desirable to determine extreme conditions for both. In conclude, based on these extremes is possible to evaluate for extreme states flowing power from the substation. The analysis can be split into three stages, it is as fallows. 4.2

Extreme conditions PVP

First stage is focused on the design methodology for evaluation of extreme conditions PVP. The methodology is described in [2]. Fig. 3 shows extreme areas of PVP production. The thresholds of these extreme conditions are defined by two regression polynomial equations, which form the top and bottom envelope curves of stochastic changes in the produced output.

Time 4:30 100

6:30

8:30

10:30

12:30

14:30

16:30

18:30

R 2 = 0,77

P (kW)

-100

Minimum Extreme

-300 -500 -700

R 2 = 0,97 -900

Maximum Extreme

-1100

Fig. 3 Extreme conditions of PVP

4.3 Thresholds of daily load diagram Further stage concentrates in regression analysis of daily load diagram. Differential thresholds are determined using 95% prediction levels. These prediction levels generally define the probability and range for output daily load diagram for ever individual value within specific time intervals. These thresholds are shown in Fig. 4.

Maximum Extreme States, Annual Production of Photovoltaic Power Plant

23

Prediction levels

Fig. 4 Extreme thresholds of daily load diagram

4.4 Extreme conditions substation Previous analyses have produced the difference of output generated by PVP and determination of thresholds for daily load diagram. Extreme conditions that might occur at the substation plant outlet correspond with the sum of extreme conditions of PVP and the daily load diagram. More information describes literature [1]. Figure 5 shows differential thresholds of possible flowing active power from the substation. 4 3,5 3

P (MW)

2,5 2 1,5 1 0,5 0 0:00 P_SUB_MAX

4:00 P_SUB_MIN

8:00

12:00

16:00

P_SUB_5 min._Power_in

20:00

0:00

P_SUB_5 min._Power_out

Fig. 5 Extreme thresholds at the substation

24

Martin Smoˇcek, Zdenˇek Hrad´ılek

Limits contain more than 95 % measured values (assumption of this study). It points out that proposed methodology is justified. The assumption of differential power change at the substation plant outlet for on-hour time interval is demonstrated in Fig. 6. Pursuant of this graph is possible for a given time interval to determine the necessary power reserve for the appropriate outlet from the substation independently of the operating PVP with more than 95% probability. 1,6 1,4

∆P (MW)

1,2 1 0,8 0,6 0,4 0,2 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Time

Fig. 6 Histogram of expected differential power

Prior studies mentioned above have shown evaluation only for bit of sample data, specifically for the month of May. Next study will be focused on spread for all months of the year. Figure 7 shows the instance of own comparison between individual monthly mean, median and mode. It is basis for future analysis. Ja

Fe

Ma Ap May Ju J-ly Au

Se

Oc

-0,1

P (MW)

-0,3 -0,5 -0,7 -0,9 -1,1

Mean

Median

Mode

Fig. 7 Comparison of monthly Mean/Median/Mode

No

De

Maximum Extreme States, Annual Production of Photovoltaic Power Plant

5

25

Conclusion

This survey concentrates in method for statistical analysis of a database measured data on PVP and output from the substation. The outcome of entire methodology is to determine the assumed difference of change active power at the substation outlet for time interval and for each annual season, accurately every month. The object of this study is to compare these extreme states with the load diagram in the area of PVP operation and assess possible impact on evaluations of support services. The conclusion reached herein will refer to the versatile methodology to define differential these change of active power.

Acknowledgements This work was supported by the Czech Science Foundation (No. GA ČR 102/09/1842), by the Ministry of Education, Youth and Sports of the Czech Republic (No. SP2013/137) and by the project ENET (Research and Development for Innovations Operational Programme (No. CZ.1.05/2.1.00/03.0069).

Reference [1] SMOČEK, M. – HRADÍLEK, Z. Photovoltaic Power Plants, Extreme Change of

Power Difference. The 7th International Scientific Symposium on Electrical Power Engineering, TU - Košice, 2013, ISBN: 978-80-553-1462-4 [2] SMOČEK, M. – HRADÍLEK, Z. Methodology for Extreme power thresholds of photovoltaic power plants. Przegląd Elektrotechniczny 10/2013, magazine of Polish electrician, 2013, ISSN 0033-2097.

Problems of connecting of SHPs to distribution system and legislative changes to support RES Problems of connecting of SHPs after 2013 to distribution system and legislative changes to support RES after 2013 Martin ak,1,Radom´ ır Goˇ no1 MartinNov´ Novák Radomír Goňo 1

Department of Electrical Power Engineering, Department of Electrical Power Engineering, FEECS, ˇ of–Ostrava, VSB Technical VŠB - TechnicalFEECS, University 17.University listopadu of 15,Ostrava, 708 33 Ostrava – Poruba {[email protected],[email protected]} 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic [email protected], [email protected]

Abstract: The paper describes possibilities of the connection of small hydro power plants to distribution system. It also overviews legislative changes in the redistribution of subsidies for renewable energy sources after 2013. Key words: Small hydro power plant, legislative changes, synchronous generator, asynchronous generator

1 Introduction In 1930s there were over 11 thousand small hydropower plants (SHP) within the territory of the Czech Republic, but gradually they were replaced by other power sources, and at the beginning of the 1980s there were only 135 SHPs in the country. The potential of our water streams which could be employed for technical purposes is estimated at1. 4 billion kWh/year. Currently, only about two thirds of this potential is being used, so there is capacity for the construction of new sources. However, the problem is to select the suitable location for the installation of new plants. In terms of the CR, a SHP is a plant with less than 10 MW of installed capacity. In the EU terms is even less than 5MW. The essential parameters for a SHP are the head, turbine flow, installed capacity and average yearly power production. By 31 July 2013 the Ministry of Industry and Trade registered 1,466 SHPs with production of up to 1MW with total installed capacity of 150.89 MW. In 2012 gross production of < 1 MW SHPs was 391.425 MWh with total installed capacity 148.5 MW with 4.86 %. share of renewable energy sources (RES) Gross production of 1 < 10 MW SHPs was 525,548 MWh with total installed capacity 162.5 MW and 6.52 % share of RES [1], [5].

2 Connection of SHPs into distribution system Small hydropower plants may work in three systems - grid-connected, in island operation (off-the-grid system), or in a combination of both. SHPs are connected to the grid via circuit breaker, such as synchronous and asynchronous generators. Synchronous generators (SG) and asynchronous generators (AG) work mostly in parallel with the grid. Asynchronous generators are not used as autonomous power c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 26–31, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

Problems of connecting of SHPs to distribution system... after 2013

27

sources. The connection of new plants is specified in ERO decree No. 51/2006 Coll. The conditions for the connection of RES concerning behaviour and reverse effects are specified in attachment 4 on regulations for parallel connection to grid. 2.1 Stand-alone Synchronous generator This generator feeds its own grid. Its alternator with disconnected excitation and taps must be started with a prime mover (combustion, diesel, etc.) before the actual connection. As soon as roughly synchronous rotation is achieved, excitation is connected and field rheostat finishes the regulation of an alternator on nominal voltage on its terminals. Constant frequency of oscillation requires constant revolution of the driving machine, constant output voltage is controlled via excitation current. Then the machine may be connected into the grid via a power device and increase load. Water stream is controlled din dependence on load and power factor is controlled by excitation. 2.2 Synchronous generator working in parallel with grid When a machine set is connected to the grid, current surges and mechanical impacts are observed. These undesirable effects must be eliminated by what experts call synchronizing to so-called stable network. That means that the voltage on alternator terminals does not change with the excitation change. With the change of alternator excitation, reactive current and power factor of an alternator change. Alternator voltage is controlled by the change of load, or the excitation current of an armature changes alternator’s stator current. This leads to the change in the magnetic field of an armature to preserve constant voltage on terminals. When terminal voltage decreases due to increased load, the excitation current must be increased; and on the contrary, when terminal voltage increases, the excitation current must be decreased. Increased capacity of a driven turbine does not increase alternator revolutions, but the power angle increases and hence active power supplied to the grid by an alternator increases. [2], [4], [6] For correct synchronizing of SG these conditions must be met: [4] a) Identical voltage of a generator and grid U G = U S – is controlled via two voltmeters connected to the machine and the grid. Generator voltage is controlled by increasing or decreasing excitation, or more precisely by changing the excitation current of the armature Ib. The difference in instantaneous voltages would cause a surge of reactive current Ij. At the worst, the alternator could be connected when instantaneous voltage values are in antiphase, i.e. out of phase with each other by a  angle. Arising current surges would exceed the short-circuit currents and cause a failure, as an alternator is designed to hold only short-circuit current on its outlet. b) Synchronous frequency of the generator and the grid ωG = ω S – is checked by a frequency meter. The frequency of the machine is set on the frequency of the grid

28

Martin Nov´ ak, Radom´ır Goˇ no

(50 Hz). Difference in frequencies would cause a surge of the active part of current Ič. These surges are caused by accumulated kinetic energy and its rapid transformation into electrical power, as at the moment of connection a substantial amount of armature mass must slow down or speed up depending on the frequencies of the alternator and the grid. This surge mechanically strains the turbine and the alternator of a SHP. c) Phase shift between the generator and the grid voltages is minimal ∆cosφ = min – When connecting the machine to the grid, the sameness (or negligible difference) of voltage phase angles is necessary. A difference in these angles would cause equalizing current. Maximum value of this current grows with a phase shift. d) Synchronous sequence of phases of a generator and LGx = LSx – with a (little asynchronous engine the correct sequence can be determined. The little engine is connected on generator terminals and then the grid. The armature of the engine turns in the direction of the torque field depending on the phase sequence. If it turns in both cases in the same direction, then the sequence is correct. If phase sequence is different, any two phases can be interchanged. The real distribution system is much more complex, and meeting all the above mentioned requirements for connecting a new power plant is impossible. Therefore a DS operator can tolerate certain permissible differences which are defined in ‘Regulations for a DS operation’ approved by ERO. These regulations, which must be respected by every operator, tolerate these differences [7]: - Voltage difference - U < ± 10 % Un - Frequency difference f < ± 0.5 Hz - Phase angle difference < ± 10° Depending on the grid impedance and generator capacity ratio, lower tolerance limits may be necessary, to avoid undesirable reverse effects on the grid. There are many methods for phasing a machine, e.g. synchronizing with lamps bright, synchronizing with lamps out, mixed synchronizing, with a synchronoscope. Power plants use synchro-check relays. Synchro-check relay – with respect to certain delay due to response time of a power switching device, a phasor must connect the alternator to the grid with a pretrigger. The pretrigger determines permissible tolerance of the difference in frequencies of the grid and the generator. Also, autonomous synchronizer that enables setting of alternator speed on the frequency synchronized with the grid frequency can be used. [2]

