Reliability Measures and Risk Assessment Andrew Slone, Engineer, Reliability Performance Analysis, NERC EPRI Workshop, February 23, 2012
Overview • Reliability Measures
2011 Risk Assessment of Reliability Performance Report 2012 State of Reliability Report Overview of ALR Metrics and Trends Severity Risk Index
• Future Integrated Reliability Indicators
2
RELIABILITY | ACCOUNTABILITY
Reliability Improvement
Reliability Measures
? Reliability ConditionDriven Indicators
? ? ? ?
?
Organization? Effectiveness Metrics
? Standards/ StatuteDriven Metrics ?
?
System/EventDriven ? Metrics
Organizational Effectiveness 3
RELIABILITY | ACCOUNTABILITY
2011 Annual Report •
Annual Report “2011 Risk Assessment of Reliability Performance”
Culminates a 3 year process to provide a view of risks to reliability Approved by the NERC Board of Trustees on August 4, 2011 Available at: http://www.nerc.com/docs/pc/rmwg/2011RMWG_Annual_Report.pdf
Condition Driven
Events Driven
Standards Driven
Integrated Measures and Analysis
State of Reliability Report 4
RELIABILITY | ACCOUNTABILITY
Data Source Integration and Analysis
Metrics
TADS Integration and Analysis
GADS
EVENTS
DADS Events Driven
Critical Infrastructure Protection 5
Condition Driven
Standards Dev & Prioritization
Standards Driven
Compliance
Events Analysis
RELIABILITY | ACCOUNTABILITY
2012 State of Reliability Report
• New in 2012 State of Reliability Report Tighter integration of sources to ascertain historic Bulk Power System reliability Focus on providing relevance for both industry and regulators Additional Data Sources
6
RELIABILITY | ACCOUNTABILITY
Transition to 2012 State of Reliability Report
7
RELIABILITY | ACCOUNTABILITY
Metrics - Improving Trends ALR
Title
1-3
Planning Reserve Margin
1-4
BPS Transmission Related Events Resulting in Loss of Load
2-5
Disturbance Control Events Greater Than Most Severe Single Contingency
6-2
Energy Emergency Alert 3
6-3
Energy Emergency Alert 2
6-11
Automatic AC Transmission Outages Initiated by Failed Protection System Equipment
6-12
Automatic AC Transmission Outages Initiated by Human Error
8
RELIABILITY | ACCOUNTABILITY
Metrics – Inconclusive Trends ALR
Title
2-3
Activation of Under Frequency Load Shedding
2-4
Average Percent Non-Recovery of Disturbance Control Standard (DCS) Events
4-1
Automatic AC Transmission Outages Caused by Protection System Equipment-Related Misoperations
6-12
Automatic AC Transmission Outages Initiated by Human Error
6-14
Automatic AC Transmission Outages Initiated by Failed AC Circuit Equipment
9
RELIABILITY | ACCOUNTABILITY
Metrics – New/No Data ALR
Title
3-5
IROL/SOL Exceedance
6-1
Transmission Constraint Mitigation
6-15
Element Availability Percentage
6-16
Element Unavailability Percentage
1-5
System Voltage Performance
1-12
Interconnection Frequency Response
10
RELIABILITY | ACCOUNTABILITY
Severity Risk Key Concepts
11
RELIABILITY | ACCOUNTABILITY
Severity Risk Index (SRI)
• Focus on significant events and quantify the historic system performance using daily outage data GADS and TADS daily outages, forced only MW load loss and restoration duration from disturbance event reports Could be considered similar to a daily, quarterly or yearly SAIDI metric for distribution systems If everything was out of service, all load was unserved, the day’s score would be 1000 If no lines or units were out of service and no load was lost the day’s score would be 0 Normal “good days” will measure higher than 0 12
