DATA ANALYTICS FOR RESILIENT DISTRIBUTION SYSTEMS

Transcription

1 NSF Real Time Data Analytics Workshop, Portland, Aug DATA ANALYTICS FOR RESILIENT DISTRIBUTION SYSTEMS 1 Chen-Ching Liu American Electric Power Professor Director, Power and Energy Center Virginia Tech (Research Professor, Washington State University) Sponsors Include National Science Foundation, Pacific Northwest National Lab, Department of Energy, European Commission

2 SMART DISTRIBUTION GRID ENVIRONMENT 2

3 Outage Management Using Smart Meters Up to several hours to collect trouble calls for outage management Smart outage management using notifications from smart meters Distribution Operating Center CIS DMS User Interface User Interface IEC DNP 3.0 SCADA network Router /Firewall OMS Router /Firewall Smart Meters Router /Firewall Customers User Interface MDMS ANSI C12.22 VMAX,ZigBee Pullman Smart Grid Smart Outage Management System (OMS) 3

4 Outage Management Incorporating Smart Meters One-line diagram of a distribution system: Brk Auto. R1 Auto. R2 FI1 Feeder L1 F1 F2 L2 F3 F4 SM11 SM12 SM1 SM2 SM3 SM4 SM5 SM6 SM7 SM8 L3 L4 SM9 SM10 SM13 SM14 F: fuse SM: smart meter L: lateral Evidence: Overcurrent flags from FI1, R2, R1; Outage reports from smart meters downstream of Fuse3. OMS: Determine the actuated protective device Determine the faulted line section

5 Multiple-Hypothesis Incorporating Smart Meters Generate Hypotheses Evidence Optimization Model Credibility of Hypotheses Most Credible Outage Scenario(s) Y. Jiang, C. C. Liu, M. Diedesch, E. Lee, A. Srivastava, Outage Management of Distribution Systems Incorporating Information from Smart Meters, IEEE Trans. Power Systems.

6 GE Grid Solutions DMS (WSU Testbed) 6

7 Service Restoration with DA 1. Fault occurs 2. Open CB 3. Find fault 4. Isolation *KEPCO: Intelligent DA System 7

8 Service Restoration with DA 5. Transfer outage area 6. Execute restoration plan 7. Field crew C.C. Liu, S.J. Lee, S.S. Venkata, An Expert System Operational Aid for Restoration and Loss Reduction of Distribution Systems IEEE Trans. Power Systems, May S. J. Lee, S. I. Lim, B. S. Ann, Service Restoration of Primary Distribution Systems Based on Fuzzy Evaluation of Multi-Criteria, IEEE Trans. Power Systems, Aug *KEPCO: Intelligent DA System 8

9 Distribution Restoration: Spanning Tree Method Radial structure of the distribution network can be represented by a spanning tree (a circuit graph). Restoration can be formulated as a problem of finding a desired spanning tree structure and a sequence of operations that change one spanning tree into another. F2 1 2 F1 Load Node Closed Switch(branch) F5 21 F4 F3 Open Switch(branch) (F1,F2,F3,F4,F5) Spanning Tree software has been integrated into PNNL GridLAB-D and available on GridLAB-D web site J. Li, X. Y. Ma, C. C. Liu, K. Schneider, Distribution System Restoration with Microgrid Using Spanning Tree Search, IEEE Trans. Power Systems, Nov. 2014, pp

10 Damages to Distribution Grids by Superstorm Sandy Downed power lines and other debris litter the streets of Seaside Heights, N.J., on 31 October 2012, two days after Superstorm Sandy made landfall in the US.* The storm surge that accompanied Superstorm Sandy sent water rushing through the streets near a substation in Brooklyn, N.Y. Restoring a flooded substation takes much longer than restoring a downed power line because of the large amounts of water, rust, and mud left trapped in the structure.* * Source: Nicholas C. Abi-Samra, One Year Later: Superstorm Sandy Underscores Need for a Resilient Grid, IEEE Spectrum, 10