2.3 Asynchronous generator working in parallel with the grid Almost every asynchronous induction motor with a squirrel-caged rotor can be used as an AG. The stator of an asynchronous machine has a three-phase winding

Problems of connecting of SHPs to distribution system... after 2013

29

similarly to a SG. The rotor has either a squirrel-cage winding or a wound rotor. By bringing three-phase voltage on the stator terminals a rotating magnetic field is induced. Currents arising at this induction rotate the armature in the direction synchronous to that of the magnetic field. When the rotation rates of armature and magnetic field are synchronous, no power is induced, but when the motor rotation is higher than rotation of the field, the machine works as a generator and supplies active power. The difference in the field and armature rotation is the slip. When the rotation rate of the motor is higher than that of the field, the machine works as a generator and supplies active power to the grid. At the same time, it consumes reactive power for its magnetization. The consumption of reactive power increases total power in the grid, which can be restricted by connecting ancillary capacitors. However, capacitors may cause overvoltage on terminals of capacitor – generator parallel combination, or selfexcitation at blackout and subsequent increasing of hydro generator rotation 3.2 Synchronizing of asynchronous generator There are two ways to synchronize an synchronous generator. It can be connected with a grid and subsequently rotated as a motor, or it is connected to the grid with an already synchronous spin, in which case it is phased as a SG. Concerning an AG, we try to minimize the first power surge either by decreasing or removing the active part of the surge current. To achieve that, the frequencies of oscillation of the AG and the grid are compared and the signal for connection is sent after they are even. Asynchronous generator frequency signal for the switch is read form the engine speed or from the voltage on the generator terminals. A generator connected in such a way has only reactive current. The generator works as follows. Initially, the water engine is closed. The generator is connected to the grid and starts to work as an electric motor reaching asynchronous rotation. Subsequently, the water inlet is opened manually or automatically, which makes the generator speed up. When it reaches super-synchronous rotation, it starts to supply active power into the grid. The rotation rate is stabilized by mechanical load. The highest efficiency is achieved when rotation is as close to nominal rotation as possible, which is much determined by proper selection of the generator capacity depending on the source potential. If the machine stops, reverse sequence is recommended. By ‘Attachment 4 on regulations for parallel connection to grid’ (ERO), an AG that is started with a driving machine must be connected with no voltage and revolutions within 95 % and 105 % of synchronous rotation. Concerning AG in island operation (not used for SHP) that are connected with voltage, the same requirements as for SG must be (see. section 2.2) [7], [10]

4 Law amendments for RES subsidy after 2013 Since 1 January 2013 a new law 165/2012 Coll. is in force. This bill draws upon EU legislature on state-supported energy sources and it is a compromise with respect to the previously passed Czech National Renewable Energy Action Plan. The Act 310/2013 Coll. was passed in the third reading on 16 August 2013. This act substitutes Act 165/2012 Coll. on state-supported energy sources. On 17 September 2013

30

Martin Nov´ ak, Radom´ır Goˇ no

the president signed this law amendment and it comes into force on 1 January, 2014. It comprises: 1. Stop of support for the new power plants (except for plants employing wind, hydropower, geothermal energy or biomass that claimed the state support on or before 31 December 2012) From 1 January 2014 the state support is stopped for the RES plants in construction, except for those employing hydropower with installed capacity up to 10 MW ( SHP). The state support is to be provided only to those licenced plants that will be put in operation before 31 December, 2015 or to plants with capacity up to 100 kW if the construction permit was granted before this law comes into force. 2. Stop of state support for heat and power producers in joint-stock companies that have no registered shares issued or those with owners of foreign nationality who are unable to provide a statutory declaration with the names of the shareholders with shares of nominal value exceeding 10 % of the authorized capital of the producer, giving the source of this information. Supposed entering into force is 1 July 2014. 3. Extension of solar tax – since 1 January 2014 power produced in solar power plants that were put in operation from 1 January 2010 to 31 December 2010 (excluding plants with installed capacity up to 30 kW) and received ‘purchasing price’ support is subject to 10 % tax (or 11 % in case of ‘green premium’ support). 4. Support stop for non-central production - the joint-stock power producers who fail to meet transparency requirements on their corporate structure (see paragraph b) above). 5. Cover for power production costs - costs cover is shared by the customer, power plant operator and electrical grid and distribution grid operator. Maximum price is set by ERO on 495Kč/MWh.[12] State support stop excludes: a) high-efficiency cogeneration plants, b) secondary energy sources, c) heat produced with RES.

Acknowledgements: this work was supported by the Czech Science Foundation (No. GA ČR 102/09/1842) and by the Ministry of Education, Youth and Sports of the Czech Republic (No. CZ.1.05/2.1.00/03.0069).

Problems of connecting of SHPs to distribution system... after 2013

31

References [1]ENERGETICKÝ REGULAČNÍ ÚŘAD. Roční Zpráva o provozu ES ČR 2012 [online]. Praha, 2013 [cit. 2013-11-12].Dostupné z: http://www.eru.cz/user_data/files/statistika_elektro//rocni_zprava/2012/RZ_elektro_2012_v 1.pdf [2] GRÉGR, Tomáš. Fázování generátoru pomocí funkce "synchro-check" terminálu vývodového pole [online]. BRNO, 2010 [cit. 2013-11-26]. Dostupné z: https://dspace.vutbr.cz/bitstream/handle/11012/18674/xgregr02.pdf?sequence=1. Bakalářská. VUT. [3] ČSN EN 50160 (33 0122): Charakteristiky napětí elektrické energie dodávané z veřejné distribuční sítě [4] VSB - TU OSTRAVA FEI KAT 410. Fázování synchronního generátoru k síti [online]. Ostrava, 2004 [cit. 2013-11-11]. Dostupné z: http://fei1.vsb.cz/kat410/studium/studijni_materialy/se/cast_B_el_stroje/se_es_c2_fazovani. pdf [5] GABRIEL, P., ČIHÁK, F., KALANDRA, P. Malé vodní elektrárny. 1. vyd. Praha: Vydavatelství ČVUT, 1998. 321 s. ISBN 80-01-01812-1. [6] Cigánek, L.; Elektrické stroje. 6. vyd. Praha: Technicko-vědecké nakladatelství, 1951 [7] ČEZ DISTRIBUCE. Pravidla provozování distribučních soustav: Příloha 4 - Pravidla pro paralelní provoz zdrojů se sítí provozovatele distribuční soustavy [online]. 2012 [8] MASTNÝ, Petr, Jiří DRÁPELA, Stanislav MIŠÁK, Jan MACHÁČEK, Michal PTÁČEK, Lukáš RADIL, Tomáš BARTOŠÍK a Tomáš PAVELKA. CVUT. Obnovitelné zdroje elektrické energie. 2011. vyd. Praha: CVUT, 2011. ISBN 978-80-01-04937-2. [9] DIGAMBER.M.TAGARE. Electricity power generation: The changing dimensions. Hobo ken, New Jersey: John Wiley & Sons, Inc, 2011. ISBN 976 - 0 - 470 - 60028 - 3. [10] Abeceda malých vodních pohonů. [online]. [cit. 2013-11-19]. Dostupné z http://mve.energetika.cz/

Actual Results of the Reliability Computation in 2013 Actual Results of the Reliability Computation in 2013

Martin Slivka1 , Radom´ır Goˇ no1 , Stanislav Rusek1 , Michal Kr´atk´ y2 , Vladimir Kral1

Martin Slivka1, Radomir Gono1, Stanislav Rusek1, Michal Kratky2, Vladimir Kral1 1 Department of Electrical Power Engineering 1 2 Department of Electrical Power Science Engineering Department of Computer 2 Department of Computer Science ˇ – Technical FEECS, VSB University of Ostrava, VŠB - Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava – Poruba 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic {martin.slivka.st1, radomir.gono, stanislav.rusek, {martin.slivka.st1, radomir.gono, stanislav.rusek, michal.kratky, michal.kratky, vladimir.kral}@vsb.cz vladimir.kral}@vsb.cz Abstract. The paper deals with the computation of distribution network components reliability parameters. Actual value of the component reliability parameters in distribution network is used for the reliability computation and also for reliability-centered maintenance system. Reliability indices are possible to retrieve only from accurate databases of distribution companies. Such a database includes records of outages and interruptions in power networks. The main problem for an analysis of these databases is the heterogeneity feature: databases of various distributors differ from one another. It is impossible to retrieve reliability parameters from this data in a direct way. In this paper there is applied a framework for the retrieving of parameters from various outage databases in the Czech and Slovak republics. There are also actual results.

Key Words: Component reliability, failure rate, mean time to repair, distribution network, and outage database

1 Introduction This work deals with the component reliability. It is necessary to observe outages and interruptions in the distribution network for retrieving the component reliability [1]. A larger database would describe the real condition of network equipment more accurately. Therefore, it is necessary to merge databases of various distribution areas. The main problem of the merging is the heterogeneity feature: databases of various distributors differ from one another, because they have different database systems and also different approaches for evaluation of outages and interruptions in their networks. In [2] there is introduced a framework that makes it possible to retrieve parameters from these various databases. This idea is developed and new results are shown here.

