23rd International Conference on Electricity Distribution

Lyon, 15--18 18 June 2015 Paper 0970

COST BENEFIT ANALYSI ANALYSIS S OF HOUSEHOLDS ENERGY BOXES DEPLOYM DEPLOYMENT IN EUROPE: IMPACT OF TH THE E SPOT PRICES. Gaspard LEBEL Raphaël CAIRE CAIRE,, Nouredine HADJSAID Univ. Grenoble Alpes, G2Elab G2Elab, F-38000 38000 Grenoble, France CNRS, G2Elab F-38000 38000 Grenoble, France [email protected] [email protected]

ABSTRACT The present paper presents a study willing to assess the profitability of replacing a conventional thermostat by a Smart Thermostat – also called Energy Boxes – in electrically electric heated households. It assumes that such replacement is a prerequisite for a dweller to take the entire advantage of a subscription at Real Time Price (RTP) tariffs. A deterministic model has analysed the Day-ahead Day ahead Nord Pool Spot data available from 2001 to end of 2014 for Denmark, Norway, Finland, Sweden, Estonia, Latvia and Lithuania. It computes these data with hourly outside temperature series of one unique weather station (Malmö) into a theoretical household. The model enables to get load profiles and annual energy bill. The dweller welfare modification is finally shortly analysed and a payback analysis is provided based on simulated bills and a benchmark of the Smart Thermostat available on the mar market. The main outcome of the study is that the profitability for the end end-user user of RTP adoption is highly dependent of the national power market context.

INTRODUCTION The large diffusion of ICT technologies into domestic housings affords new possibilities of end-users end users services, centered on the Connected Home. The concept of the Internet of Things (#IoT) gives a new look to the old oldfashioned concept of home automation would have never really convinced the dwellers over the last decades. Even though the success of the Connected Home is still not guarantee, the ease of use of smartphones and the reduction of the communication cost increases its chance of success. Among the portfolio of #IoT devices, the present paper expects to look at load controllers dedicated for space heating control. These ones are now usually commercialized under the designation of Smart Thermostats. Basically, a thermostat is a device which controls the heating system by sending it ON/OFF set points thanks to a steady-state steady state relay which is driven by a temperature sensor and eventually a schedule or a presence sensor. On top of these initial components components,, a Smart Thermostat embeds as well computing capabilities in a controller and a communication channel. The addition of a current sensor to such such Smart Thermostat enables to monitor as well the heaters power consumption. The embedment of

CIRED 2015

Stéphane BEDIOU, BEDIOU Alain GLATIGNY Schneider Electric Industries Energy Division stéphane.bediou@schneider sté[email protected] electric.com this functionality leads to switch to connected devices also known an Energy Boxes. For the objective of optimized space heating, the availability of a communication communication channel could first enable to receive dynamic power tariff information from the retailers, like Real Time Price (RTP). It enables then to get outside temperature forecast. Such information, associated with heaters power consumption and indoor temperature temperature historic enable to compute the thermal characteristic of the accommodation. From there, it makes possible to forecast accurately the hourly heat load required to comply with an indoor temperature set point over several hours. The core idea of the present present paper is so to assess the added value for an end-user end user heated by electric heaters, to dispose of these functionalities by replacing its basic thermostat by a Smart Thermostat. The questions asked are finally: • Is such investment willing to reduce it itss energy bill? • What would be the pay-back pay back of such investment? • Are there any other success condition conditions to consider, like the power market context context?

SIMULATION MODEL To answer these questions a specific hourly hourly-based based deterministic model has been developed. It assesses the profitability of domestic endend-users users flexibility, enabled by Energy Boxes commissioning. The object used for the simulation is a theoretical domestic hhousehold ousehold where the electric heaters are controlled a by power power-monitored monitored Smart Thermostat, i.e. an Energy Box Box.. The model assumes a perfect knowledge of the outside temperature, power market prices and the thermal characteristics of the household. In such a way the result obtained are the ideal ones. Fig. 1 provides a global view of the system simulated.

Figure 1: Global view of the simulated system Simulation model

1 1/5

23rd International Conference on Electricity Distribution

Lyon, 15--18 18 June 2015 Paper 0970

Model of the Household The modeled household [6] is a standard Swedish household located ar around ound Malmö (Sweden). It is heated day and night at least at 21°C by electric heaters. The heating system has a nominal power of 6kW/household. The decision not to consider a night cooling-off cooling off in the simulation has been inspired by open discussions with end-users end users consumption experts from Vattenfall Market Analysis and Future Power System team that occurred in 2012. Moreover, aall ll of these heaters are already connected to a central thermostat through local connectivity. connectivity. This is a key restriction to keep in mind. The heating season for which heat costs are calculated is considered per civil year and leads from Jan 1st to Apr 30nd and from Oct 1st to Dec 31st, i.e. 5064 hours long. The household is characterized by two thermal coefficients: coeffici λλ, its global heat losses coefficient (in W/°C) and τ,, its thermal constant (in hr) hr). λ corresponds to the mean heat losses coefficient of Southern Sweden households, and has been calculated (1) by considering the mean annual power consumption used in 2010 for electric heating in the households surrounding Malmö ([1], ( table 3.34) divided by the Heating degree days of the area. The value of degree days has been calculated by summing the difference between an inside set temperature of 21°C and the outsid outsidee daily temperature of Malmö in 2010 for each day of the heating season. The value chosen for τ (150hr1) corresponds to a time constant of concrete concrete-built households [2]..

Data sets The hourly Spot price and volume data of Denmark, Norway, Finland, Sweden, Estonia, Latvia and Lithuania have been used for the simulations [3]. In the countries where the Spot markets are split in several prices area, only the ones where the prices are the cheapest and the most expensive have been considered. The bidding areas considered are listed in Table 1. 1. These Spot data have been matched with one unique set of outside temperature: the hourly temperatures measured in Malmö [4]. The period of study lleads eads from 2001 to 2014. The choice to test the market data of several countries on the same thermal system has been motivated by the wish to assess the impact of the market framework on the Energy Box profitability.

End-user End user tariff definition The end-user user tariffs applied to the model include only the part of the tariff related to power production. Thus it does not include the transmission and distribution taxes neither neither the public authorities taxes. Two tariffs are used. First, the Real Real Time Price rice (RTP) tariff iff which is strictly equal to the Day Day-ahead ahead Spot prices. Then the Fix tariff, 1

It has been noticed during the simulation that τ does not affect the results on a significant manner for any values contained in a range [75;150hr]

CIRED 2015

which correspond to the mean Day Day-ahead ahead Spot prices, pondered by the turnover over the year. This Fix tariff has been calculated calculated for each year according to the equation (2). Such Fix tariff aimed to represent the mean cost of power production regardless its time of production. These annual fix prices are displayed in Table 1. 1 For simplification issues, the hourly spot price taken for the countries split into several bidding area (Denmark, Norway, Sweden), is the average of the price in application in each of the bidding area area,, regardless the turnover per price area. area Equation quation (3) presents the method of computation. Moreover, it it is assumed that the market penetration of RTP tariff for domestic consumers stays marginal, so that Smart Thermostat spreading is not willing to modify the Day-ahead Day ahead market level of prices and profiles. (1)



%( ) %( * % %

∑+∈-1;1202∪-273;36522

TUVW * , '°R

∑?@AB % CDE 8 (9:; %,X ∗8 (9ZU[V\> %,X

(2)

∑?@AB 8 (9 (9ZU[V\> %,X CDE 8

∀

(3)

⁄%( ). 9 N ( * 'Q._`a% 'Q

'°RS



2001 2001 … 2014

FGH