RELIABILITY | ACCOUNTABILITY
Severity Risk Index (SRI)
13
RELIABILITY | ACCOUNTABILITY
Integrated Reliability Indicators (IRI)
• Develop the conceptual model of reliability risks • Establish quantitative measures for evaluating the performance of the indicators • Support risk-informed decision making • May be able to correlate over the long term to predictive measures of reliability risks and if so… • Estimate the effectiveness of reliability risk reduction and/or mitigation
14
RELIABILITY | ACCOUNTABILITY
EDI Calculation DurationDays SRI EDI DurationDays • The EDI is derived from Severity Risk Index (SRI) • SRI = wT ∙(NT) + wG ∙(NG) + wL ∙(LLS) • Where: 15
SRI = severity risk index for a specified event, wL = weighting of load loss, LLS = normalized MW of Load Loss in percent, wT = weighting of transmission lines lost, NT = normalized number of transmission lines lost in percent, wG = weighting of generators lost, NG = normalized number of generators lost in percent RELIABILITY | ACCOUNTABILITY
SRI Calculation SRI
Gen T Weighted _ Loss MW _ LOSS 0.3 0.1 0.6 LLS Genloss _ base T loss _ base
• Load Loss and duration (for 80% of load restored) Restoration Promptness Level (RPL) o RPL = 1, if restoration < 4 hrs o RPL = 2, if 4 =12 hours
Load Loss Severity (LLS) = (RPL/3)*(Load Loss MW/Daily Peak Load)
16
RELIABILITY | ACCOUNTABILITY
SRI Base (Denominator) for Normalization
• Normalization Base Use TADS inventory (200 kV and above) for Tloss_base o Yearly AC Circuits and Transformers inventory data o Average MVA rating for each voltage class (see next slide)
Use unit MW Ratings in ESD as Gloss base o ESD – Electricity Supply and Demand database
• Aggregation level Interconnection NERC -Wide 17
RELIABILITY | ACCOUNTABILITY
SRI Base (Denominator) for Normalization (2)
• Average MVA capacity for AC Circuits o 765 kV → 3000 MVA o 500 kV → 2000 MVA o 345 kV → 1300 MVA o 230 kV → 700 MVA
• Example of AC circuit loss event in EDI: 100 – 230kV, 40 – 345 kV, 10– 500 kV, 3 – 765 kV: NT
18
100 700 40 1300 10 2000 3 2000 2708 700 1078 1300 344 2000 33 3000 151,000 4,084,000 0.037 RELIABILITY | ACCOUNTABILITY
Event Driven Index (EDI) - Example
• Using the combined risk index score SRI, the 2009 risk events summed to 506.117 points; the top five day’s SRI values were: 1/27/09 4.98 SRI points, 1/28/09 3.67 SRI points, 12/15/09 3.22 SRI points
7/1/09 3.70 SRI points 2/11/09 3.32 SRI points
• If EDIyear = success rate for BPS risk events, then EDI2009 = 100 x (365 x 1000 – 506.117)/(365 x 1000) ≈ 99.8613
19
RELIABILITY | ACCOUNTABILITY
Three-Tier Risk Measures
20
RELIABILITY | ACCOUNTABILITY
The Risk Control Reduction Cycle
• Find potential risks to reliability.
• Prioritize the risk clusters to find those risks which are the most severe.
Risk Cluster Identification
Prioritization
Intelligence and Analysis
Solve Problems • Solve the problems to eliminate potential risks to reliability.
21
Actionable Risk Control Steps
• Develop actionable risk control steps.