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 32–38, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

Actual Results of the Reliability Computation in 2013

33

2 Reliability Analyses A majority of reliability computations is performed in the following way. The reliability computation of the whole system is executed on the basis of components reliability that is included in the system. That is the reason why the reliability is computed in two phases. The first phase represents the retrieving of component reliability parameters and the second phase is the reliability computation itself. Other phases may include the evaluation of computed results and an improvement of the supply quality. In virtue of experience, it is necessary to state that in most cases, the retrieving reliability parameter is far more complicated than the reliability computation itself.

2.1 Input Data for Computations There are various methods for input data retrieval which are based on the type of an examined object, available data of an examined object, etc. Reliability is divided into two basic groups in compliance with the method of input data retrieval: • Empirical reliability – input data for the reliability computation is retrieved from data on equipment, or similar equipment operating under similar conditions • Predetermined reliability – the probability of outage-free operation is expressed on the basis of knowledge about component status. Obviously, incorrect input data leads to poor results, even when a correct computation method is applied. Moreover, in many cases of reliability computations in electrical power engineering, we face the problem of insufficient data size for a component, e.g. an insufficient number of historical records.

2.2 Reliability Computation In the case of empirical reliability, we need data on operations and outages of components occurring in the reliability diagram. In the case of power system components, data must be available for outages of breakers, disconnectors, transformers, lines, etc. for a set type and voltage level. Moreover, there is another type of data necessary for the reliability computation. We need to have knowledge of the power network itself. For example, we must know the number of pieces of equipment for a set type, the total length of a line type, voltage level and so on. It is possible to compute basic reliability parameters of particular components from this data. The number of outages per period is retrieved from the database. The period is usually defined depending on requirements concerning the reliability computation. An additional value necessary for the failure rate computation is the number of components for a set type and area. This value is possible to retrieve from the equipment owner (usually system operator). As the numbers of components change in the real power network during a period, we update it annually. Other important information is possible to retrieve in more detailed databases, e.g. the most frequent cause of outages, areas of the greatest amounts of undelivered energy, etc.

34

Martin Slivka et al.

Regulations 2/74 include reliability parameters for basic equipment. These parameters were set in 1980 and are very outdated. It is necessary to update these parameters using an analysis of outage databases.

2.3 Heterogeneous Outage Data A common way of addressing the problem of heterogeneous data is to develop a common relation scheme and different data transform into the relation. It enables querying and analysis. We have selected 31 attributes [3]. For the component reliability only few attributes are necessary: • Distribution Company - anonymous code of distributor • Outage Identification - unique code of event • Outage Type - accidental, planned or forced • Equipment Voltage - 0.4 kV, 22 kV... • Outage Cause - foreign influences, causes before starting operation... • Equipment Type - overhead line, underground line... • Failed Equipment - specific device - conductor, switch, pole, fuse... • Failed Equipment Type - further specification - wooden pole, steely pole... • Amount of Failed Equipments • Producer - Siemens, ABB... • Production Year - age of the component • Beginning of outage • End of outage - time of restoration of supply to all consumers • End of equipment failure - time of repair of the device • Failure Type - with or without equipment damage Some other attributes are included for continuity of supply analyses and some for future expansion purposes.

3. Problems of data transformation It is necessary to pay attention to valid data transfer because of different approach to failure records by distribution companies. Incorrect transformation disfigure records in database and as a result the value of such a database as a whole is decreasing. Therefore transformation correctness must be checked and causes of malfunction must be analyzed. Then recognized failings must be displaced. As there are huge amount of records, it is very difficult to check data manually and search for causes of incorrectly operating transformation. That is why there was developed software Debugger (Fig. 1) at the Department of Computer Science in order to make easier to check particular data transformations to database. When the Debugger is started, also explorer window is opened. It is possible to choose transformation files and upload them. Then there is written transformation code in part Transformation Program. In the Input tuple field the original data row is displayed. On the right by Transformation Program values of appropriate cells and conversion according to codebooks are displayed. There are values listed in the Out-

Actual Results of the Reliability Computation in 2013

35

put tuple field after transformation is performed. Actual position at the step-by-step operation of the software is highlighted by yellow color in the background. Step-bystep operation is provided by F11 key.

Fig. 1. Software Debugger

Sometimes there are incorrectly entered or eventually missing items at the data from distribution companies. The software is not able to manage with these records and generally whole one case deletes. By step-by-step operation can be easily found out inconsistency in input data. During verification there were found certain amount of data that were not transformed to the application. Purposely there are removed records that are listed multiple times for example because more equipment were affected. There was found by verification of one distribution company data that from total number of 1,594 records 14 were removed on purpose and 12 records are missing. Then loss is approximately 0.76%. Recognized missing records are E1 type, it means where none equipment was damaged.

36

Martin Slivka et al.

4. Results The results include the rates and mean durations of equipment outages. The actual data collection includes outage data from distributors from the Czech Republic and one from the Slovak Republic. We have retrieved data from eight distribution areas. Distributors have delivered their data in xls files twice a year. Today database contains more than 450 thousand records (from 2000 to 2013) on voltage levels 110 kV, MV and partially LV.

4.2 Framework Results The graphic representation of all distribution regions reliability indices from the above-mentioned data for the 22 kV cable is given in Fig. 2. From the significant differences in particular years it is possible to observe the contribution of our analyses. The divergence of reliability indices is eliminated during long-term observation.

Fig. 2. The value tendency of reliability indices of the 22 kV cable

These parameters could update reliability indices from old Regulations 2/74 [4]. There is a comparison of both databases, 1975 - 1990 and 2000 - 2013, in Table 1.

Actual Results of the Reliability Computation in 2013

37

Table 1. Comparison of results ČEZ 22/80

Equipment

2000 - 2013

-1

14.5 215

4.724 5.063

-1

14 3

2.856 4.108

-1

5.2 3.5

0.274 1.804

-1

0.03 2500

0.055 0.808

-1

0.04 1300

0.056 0.276

-1

0.015 30

0.012 24.707

-1

0.01 100

0.022 23.418

22 kV cable

l (year ) t (h)

22 kV overhead line

l (year ) t (h)

110 kV overhead line

l (year ) t (h)

MV/LV transformer

l (year ) t (h)

110 kV/MV transformer

l (year ) t (h)

22 kV circuit breaker

l (year ) t (h)

110 kV circuit breaker

l (year ) t (h)

In Table 1, we can observe that the current reliability indices are rather more superior. One of the results of analyses is structuring failures according to their causes. The most common cause of outages is “Operation and maintenance causes”. It is possible to provide also comparison of distribution regions - REAS (Fig. 3). The Energy Regulatory Office could find these results useful for justifying of renewal costs among distribution system operators.

Fig. 3. Comparison of distribution regions

We can also obtain other information important for operators, such as the faulty equipment series from a specific producer, areas of the greatest amounts of unsupplied energy, distribution of outages according to their duration, etc.

38

Martin Slivka et al.

5. Conclusion A statistical significance of an outage database depends on the number of records in the database. A larger database would describe the real condition of the network equipment more accurately. The main problem is the heterogeneity feature: databases of various distributors differ from one another. The results include the rates and mean durations of equipment outages. We can also obtain other significant information for operators. We compared the new results to the original results in this paper. Acknowledgements: This work was supported by the Czech Science Foundation (No. 102/09/1842), by the Grant of SGS VŠB - Technical University of Ostrava (No. SP2013/137) and by the project ENET (No. CZ.1.05/2.1.00/03.0069).

References 1. R.E. Barlow, & F. Proschan, Statistical theory of reliability and life testing: probability models (New York, USA: Holt, Rinehart and Winston, 1975) 2. M. Kratky, R. Gono, S. Rusek, & J. Dvorsky, A framework for an analysis of failures data in electrical power networks. Proc. PEA Conf. on Power, Energy, and Applications, Gaborone, BW, 2006, 45-46 3. R. Goňo, M. Krátký, & S. Rusek, Analysis of Distribution Network Failure Databases. Przegląd elektrotechniczny, 86(8), 2010, 168-171 4. J. Piskac, & J. Marko, Regulations for electric power system no. 2 – failure statistics at electricity distribution (Prague, CZ: CEZ, 1974)

Design, realization realization and analysis of Measuring Heat Design, and analysis of Measuring Energy Balance HeatPump Pump Energy Balance JanˇŠrámek, Zdeněk Hradílek. Jan Sr´ amek, Zdenˇek Hrad´ılek Department of Electrical Power Engineering, Department of electric power 410, FEI, VŠB-TU Ostrava, 17. Listopadu ˇ engineering FEECS, VSB – Technical University of Ostrava, 15/2172, 708 33 Ostrava Poruba, Czech Republic, 17. [email protected], 15/2172, 708 33 Ostrava-Poruba, Czech Republic [email protected] [email protected], [email protected]

Abstract. Because applications and operation of heat pump, especially air/water type, are connected with problem based on the principle of these heat sources, we designed methods for measuring complete heat pump energy balance. We applied this designed methods on real property with air/water heat pump as primary heating source. This will allow us clear up real operation characteristic of this type of heat pump and we gained real data for analysis. This article presents realization from design to analysis step by step. Gathered data allow us present possibilities for improving operation of air/water heat pump with accumulation of excess energy.

Key words: heat pump, measuring system, thermal contemporaneity, accumulation, energy balance

1 Location and term of realization The energy balance measurement was conducted within an object with intermittent utilisation. The primary heating system in this object comprises air/water heat pump only. The object is situated in the area of Beskydy, Ostravice-Staré Hamry (CZE); and its built-up area amounts to the total of 131 m2. The total heat loss incurred by the object is equal to 4.5 kW in accordance with standard [2]. The measurement period lasts from autumn 2012 to spring 2013. [1] 1.1 Measured heat source-heat pump Heating is secured by air/water heat pump with nominal heat power 7,6 kW (A2/W35) and as bivalence source is used electric heating element. Heat pump is involved in preparation of hot domestic water (HDW).

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 39–45, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

40

ˇ amek, Zdenˇek Hrad´ılek Jan Sr´

Table 1. Heat pump specification. [4]

Heat pump specification Producer/supplier HOTJET (Czech Republic) air/water (A/W) Type of heat pump Compact Nominal thermal power (A2/W35) 7,6 kW Bivalence source Electric heating element 7,5 kW Heating system Under floor heating Temperature gradient (°C/°C) 45/35

Fig. 1. Measured object-air/water heat pump.

2 Measurement engineering design and installation This part describes the design and schematic representation of the measuring apparatus. The most serious problem was the installation of hydraulic components into the already finished pipeline from heat pump. Provisions for monitoring of the data necessary were based on the existing measuring equipment and components: [4] Heat meter SIEMENS Megatron 2 (metrologically proven gauge) Monitor of the distribution network (MDS-U) Measuring bridge ACS tools SIEMENS Laptop (operation system XP) Cable (data) and serial convector USB/RS 232 MDS-U software Software ACS tool MS excel

Design, realization and analysis of Measuring Heat Pump Energy Balance

41

Fig. 2. . Measuring block diagram

The whole measurement system has been designed as automated. In spite of that, we had to spend the first weeks with more frequent monitoring and fine tuning of the system to eradicate some errors, especially in settings of the software equipment. Continuous download of data will enable us conduct checks during measurement and data analysis. Practical installation of measuring components was per-formed at the break of September/October 2012. The process dealt mainly with integration of heat gauge into the heat pump return pipe. MDS-u was connected to the distribution board intended solely for supply of power to the heat pump and its components.

Fig. 3. Installation of measuring system, left fig. shows MDS-U, right fig. shows heat meter.

42

ˇ amek, Zdenˇek Hrad´ılek Jan Sr´

3 Method and objective of measurement The whole measurement system has been designed to enable retrieval of all heat and electric power data during heat pump operation. The measurement is to retrieve the following data required: Electric input power of heat pump o Electric current (A) o Electric voltage (V) o Electric power (kVAr, kVA, kW) Produced energy by the heat pump o Total produced energy (kWh) o Actual heat power (kW) Temperature o Outdoor and indoor temperature (°C) o Temperature of supply and return pipe of heat pump (°C) o Temperature in accumulation tank (°C) Time of compressor operation (on/off) The measurement focuses on two essential objectives that can be described as monitoring of full heating season and shorter periods. That will enable us retrieve and measure data for assessment of the full heating season as a whole with simultaneous monitoring of rapid changes in the heat pump on/off mode.

4 Measured Data Analysis Whenever possible the data measured of interest for us should define full range of temperatures for heat pump operation (-18°C to +20°C). The temperature range in December was limited from -10,6 up to +9.55°C, which is interesting with reference to functioning principle of the air/water pump. The data obtained in December was used to perform partial calculations. These produced the average performance factor, the total energy produced and the total energy consumed. As far as the performance factor is concerned, the data file was processed using the coefficient values of 1-4. That helped us filter out the performance factor beyond these threshold conditions. The initial data file comprises 37,814 values/lines. The data was measured within 1minute intervals. The data file is not perfectly complete but it includes a continuous section from 1.12.2012 till 29.12.2012. The interval measured (1 minute) seems sufficient for description of heat pump operations dynamics and we have taken on adjustments to eliminate incorrect and incomplete data only. Apart from filtering the incomplete data out, the data file has been further split to heat pump operation and downtime conditions. The data file showing mere operation of heat pump then includes 10,227 lines only.[5]

Design, realization and analysis of Measuring Heat Pump Energy Balance

43

Table 2. Measured data (December 2012).

Data File for December 2012 Number of values/lines

37,814

-

Heat pump operation

10,227

-

Off heat pump operation time

27,002

-

Removed data

585

-

Outside temperature (min.)

-10,6

°C

Average temperature

0,85

°C

COP A2/W45 measured (producer)

2.53 (3,1)

-

Produced energy (ACS-TOOL)

1461.75

kWh

Electric energy (MDS-U)

665.78

kWh

Fig. 4. The dependence coefficient of performance (COP) and power according to outside temperature.

First result (Fig. 5) are nearly about 20% lower values of COP and heat power than guarantee by producer (in accordance with norm CSN EN 14511).

44

ˇ amek, Zdenˇek Hrad´ılek Jan Sr´

Fig. 5. Coefficient of performance (COP) measured (guarantee by producer).

5 Conclusion The methodology of measurement presented in this article and the data analysis outlined here represent actual presentable opportunity to obtain data about operation of air/water heat pumps to be further developed to establish the heat asynchrony in heating of the object concerned. The objective is to obtain the data for full heating season to establish the functional dependency of heat pump output within basically the whole temperature range as required. The functional dependency, together with mode/s of heating in the object concerned will enable us innovate sizing of heat pumps to ensure more effective utilization of their options. This is also associated with monitoring of parameters of charging and discharging of the heat capacity integrated within the system. We can further work on improved effectiveness of design with respect to tank size.

Acknowledgments This work was supported by the Ministry of Education, Youth and Sports of the Czech Republic (No. SP2013/137) and Czech science foundation (GAČR No.102/09/1842) and by the project ENET (Research and Development for Innovations Operational Programme (CZ.1.05/2.1.00/03.0069).

Design, realization and analysis of Measuring Heat Pump Energy Balance

45

References [1] ŠRÁMEK, Jan. HRADÍLEK, Zdeněk. Methods for Measuring Heat Pump Energy Balance. In. Proceedings of the 14th International Scientific Conference EPE 2013 May 28-30, 2013 Dlouhé Stráně, Czech Republic. ISBN: 978-80-248-2988-3 [2] ČSN EN 12831 Heating systems in buildings – Method for calculation of the design heat load [3] ŠRÁMEK, Jan. Design, realization and analysis-methods for measurement heat pump energy balance. In. Proceedings of the 11th annual workshop WOFEX 2013, Ostrava, Czech Republic. ISBN: 978-80-248-3073-5 [4] ŠRÁMEK, Jan. HRADÍLEK, Zdeněk. Heat pump energy balance measurement. In. Proceedings of the 9th Workshop ELNET 2012, Ostrava, Czech Republic. ISBN: 978-80-2482926-5 [5] ŠRÁMEK, Jan. HRADÍLEK, Zdeněk. Methods for Measuring Heat Pump Energy Balance. In. Proceedings of the International Scientific Conference Forecasting in electric power engineering 2013, September 11-13, 2013 Podlesice, Poland.

Improved Physical Design of Outage Database*

Improved Physical Design of Outage Database? Peter Chovanec, Michal Kr´atk´ y, Pavel Bedn´aˇr

Peter Chovanec, Michal Kr´ atk´ y, and Pavel Bedn´ aˇr Department of Computer Science, Department of Computer Science ˇ – Technical FEECS, VSB University of Ostrava, ˇ – Technical VSB University of Ostrava, Czech Republic 17. Listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic {peter.chovanec, michal.kratky, pavel.bednar}@vsb.cz {peter.chovanec, michal.kratky, pavel.bednar}@vsb.cz Abstract. The reliability computation is applied for the maintenance of equipments in power networks. The reliability computation is calculated from a database of outages in electrical power networks. In the case of the outage database described in this paper, data are stored in a relation with 31 attributes. The reliability computation requires to process up to hundreds of range queries over this data. Therefore, the efficient processing of these queries is necessary. Since multidimensional range queries are used, a multidimensional data structure, the R-tree, has been applied in our previous work. In this paper, we introduce an improved physical design of the outage database: the Signature R-tree is utilized and compared with the well-known R-tree. Key words: power networks, reliability computation, outage data, Rtree, Signature R-tree

1

Introduction

Institutional changes taking place all over the world drastically effect the approach to power supply quality. It is developing towards a purely commercial matter between suppliers and their customers. The supply that does not comply with agreed qualitative parameters will lead to trade disputes and financial settlements. Undelivered energy, including its valuation, has arrived on the scene. The two following aspects of supply quality may be considered: 1. Supply reliability – relating the availability of electricity in the given location. 2. Voltage quality – relating to the purity of characteristics of the voltage waveform, including the absolute level of voltage and frequency. This document deals with the first aspect in more detail. Worldwide centers of reliability computation1 provide databases of information about the availability of electronic and non-electronic components and distribution functions for various failure types. They include the result failure rate and we can retrieve information about the producer, operation conditions, etc. These databases are ?

1

This work was supported by the Czech Science Foundation (No. GA102/09/1842) and by the Ministry of Education, Youth and Sports of the Czech Republic (SGS, No. SP2013/137). For example Alion System Reliability Center, http://src.alionscience.com/

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 46–54, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

Improved Physical Design of Outage Database

47

applicable to the availability prediction of complicated systems. However, these databases do not include data about power equipments. The IEEE standards define a host of reliability indices applied to distribution reliability. IEEE P1366 [17] explains the reliability indices applied to measurement of distribution system reliability, and a way of calculating reliability indices. Although authors introduce a discussion about some factors influencing these indices, reliability parameters for power system equipment are not depicted. The Canadian Electrical Association2 introduces a collection of reliability parameters for power system equipment. This is useful for North America; however, it is almost impossible to compare conditions and equipment in North America and Central Europe. In many cases, it is necessary to compute electrical energy not supplied to consumers; probability computation of not supplied energy is only possible on the basis of the reliability computation results. Consequently, we need to observe failures and outages in the transmission and distribution of electrical energy3 for retrieving the component reliability [1]. In [13], we introduced a framework for retrieving reliability parameters in distribution networks. Consequently, we improved the approach by a new embedded DBMS, called QuickDB, presented in [11]. In [6], we depicted preliminary results of multiple range queries over the outage database. Since we store outage data in a relation with 31 attributes and we use multidimensional range queries to query the relation, a multidimensional data structure, the R-tree, is utilized as a storage of the data. In this paper, we show results of the Signature R-tree [14, 7] and a comparison with the R-tree over the outage data. This paper is organized as follows. In Section 2, we briefly describe QuickDB and the R-tree. In Section 2.3, we outline principles of the Signature R-tree. In Section 3, we introduce the outage database and describe the reliability computation. In Section 4, we put forward preliminary results of the approach. In the last section, the paper content is resumed and the possibility of a future work is outlined.