RELIABILITY | ACCOUNTABILITY
IRI Concept
Event Driven Index (EDI) Measures Risk from Major System Events
Uses the roster of reliability metrics developed by RMWG and approved by OC/PC
22
Condition Driven Index (CDI) Monitors Risk from Key Reliability Metrics
Based on event severity risk index (SRI) values and turns into an availability index
Standards/Statute Driven Index (SDI) Measures Risks from Severe Impact Violations
Identify a subset of standards that have highest impact to reliability and create an index of standards violations
RELIABILITY | ACCOUNTABILITY
IRI Concept (2) • BPS Integrated Reliability Index (IRI): Event Driven Index (EDI): Based on event severity risk index (SRI) values Condition Driven Index (CDI): Use a subset of metrics based on selection criteria Standards/Statute Driven Index (SDI): Identify a subset of standards that have highest impact to reliability and create an index of standards violations
23
RELIABILITY | ACCOUNTABILITY
EDI Trends by Quarter
Event Driven Index (EDI)
100.00 99.95
99.90 99.85 99.80 99.75 99.70
24
RELIABILITY | ACCOUNTABILITY
Standards/Statute Driven Index (SDI) Proposed Approach
• Severe Reliability Impact Statement (RIS) RIS indicates significance of impact on BPS
• • • •
25
Uses Violation Risk Factor (VRF) Demonstrates Violation Severity Level (VSL) Similar to daily SRI calculation Almost 4 year history violation history of 4,410 confirmed violations
RELIABILITY | ACCOUNTABILITY
SDI Calculation • SDI = (Total Compliance - ∑(wV∙NV/NR))/(Total Compliance) • Range from 0 to 100 • Total Compliance = 1.0 * Number of Days in a specific period • wV = weighting of a particular high risk req. violation • NV = number of violations for the selected high risk req. • NR = number of registered entities required to comply with the high risk req. 26
RELIABILITY | ACCOUNTABILITY
SDI Calculation Example • SDI = 100 ∙ (Total Compliance - ∑(wV∙NV/NR))/(Total Compliance) • SDI = 100 ∙ (1.0 ∙ Days in a Quarter - ∑(wV ∙ NV/NR)) / (1.0∙Days in a Quarter) • 1Q2009: 40 violations (PRC-005 R1) RIS = Severe, VRF=High, VSL=Severe A total of 335 TO, 828 GO registered entities SDI = 100 x (1.0 x 90 - 0.038 x 40/(335+828))/(1.0 x 90) = 99.99
27
RELIABILITY | ACCOUNTABILITY
Possible SDI Trend by Quarter
28
RELIABILITY | ACCOUNTABILITY
Condition Driven Index (CDI)
• Uses metrics approved by OC/PC • Applies SMART criteria to rank the relative importance and weighting for performance • 5 trend ratings are determined per metric performance 29
Significant improvement Slight improvement Inconclusive Slight Deterioration Significant Deterioration RELIABILITY | ACCOUNTABILITY
Possible IRI Attributes • • • •
Scale from 0 to 100 100 would indicate a perfect performance Weighting should be developed and can be adjusted Aggregation at NERC, Interconnection and Region levels • Reporting period – targeting quarterly, reported at year end • Trial indicators until sufficient history is experienced
30
RELIABILITY | ACCOUNTABILITY
IRI Objectives
•Inform industry leaders •Increase transparency •Quantify action effectiveness of risk reduction •Meaningful bulk system performance trends •Guidance on improving system reliability •Support with risk‐informed decision making
31
RELIABILITY | ACCOUNTABILITY
Working with EIA and Sandia
• EIA’s and Sandia's statistical analysis experience and risk-control technologies Leverage metric development experience for the Nuclear Regulatory Commission (NRC) and the National Aeronautics and Space Administration (NASA) Statistical significance and confidence interval determination Integrated strategy to apply science, technology, and engineering capabilities
32
RELIABILITY | ACCOUNTABILITY
IRI Future Steps Present • Severity Risk Index • Adequate Level of Reliability metrics • TADS • GADS • Event Analysis Database
33
IRI Process • Statistically link reliability reduction cause & effect • Link event, precursor indicators, and outage datasets to explore initiating events • Find top initiating events • Develop reliability index framework
Future • Use IRI to quantitatively measure BES reliability • Focus on improving top initiating event performance • Help focus industry on relevant reliability risks to improve BES reliability
RELIABILITY | ACCOUNTABILITY
Questions and Answers
34
RELIABILITY | ACCOUNTABILITY