2

Database System for Handling Outage Data

2.1

Introduction

In [13], we have introduced a framework for storage and querying outage data [9, 8]. Databases of various distributors are transformed into the common relation scheme with 31 attributes. Since then, several works have been presented [12, 2, 4]. In [11], we introduced a new data storage based on multidimensional data structures [15]. A variant of the R-tree [10], called the R∗ -tree [3], has been 2 3

http://www.canelect.ca/ We have used the term ’outage database’ instead of the preferred phrase ’database of failures and outages in the transmission and distribution of electrical energy’ in this paper.

48

Peter Chovanec, Michal Kr´ atk´ y, Pavel Bedn´ aˇr

applied for the implementation. In [6], we depicted preliminary results of multiple range queries over the outage database. In [5], we introduced complete algorithms, cost model, and results of multiple range queries. 2.2

R-tree and its Variants

Since 1984 when Guttman proposed his method, R-trees [10] have become the most cited and most used as reference data structure in this area. The R-tree is a height-balanced tree based on the B+ -tree with at least 50% utilization guaranteed. This data structure supports point and range queries and some forms of spatial joins as well. A general structure of the R-tree is shown in Figure 1. R2 p9

p5

R1 R2

R5

p8

p1

p4 p2

R3 R4

R3 R4

p6

R6

p7 p10

p2 p4 p8

p10 p6

R5 R6

p9 p1 p5

p7 p3 p11

p3

R1 p11

Fig. 1. A planar representation and general structure of the R-tree

It is a hierarchical data structure representing spatial data by the set of nested n-dimensional minimum bounding rectangles (MBR). If N is an inner node, it contains pairs (Ri , Pi ), where Pi is a pointer to a child of the node N . If R is the inner node MBR, then the boxes Ri corresponding to the children Ni of N are contained in R. Boxes at the same tree level may overlap. If N is a leaf node, it contains pairs (Ri , Oi ), so called index records, where Ri contains a spatial object Oi . Each node of the R-tree contains between m and M entries unless it is the root and corresponds to a disk page. Other properties of the R-tree include the following: – Whenever the number of node’s children drops below m, the node is deleted and its descendants are distributed among the sibling nodes. The upper bound M depends on the size of the disk page. – The root node has at least two entries, unless it is a leaf. – The R-tree is height-balanced; that is, all leaf nodes are at the same level. The height of an R-tree is at most blogm N c−1 for N index records (N > 1). Many variants of the R-tree have been proposed during the last decades. Although original algorithms of the R-tree tried to minimize the area covered by MBRs, R∗ -tree [3] takes other objectives into account, e.g. the overlap among

Improved Physical Design of Outage Database

49

MBRs. R+ -tree [16] was introduced as a variant that avoids overlapping MBRs in intermediate nodes of the tree and an object can be stored in more than one leaf node. 2.3

Signature R-tree

The range query algorithm traverses the tree from the root node and it follows relevant items in each node. The item is relevant if its MBR is intersected by the query rectangle. The algorithm recursively traverses all subtrees and it is finished after all relevant subtrees are processed. The R-tree has some features which can significantly slow down the range query processing. MBR of nodes often overlap and cover the dead space (the space without any items). It can lead to the retrieval of nodes without any relevant items. In [7, 14], the Signature R-tree was introduced for efficient processing of range queries. The Signature R-tree is a variant of the R-tree including multidimensional signatures for a more efficient filtration of irrelevant tree nodes. A general structure of the Signature R-tree is presented in Figure 2. Each leaf node of the tree is described by a multidimensional signature, that is built from signatures of particular node items for each dimension. Each inner node item includes the MBR as well as the multidimensional signature. Consequently, such a tree contains two hierarchies, the hierarchy of MBRs and hierarchy of signatures.

n-dimensional signature of tuples in the super-region

super-region

n-dimensional region signature of tuples in (MBR) the region

T

Rl:Rh S

...

... ...

...

T

Rl:Rh S

...

...

index – hierarchy of MBRs and n-dimensional signatures

... ...

Rl:Rh S T

...

Rl:Rh S

T

...

Rl:Rh S T

...

T

... ...

Rl:Rh S T

...

T

indexed tuples

tuples in the region

Fig. 2. Structure of the Signature R-Tree

3

Reliability Computations

The majority of reliability computations is performed in the following way. The reliability computation of the whole system is executed on the basis of components reliability that are included in the system [8]. That is the reason why the

50

Peter Chovanec, Michal Kr´ atk´ y, Pavel Bedn´ aˇr

reliability is computed in two phases. The first phase represents the retrieving of component reliability parameters and the second phase is the reliability computation itself. Other phases can include an evaluation of computed results and an improvement of the supply quality. In virtue of experience, it is necessary to state that in most cases, retrieving a reliability parameter is far more complicated than the reliability computation itself. It consists from a set of non-trivial queries over the data collection, e.g. Figure 3 shows a form for the reliability computation generating 120 range queries.

Fig. 3. A form of the reliability computation

4

Preliminary Results

In our experiments4 , we compare the R-tree and the Signature R-tree over the outage database. As a storage, we used embedded QuickDB presented in [11]. The embedded DBMS has been implemented in C++. The outage database includes approximately 330,000 records with 31 attributes. In our test, we measure processing times of the passportization computation for individual distributors (with the REAS xx abbreviation). It typically includes 1,000 range queries per one computation. The efficiency of the reliability computation has been measured by the database time and the response 4

The experiments were executed on an Intel Xeon E5430 2.66Ghz, 12.0 MB L2 cache; 8GB of DDR333; Windows 2003 Server R2.

Improved Physical Design of Outage Database

51

Table 1. Results of passportization

REAS REAS REAS REAS REAS REAS REAS

03 04 05 07 08 10 11

R-tree Sig. R-tree Speedup Processing Time [ms] Processing Time [ms] Database Response Database Response Database Response 73 819 29 791 2.51 1.03 1,874 3,375 440 1,957 4.26 1.73 1,439 3,099 316 1,993 4.55 1.55 1,268 2,339 219 1,318 5.79 1.78 689 2,207 249 1,803 2.77 1.22 632 2,354 223 1,975 2.84 1.19 497 2,159 139 1,819 3.57 1.19

time. The database time means the processing time of queries in the data structure, the response time means the complete time of the web application between starting the computation and displaying a result to a user.

Average Database Time for Passportization in period 2000 - 2012 2 000 1 800 1 600

Time [ms]

1 400

1 200 R-tree

1 000

Sig. R-tree 800 600 400 200

0 REAS03

REAS04

REAS05

REAS07

REAS08

REAS10

REAS11

Fig. 4. The average database time for passportization in period 2000 – 2012

We compare performance of the passportization computation in period of years 2000 – 2012. All tests have been executed 10×, the average results are shown in Table 1. In Figures 4 and 5, we show the database time and the response time, respectively, of the R-tree and the Signature R-tree. Figure 6 shows the average speedup of both trees for the database and response times. The database time is approximately 2.5× to 5.8× more efficient for the Signature R-

52

Peter Chovanec, Michal Kr´ atk´ y, Pavel Bedn´ aˇr

tree compared to the R-tree. The response time has slightly better performance ranging from 3% to 20%.

Average Response Time for Passportization in period 2000 - 2012 4 000 3 500 3 000

Time [ms]

2 500

R-tree

2 000

Sig. R-tree 1 500

1 000 500 0 REAS03

REAS04

REAS05

REAS07

REAS08

REAS10

REAS11

Fig. 5. The average response time for passportization in period 2000 – 2012

5

Conclusion

The outage database is a collection of outages in power networks in the Czech and Slovak Republics. Its existence is necessary for the reliability computation of a wholesale-consumer connection; therefore, demand for this computation increases. A significant number of complex queries is necessary to process during the computation; a sophisticated storage of the collection and query processing are necessary. In [11], we introduced a new embedded DBMS, called QuickDB, for handling the outage database. The R-tree data structure has been used as a storage of the data. In this paper, we compared the Signature R-tree with the R-tree: the query processing time is approximately 2.5× to 5.8× more efficient for the Signature R-tree compared to the R-tree.

References 1. R. E. Barlow and F. Proschan. Statistical Theory of Reliability and Life Testing: Probability Models. Holt, Rinehart and Winston, Inc., 1975. 2. R. Baˇca, M. Kr´atk´ y, and V. Sn´aˇsel. Bulk-loading of Compressed R-tree with ˇ – Failure Data. In Proceedings of the 4th Workshop ELNET 2007. FEECS, VSB Technical University of Ostrava, 2007.

Improved Physical Design of Outage Database

53

Average Speedup for Passportization in period 2000 - 2012 6

5

Speedup

4

Database Time

3

Response Time

Base Line

2

1

0 REAS03

REAS04

REAS05

REAS07

REAS08

REAS10

REAS11

Fig. 6. The average speedup for passportization in period 2000 – 2012

3. N. Beckmann, H.-P. Kriegel, R. Schneider, and B. Seeger. The R∗ -tree: An efficient and robust access method for points and rectangles. In Proceedings SIGMOD 1990, pages 322–331. ACM Press, 1990. 4. P. Chovanec and M. Kr´atk´ y. Benchmarking of Lossless R-tree Compression for Data of Failures in Electrical Power Networks. In Proceedings of the 7th Workshop of ELNET, Czech Republic, 2010. 5. P. Chovanec and M. Kr´atk´ y. On the Efficiency of Multiple Range Query Processing in Multidimensional Data Structures. In Proceedings of the 17th International Database Engineering & Applications Symposium, IDEAS ’13, pages 14–27, New York, NY, USA, 2013. ACM. 6. P. Chovanec, M. Kr´atk´ y, and P. Bedn´aˇr. Querying Outage Data using Multi Queries - Preliminary Results. In Proceedings of the 9th Workshop of ELNET, Czech Republic, 2012. 7. P. Chovanec and M. Kr´atk´ y. Efficiency Improvement of Narrow Range Query Processing in R-tree. In Proceedings of the Dateso 2009 Annual International Workshop on DAtabases, TExts, Specifications and Objects, volume 471. CEUR Workshop Proceedings, 2009. 8. R. Goˇ no and S. Rusek. Analysis of Power Outages in the Distribution Networks. In Proceedings of the 8th International Conference on Electrical Power Quality and Utilisation (EPQU2003), Cracow, Poland, 2003. 9. R. Goˇ no, S. Rusek, and M. Kr´atk´ y. Reliability analysis of distribution networks. In Proceedings of the 9th International Conference on Electrical Power Quality and Utilisation, EPQU 2007. Barcelona, Spain. IEEE Press, 2007. 10. A. Guttman. R-Trees: A Dynamic Index Structure for Spatial Searching. In Proceedings of the International Conference on Management of Data, ACM SIGMOD 1984, Boston, USA, pages 47–57. ACM Press, 1984.

54

Peter Chovanec, Michal Kr´ atk´ y, Pavel Bedn´ aˇr

11. M. Kr´atk´ y, R. Baˇca, and P. Chovanec. Efficiency of the Embedded Database System for Handling Outage Data. In Proceedings of the 8th Workshop of ELNET, Czech Republic, 2011. 12. M. Kr´atk´ y, R. Goˇ no, and S. Rusek. A Framework for Querying and Indexing Electrical Failure Data. In Proceedings of ELNET 2006. Ostrava, Czech Republic, 2006. 13. M. Kr´atk´ y, R. Goˇ no, S. Rusek, and J. Dvorsk´ y. A Framework for an Analysis of Failures Data in Electrical Power Networks. In Proceedings of the International Conference on Power, Energy, and Applications Conference, ELNET/PEA 2006. IACTA Press/IASTED, 2006. 14. M. Kr´atk´ y, V. Sn´aˇsel, J. Pokorn´ y, and P. Zezula. Efficient Processing of Narrow Range Queries in the R-Tree. In Proceedings of the tenth International Database Engineering & Applications Symposium, IDEAS 2006. IEEE Computer Society Press, 2006. 15. H. Samet. Foundations of Multidimensional and Metric Data Structures. Morgan Kaufmann, 2006. 16. T. K. Sellis, N. Roussopoulos, and C. Faloutsos. The R+ -Tree: A Dynamic Index For Multi-Dimensional Objects. In Proceedings of VLDB 1997, pages 507–518. Morgan Kaufmann, 1997. 17. The Institute of Electrical and Electronics Engineers. Guide for electric distribution reliability indices, http://ieeexplore.ieee.org/xpl/ articleDetails.jsp?arnumber=1300984, 2003.

Automatic Automatic Consumption Consumption Optimization Optimization with with regard regard to to the the Green Green Premium Premium Policy Policy 1 2 Miroslav Pr´ yPr´ mek , Aleˇ ak1a,kand Luk´ ˇs Prokop 1 s Hor´ 1 2 Miroslav ymek , Aleˇs Hor´ , Luk´ aˇsaProkop 1

Faculty of Informatics, University Brno 1 Faculty ofMasaryk Informatics Botanick´ aMasaryk 68a, 602University 00 Brno, Czech Brno, Republic {xprymek,hales}@fi.muni.cz Botanick´ a 68a, 602 00 Brno, Czech Republic 2 Faculty of Electrical Power of Engineering Computer Science 2 Department Computer and Science VSB – Technical University of Ostrava ˇ FEECS, VSB – Technical University of Ostrava, listopadu15/2172, 15, 708 33 – Poruba, Czech Republic 17.17. Listopadu 708Ostrava 33 Ostrava-Poruba, Czech Republic [email protected] fxprymek,[email protected], [email protected] Abstract. The increasing number of local renewable energy sources connected into the power grid brings new challenges – organizational, technical, political. One of the challenges corresponds to the problem of efficient localization of the power source nearest to the consumption points with the aim of lowering the pressure put on the distribution grid. To support this effect, the Energy Regulatory Office of the Czech Republic declares the policy of the ”Green premium” (bonus) which motivates local consumption of the energy produced by renewable sources. The green premium offers financial support of direct consumption of the local produced power. In the paper, we briefly present the technical and organizational rules of this policy and a case study of an automatic system for intelligent demand-side management adapted to these rules.

1

Introduction

In the presented research, we have focused on the lowest level of the smart grid structure – a household. The main aims of the control logic on this level are: – control the operation of particular appliances according to the actual power supply and its economical parameters – minimize demand peaks, make the overall consumption profile as fluent as possible (by coordination of particular appliance operation) – flexibly react to external effects (outage, brown-out, etc.) to minimize possible losses – reach the given goals without a significant impact on user experience In the following text, we describe the Priority-driven Appliances Control System (PAX [5, 4]). The PAX system is designed for small and cheap microcontrollers for particular appliance control, yet flexible enough to fulfill the above-stated criteria well. The controlling core of the system can mix real and virtual appliances, so the system can be used as a smart home simulator as well as a real, physical appliance controller. Later in the paper we describe how real world data are used as a basis for smart home modeling. c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 55–59, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

56 2

1.1

Miroslav y aak, ˇs Prokop Miroslav Pr´ Pr´ ymek, mek, Aleˇ Aleˇss Hor´ Hor´ k, Luk´ and aLuk´ aˇs Prokop

Data Sources for Initial Testing

Several measurements were performed to collect real household data to test and demonstrate PAX functionalities. Each household appliance in a selected real household was monitored using power networks analyzer MDS-U [1] during a long time period. MDS-U is a power analyzer which is able to measure voltage, current and power factor and calculate electrical quantities like active, reactive and apparent power for selected time interval. In this case, one minute time interval was used for data collection. Averaged power consumption curves were defined based on long time power consumption data for 20 most common household appliances (refrigerator, cook top, wall oven, personal computer, washer, dishwasher, vacuum cleaner, etc.). During the long time measurement, the switching scheme was evaluated for all monitored household appliances. Usual power consumption for each time can be defined based on averaged power consumption curves and the switching scheme for most common household appliances. The power consumption curve of each household appliance are the fundamental data for PAX testing. Together with household appliances monitoring there were selected renewable power sources monitored during 1 year time period. According to the actual trends in renewable energy sources utilization, photovoltaic (PV) and wind power plants (WPP) were chosen. PV consists from mono crystalline panels Aide Solar (P MAX 180 W, I MP 5 A, V MP 36 V, I SC 5.2 A, V OS 45 V) and PV has 2 kWp rated power. WWP uses synchronous generator with permanent magnets of 12 kVA, voltage 560 V, current 13,6 A, torque 780 Nm and 180 rpm. A detailed description of the small off-grid power system test bed can be found in [3] or [2]. 1.2

The PAX Implementation

As we have mentioned above, the PAX system is designed to support research of the smart home automation from the simulation phase up to device control in real-time. For the later usage scenario, the scheduler is driven by real-time clock, the events are induced by external sources or occur at predefined times. When used as a simulator, without connections to real devices, there is no need to restrain the process with real-time clock. The (simulated) events should occur and be processed as fast as possible to make mass data processing possible. For this purpose, the PAX core implements an event queue. The queue is filled with simulated or real-world measured data and the simulation consists of sequential processing of the queue events. Since the PAX core heavily depends on (asynchronous) message passing, it is possible to convert almost any code sequence into a sequence of events in the event queue without any special changes in the code. The same code can thus be used for offline data processing (simulation) and online device control. With this pure event-driven design, it is possible to implement arbitrary time precision simulation and simulate every possible aspect of the control system. The grid appliances are classified into four types according to their control mechanism, user expectancies and power consumption profiles: interactive appliance (electric light, television), intelligent interactive appliance (computer),

Automatic to the Green AutomaticConsumption ConsumptionOptimization Optimization...wrt.the Green Premium Premium Policy Policy

573

Fig. 1. Subsidy/cost of particular consumption and generation components

deferrable-operation appliance (washing machine, dryer), and feedback-controlled appliance (refrigerator).

2

PAX Applications to Multi-Tariff Environments

In recent years, the number of new installations of renewable energy sources has grown rapidly. Especially photovoltaic station (PV) with the power in kWp are installed on family house roofs. Such PVs can be subsidized by the government in two regimes – first, a guaranteed purchase price is given, or second, the PV energy can be used in the green premium regime. In this regime, the PV energy is consumed directly by the house and only surplus power is sold. Moreover, the PV in-house-consumed PV energy is also subsidized. In this way, the photovoltaic station is economically most efficient, when most of its energy is consumed in the house – such energy receives the subsidy and saves payed energy from the global grid. An example of particular purchase price comparison of the PV consumed energy and global grid energy consumption is stated in Figure 1. We can see clearly that the described ”Green Premium” subsidy strongly motivates the user to consume the generated power locally as much as possible and so do not stress the distribution grid. Hence this policy is giving good opportunity for consumption profile optimization. The pressure for local consumption is even bigger then the motivation for savings (in the surplus hours). We have tested the application of the PAX consumption management on a family house with regard to the Green Premium subsidy. By setting the priority of the PV source higher than the priority of the grid power, the PAX system automatically moves the consumption into the area with unused PV power where possible.

58 4

Miroslav y aak, ˇs Prokop Miroslav Pr´ Pr´ ymek, mek, Aleˇ Aleˇss Hor´ Hor´ k, Luk´ and aLuk´ aˇs Prokop

Table 1. Optimization Results

generated consumed purchased sold premium overall price saved energy [kWh] [kWh] [kWh] [kWh] [CZK] [CZK] [kWh] [%] original 2 865 5 243 4 027 1 649 6 991 10 955 0 0% optimized 2 865 5 243 3 770 1 392 6 991 9 902 257 8,96%

The deferrable-operation appliances (DOA) is the best group for the optimization we are doing here – because we are trying to move as much consumption as possible to the PV output peek (noon) we must be able to defer operation of the devices by several hours. This is possible only with DOAs hence this type of appliances yields the most of the achieved optimization. The more appliances of this type a household has, the better the result of the optimization will be. For the household type for which this model was made, the results of the PAX optimization are summed up in the Table 1. In the first row, there are original values taken from the measurement of the given household consumption profile approximated for one year. As we can see, the PV power is about 50% of the household consumption and about 75% of the consumed power must be purchased from the grid operator at high price and about one half of the PV production is sold for eight times lower market purchase price. Although the household has a PV which covers substantial part of the household consumption, the PV’s profitability is in great part directed by the Green Premium subsidy. Without it, the overall power costs would remain high: without PV it would be 24 654 CZK and with PV 18 936-990=17 946 CZK, i.e. decreased only by 27%. The next row of Table 1 (”optimized”) shows the results of the model PAX optimization. As we can see, the amount of the purchased energy was lowered by 257 kWh, i.e. almost 9% of the PV production increasing the profitability of the PV ownership without any changes to the real consumption. As we have said, this number would be different for households with different appliance structure. The primarily aim of this work is to demonstrate that the intelligent demand-side consumption control is not just a theoretical concept but that it has a practical importance and a specific financial impact. Due to the PAX’s ability to combine the real and virtual appliances, the financial impact of the appliances control system application is easily predictable. The predictability of the outcome is surely the important property of the overall PV investment profitability judgment.

3

Conclusions

We have presented the application of Priority-driven Appliances Control System (PAX) to the problem of optimizing a family house power consumption with renewable energy sources with regard to the Green Premium policy, i.e. government subsidy supporting local consumption of the renewable energy.

Automatic to the Green AutomaticConsumption ConsumptionOptimization Optimization...wrt.the Green Premium Premium Policy Policy

595

On a model situation based on real family house measurements, we have shown that the application of the PAX demand side management can reduce the sum of expensive energy from the grid by 9% without any changes to the actual household consumption. An important aspect here is keeping the responsiveness of the system on a very high level. This means that a family house with higher number of deferrable-operation appliances than in our measurements can reach even substantially higher savings.

Acknowledgment This work has been partly supported by the Czech Science Foundation under the project 102/09/1842.

References 1. Egu Brno, s.r.o. Power Network Analyzer. Available from http://www.egubrno. cz/sekce/s005/pristroje/mds/mds_ostatni_3_5_u.html. 2. S. Misak and L. Prokop. Technical-Economical Analysis of Hybrid Off-grid Power System. In 11th International Scientific conference Electric Power Engineering, pages 295–300, 2010. 3. L. Prokop and S. Misak. Energy Concept of Dwelling. In 13rd International Scientific conference Electric Power Engineering, pages 753–758, 2012. 4. Miroslav Pr´ ymek and Aleˇs Hor´ak. Modelling Optimal Household Power Consumption. In Proceedings of ElNet 2012 Workshop, Ostrava, Czech Republic, 2012. VSB Technical University of Ostrava. 5. Miroslav Pr´ ymek and Aleˇs Hor´ak. Priority-based smart household power control model. In Electrical Power and Energy Conference 2012, pages 405–411, London, Ontario, Canada, 2012. IEEE Computer Society.

Accumulation of Electrical Energy from Solar Power Jan Vaculík1, Zdeněk Hradílek2, Petr Moldřík3 VŠB - Technical University of Ostrava, Accumulation of Electrical Energy from Solar Faculty of Electrical Engineering and Computer Science, Power Department of Electrical Power Engineering, 17. listopadu 15, 708 33 Ostrava, Czech Republic http://fei1.vsb.cz/kat410/ Jan Vacul´ık, Zdenˇek Hrad´ılek, Petr Moldˇr´ık 1) tel: +420 597 329 326 , email: [email protected] 2) Department of Electrical Power Engineering, tel: +420 597 235 , email: [email protected], ˇ 321 FEECS, VSB – Technical University of Ostrava, 3) tel: +420 596 999 email: [email protected] 17. Listopadu 15/2172, 708 320, 33 Ostrava-Poruba, Czech Republic [email protected], [email protected], [email protected]

Abstract. This paper describes our experience gained through practical operation of a low-temperature electrolyzer HOGEN GC600 with a proton exchange membrane that has been used for production of gaseous hydrogen in the fuel cells laboratory at VŠB -Technical university of Ostrava. The measurements contained in the paper illustrate laboratory research of hydrogen generation in the above mentioned electrolyzer and options for storage of hydrogen into various containers. The matter comprises research on impact of changes to parameters of this electrolyzer on efficiency of gaseous hydrogen production. Electric power needful for the electrolyzer supply is delivered from photovoltaic panels.

1 Introduction Solar radiation strikes the Earth's surface unevenly and its intensity depends on the season, time of the day, and local weather conditions. As far as utilisation of solar radiation for production of electric power by means of photovoltaic power plants is concerned, the process usually occurs upon request from the linked electric power system regarding supply of power at minimum or zero level. To eliminate negative impacts on operation of electric power networks due to unsolicited supply of power, the power output needs to be limited by certain means. Should the desired solution not include non-economical disconnection of photovoltaic power plants from the network, the power generated by these as excessive at the certain moment must be stored. One of the options, which is still undergoing the research stage, deals with storage featuring hydrogen technologies. Besides the description of laboratory system for storage of electric power into hydrogen built in our facility, this paper deals with further analysis of data obtained through operation of the system. Attention is paid mainly to the issue of efficiency of hydrogen production using electrolysis of water and potential means for its improvement. The existing operation of this system has revealed that the weakest link used in

c Zdenˇek Hrad´ılek (Ed.): ELNET 2013, pp. 60–66, ISBN 978–80–248–3254–8.

ˇ – Technical University of Ostrava, FEECS, 2013. VSB

Accumulation of Electrical Energy from Solar Power

61

the system is the hydrogen generator - low-temperature electrolyzer of PEM type. Research in the field of electrolytic production of hydrogen and its subsequent utilisation in fuel cells will be soon supported by launching the new laboratory of hydrogen technologies, which is being built within the Technological Centre Ostrava. Electrolyzer is, in terms of efficiency, the weakest element of electrical power hydrogen accumulation cycle. For this reason we have focused our attention on research aiming at enhancing the efficiency of connected Hogen GC600 electrolyzer. During last research we have found that the total efficiency of our island hydrogen system is less than 9 %. This very low efficiency is caused by production cycle of hydrogen. Our current research consists in the effort to improve the efficiency of used hydrogen generator (electrolyzer) Hogen GC600, because this electrolyzer is the least efficient part of hydrogen storage system. Hydrogen storage methods are also important part of our research. We use pressure vessels and metal hydride containers for gaseous hydrogen storage.

2 Hydrogen Storage System To accumulate electrical power with help of hydrogen technologies and at the same time use the advantage of zero-emission renewable energy sources (RES), hydrogen must be generated by water electrolysis. An osmotic unit in an electrolyzer processes modified water at the time of low load in electricity system or with use of electrical power from photovoltaic panels (at the Fuel Cells Laboratory, VSB-TU). Electrolyzer is the main component of hydrogen accumulation assembly as it affects the resultant technical and economic parameters of the hydrogen accumulation cycle. The performance of electrolyzer is directly proportional to surface area of electrodes, which represent the major share of the final price of this appliance. On the contrary, the amount of energy accumulated depends on the size of hydrogen storage container only. Hydrogen generation cycle contains two basic procedures: a) Generation - H2O electrolysis takes place in an electrolyzer which transforms electrical energy into chemical one the basis of synthetically produced fuel hydrogen. b) Storing, transportation of hydrogen resp.

2.1 Methods for Hydrogen Storage Storage of hydrogen is associated with specific hindrances. Hydrogen is highly reactive element of low density. Its molecules are small allowing hydrogen to diffuse through certain materials (plastic, some metals) both in liquid and gaseous state. If causes the so called „hydrogen embrittlement“ of metal structures it comes into contact with. There are several technical options to solve hydrogen storage. The most common method used deals with storage of liquid hydrogen in steel vessels under pressure of approximately 200 bar. (20 MPa).

62

Jan Vacul´ık, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

Welding vessels are used for storage of larger amounts of hydrogen. These vessels feature layered laminated walls. The internal wall layer is made from stainless steel to resist effects of pressure hydrogen. The exterior wall is made from steel suitable to withstand such pressure. In our environment, hydrogen is normally stored and distributed in pressure vessels at 200 to 350 bar. The use of pressure electrolyzers does not require any compression unit, as the gas is being compressed right inside the electrolyzer. This electrolyzer is used by the fuel cells laboratory at VSB - TU Ostrava: Hogen GC600 type. A more modern and very perspective alternate option is storage of hydrogen by means of the so called „metal hydrides“, when hydrogen becomes a part of chemical structure of selected metal alloys. Metal hydride storage systems make use mainly of metal alloys of nickel, magnesium, lanthanum, iron and titanium. This is the safest method for hydrogen storage, which is based on easy absorption of hydrogen into certain materials on metal basis, under higher pressure and lower temperature conditions. Therefore this is an exothermal reaction, i.e. absorption generates heat that needs to be drawn away. A reverse event – desorption, i.e. release of hydrogen from the specific material, is achieved by supplying heat within. Storage of hydrogen in metal hydrides does not require any extremely high pressure or cryogenic temperatures as if stored in pressure vessels or liquid state. As far as desorption of hydrogen is concerned, metal hydrides are divided into high- and lowtemperature ones. [1] Small quantities of hydrogen can be stored in simple pressure free float containers with water closure or underground gas storage reservoirs. Capacities of underground reservoirs range up to 108 m3.

2.2 Electrolyzers Hydrogen is acquired in a process of decomposition of demineralized water - electrolysis. If the energy used for the hydrogen production is generated with renewable energy sources, such as wind power generator or photovoltaics and burning of fossil fuels is avoided, the process of hydrogen generation is very clean and emission - free (no CO2, SO2, NOx, etc. emissions). Electrolyzer is a series of cathodes and anodes immersed in water with added electrolyte (often KOH potassium hydroxide) to increase conductivity. A polymerous electrolyte with ion exchange membrane seems to be very promising. Porous electrode layers are applied on both sides of the membrane and the current is conducted with H3O+ ion at the cathode and OH- ion at the anode. The thickness of the membranes is > 1mm. [3] These electrolyzers provide for the chemical reaction listed below: • Cathode: 4 H2O + 4 e-  2 H2 + 4 OH • Anode: 4 OH-  O2 + 2 H2O - 4e• Overall reaction: 2 H2O  2 H2 + O2 The negative electrode is usually made from nickel with platinum plating as catalyst to enable bonding of atomic hydrogen into molecules of H2 on the electrode surface to increase hydrogen production. If the cathode lacked catalyst, the atomic hydrogen

Accumulation of Electrical Energy from Solar Power

63

would accumulate on the electrode resulting in blocked current flow. The positive electrode is mostly made from copper and nickel. Its surface is covered with oxides of manganese, ruthenium or tungsten. These metals enable bonding of atomic oxygen to form molecules of O2. Separation of these two parts during unobstructed flow of ions inside electrolyzer is ensured by means of a diaphragm, based on asbestos and resistant to temperatures > 80 °C to prevent mutual blending of oxygen and hydrogen.[1] Potential compensation of electric power savings can be represented by heat supplied into the reaction. It is convenient mainly for the reason that the cost of heat is lower compared to electric power and rising temperature contributes towards efficiency of electrolysis. The minimum voltage for reaction is 1.228 V at 25 °C (298 K), yet the heat is insufficient, therefore it has to be supplied from ambient environment to prevent zero yield from production. Voltage increase to 1.47 V at the same temperature of 25 °C (298 K) results in increase of temperature in reaction, so there is no need for additional heat supply. If the voltage level rose even more, the excess heat would dissipate into the ambient environment. [2]

3 Laboratory measurements The task of laboratory measurements was to analyse the changes of efficiency of Hogen GC600 electrolyzer with different pressures on the electrolyzer panel. The pressure can be adjusted within the span 3 - 13.8 bar. This electrolyzer is shown in Fig. 1 and it´s parameters are shown in Tab. 1. Hydrogen flow within the span 0 - 600 cm3/min is also dependent on the set pressure. The level of the flow is displayed on the flowmeter connected between the electrolyzer and the hydrogen-storing cylinders. At the same time the date on the flow are stored in a computer with expert software. The data can be further analysed for needed overview and graphs.

Fig. 1 Hogen GC600 Electrolyzer

64

Jan Vacul´ık, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

Tab. 1 Product Specification Hogen GC600 [3] Maximum Hydrogen Flow Rate Delivery Pressure Hydrogen purity DI Water Tank Capacity Water Consumption (approximate) Power Outdoor temperature

0 - 600 cm3 / min 3 - 13.8 bar (45 - 200 psig) ± 5 % Full Scale Output, < 0.5 ppm Water Vapor 100.00 % 1.9 liters (approx 0.5 gallon) - Full Level to Shutoff Level 0.6 cm3/min at Full Rated Output, Equivalent to 0.9 liters per 24 Hours of Operation 100 - 240 VAC, 47 / 63 Hz 10 ° C / 35 ° C (min. / max.)

The voltage for Hogen GC600 electrolyzer use for hydrogen production was supplied from solar polycrystalline panels type Schott Poly 165 (it´s parameters are shown in Tab. 2) with the total installed output of 1980 Wp. The circuit was further provided with lead batteries determining the voltage level on direct bus as well as alternating load (200 W lighting units). The diagram in Fig. 2 shows mutual linkage of individual system parts, including location of measuring points. Electric values measured are processed using the NI USB-6218 measuring card. All the energy from solar panels was intended for hydrogen production using the electrolyzer, so there was no load connected to the alternate current bus. The intensity of solar radiation Me (W/m2) was sufficient both to supply the electrolyzer as well as to recharge batteries to 56 V as required. Table 2 The parameters of photovoltaic panels Type Nominal power (Wp) Voltage at nominal power (V) Current at nominal power (A) Open - circuit voltage (V) Short - circuit current (A) Module efficiency level (%)

SCHOTT POLY 165 ≥165 35.10 4.70 43.60 5.27 12.60

Fig. 2 Block Diagram of a Solar Hydrogen Storage System

Accumulation of Electrical Energy from Solar Power

65

3.1 Efficiency evaluation Graphics and table images illustrate generation cycle of hydrogen in Hogen GC600. The starting value of pressure is 7 bar (so-called “cold” condition) and the maximum value is 13.79 bar. Other adjusted values are taking into account a warmed up electrolyzer with pre-heated water which means higher efficiency of the process. Tab. 3 shows comparison of final values of all pressures as measured over the period of 30 minutes. Efficiency indicator was the most important one. The highest efficiency of hydrogen production cycle was achieved at the pressure of 10 bar. This is the hydrogen output pressure on electrolyzer. Efficiency is closely connected with another critical parameter - electrical power consumption during the generation cycle. As Fig. 3 shows, consumption decreases with rising pressure manually adjusted on the electrolyzer unit. Consequently (see Fig. 4), hydrogen generated during electrolysis is compared with equivalent energy quantum (EEQ). Tab. 3 Final values Consumption (Wh) 7 bar 8 bar 9 bar 10 bar 13.79 bar

233 232 200 176 190

Fig. 3 Comparison of consumption for various adjusted pressures

H2 volume (liters) 18,7 21,55 19,08 18,8 19,9

EEQ (Wh)

Efficiency (%)

60,03 69,18 61,25 60,35 63,88

25,76 29,82 30,63 34,29 33,62

Fig. 4 Comparison of hydrogen volume produced

5 Conclusion Measurement comprised setting of operation pressure levels of hydrogen on electrolyzer, specifically within the range between 7 and 13.79 bar. The duration of hydrogen production period per set pressure level was 30 minutes.

66

Jan Vacul´ık, Zdenˇek Hrad´ılek, Petr Moldˇr´ık

The measurement process experienced several drop-outs of the electrolyzer off the operating mode, when the cause for such failures was mainly due to low level of input de-mineralised water. Remedy of failure by operator always required repetition of the electrolyzer pressurising cycle. That was projected in increased consumption of electric power, which resulted in decreased efficiency of hydrogen production in electrolyzer. As far as research is concerned, such drop-outs can be deemed remote observations which were not considered during data evaluation. The effect of so called „blowoffs“ in electrolyzer leading to minor escape of hydrogen gas into ambient atmosphere, detected by the relevant apparatus. However, the resultant concentration of hydrogen in air never reached even the 10 % threshold of explosiveness that matches settings of safety sensors inside laboratory. The final stage was associated with comparison analysis of data obtained by measurement for all hydrogen pressure levels set on the electrolyzer output.. The data obtained from measurement was used to evaluate the efficiency of hydrogen production. The weakest efficiency (25.76 %) was achieved with pressure of 7 bar. The production cycle for pressure of 7 bar shows the worst values in all aspects. Except for the highest pressure (13.79 bar), the efficiency of hydrogen production increased with rising pressure, while the electric power was decreasing. As far as efficiency (34.29 %) is concerned, the best results were rendered with hydrogen pressure level of 10 bar. That did not confirm the assumption that the highest efficiency of hydrogen production was achieved at the highest pressure level (33.62 %). Another parameter definitely affecting efficiency of hydrogen production is the temperature of reaction de-mineralised water at input to the Hogen GC600 electrolyzer. Further research will be aimed in this respect.

Acknowledgements This work was supported by the project ENET - Research and Development for Innovations Operational Programme No. CZ.1.05/2.1.00/03.0069, by the Czech Science Foundation - project No. GAČR 102/09/1842, and by the Ministry of Education, Youth and Sports of the Czech Republic (No. SP2013/137).

References 1.

2. 3.

Vaculík, J., Moldřík, P., Hradílek, Z., Minařík, D.: Storing solar energy in hydrogen. In Proceedings of the 7th International Scientific Symposium on Electrical Power Engineering ELEKTROENERGETIKA 2013. Technical University of Košice, 2013, vol. 7., výhledově v časopise Acta el.et inf. (ISSN 1335-8243 / 1338-3957) , p. 172175, ISBN 978-80-553-1441-9 BALAJKA, Jiří. Vodík a iné nové nosiče energie. Bratislava : ALFA, 1982. 303 s. HOGEN® GC 300 and 600 Laboratory hydrogen generators User’s Manual.

Author Index

Bedn´ aˇr, Pavel, 46

Nov´ ak, Martin, 26

Goˇ no, Radom´ır, 7, 26, 32

Prokop, Luk´ aˇs, 55 Pr´ ymek, Miroslav, 55

Hor´ ak, Aleˇs, 55 Houdek, V´ıt , 7 Hrad´ılek, Zdenˇek, 1, 20, 39, 60

Rozehnal, Petr, 13 Rusek, Stanislav, 7, 32

Chovanec, Peter, 46 Janˇsa, Jiˇr´ı , 1 Kral, Vladimir, 32 Kr´ atk´ y, Michal, 32, 46 Krejˇc´ı, Petr, 13 Moldˇr´ık, Petr, 1, 60 Mozdˇreˇ n, Tom´ aˇs, 7

Slivka, Martin, 32 Smoˇcek, Martin, 20 ˇ amek, Jan, 39 Sr´ Unger, Jan, 13 Vacul´ık, Jan, 60

Editor:

Zdenˇek Hrad´ılek

Title:

ELNET 2013

Place, year, edition:

Ostrava, 2013, 1st

Page count:

80

Edit:

ˇ – Technical University of Ostrava, VSB 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic

Impression:

100

ISBN 978–80–248–3254–8