11 Typical Outages and Catastrophic Outages Due to Extreme Events Typical Outages Single fault: In most cases, there is only one faulted component. Small amount of load and a small number of customers are involved. Power is available: Most power sources are working and stay connected. T&D network remains intact: Outage loads are easily connected to sources. Easy to repair and restore Catastrophic Outages Multiple faults: Multiple electrical facilities are damaged. Large amount of load and a large number of customers are out of services. Lack of power: Power sources can not access the load or are out of service. T&D network damaged: Overhead lines, transformers, substations are damaged. Difficult to repair and restore C. C. Liu, Distribution Systems: Reliable But Not Resilient, In My View, IEEE Power and Energy Mag., May/June

12 Resilience: Definition and Metric PPD21 Definition:..ability to prepare for and adapt to changing conditions and withstand and recover rapidly from disruptions.. * Metric: System performance and resilience Resilience refers to total amount of ENERGY served to critical load during the preset time period For distribution power systems, system performance F(t) can be specified as the total amount of power serve to critical load at time t * Office of the Press Secretary of the White House, Presidential Policy Directive 21 Critical Infrastructure Security and Resilience [Online]. Available: 12

13 Resilience: Definition and Metrics System Performance in Natural Disasters System Performance pre-event event progress restorative state infrastructure recovery t e t r t ir Metric: cumulative service time to critical loads weighted by priority** Critical Load: those necessary for maintaining basic societal functions, such as hospitals and street lighting** Time * Office of the Press Secretary of the White House, Presidential Policy Directive 21 Critical Infrastructure Security and Resilience [Online]. Available: ** XU Y, LIU C-C, GAO ** H. Reliability analysis of distribution systems considering service restoration, in Proc. IEEE PES Conf. Innovative Smart Grid Technologies, Feb. 2015, pp

14 Microgrid Supports Fast Recovery of Distribution Systems Opportunities When a major outage occurs, microgrids can be controlled to provide an efficient service restoration strategy to restore critical loads in a distribution system and improve resilience. Challenges: Distributed generators (DGs) in a microgrid have relatively small capacity. They have limited ability to absorb shocks and maintain stability Scarcity of generation resources: Fuels for generators, e.g., diesel and natural gas, electric energy in storage devices are limited and hard to support after an extreme event Restoration scheme considering DERs and Microgrids 14

15 Microgrid Supports Fast Recovery of Distribution Systems Problem Formulation Objective: Maximize cumulative service time to critical loads weighted by their priority Constraints: Dynamic, generation-resource, operational, and topological constraints are considered. Algorithm Introduce concept of restoration tree and load group. Transform restoration into maximum coverage problem, which is solved by mixed integer linear program (MILP). Case Study & Application PNNL 1069-node test system. Pullman Regional Hospital and City Hall, WSU Microgird field test max c i t i XU Y, LIU C-C, SCHNEIDER K P, et al. Microgrids for service restoration to critical load in a resilient distribution system. IEEE Trans. Smart Grid. i 15

16 Microgrid Supports Fast Recovery of Distribution Systems PNNL 1069 Test System Restoration Path M1 -Z39-Z7-Z6-Z23-Z5- Z18-Z8-Z30-Z4-Z21 -Z37- Z75-Z48-Z58-Z45-Z90- Z92(CL3) M1 -Z39-Z7-Z19-Z2(CL1)- Z14-Z16-Z9-Z78-Z54-Z56- Z49(CL2) M3-Z93(CL4) CL5 cannot be restored. CL5 is closer to M4 and operational constraints are satisfied. However the dynamic constraints are not satisfied due to the limited capacity of Microgrid 4 16

17 Microgrids for Resilience Enhancement -- PNNL 1069 Node Test System System function curve of distribution system with microgrid System performance 100% 90% Time (hour) System function curve of distribution system without microgrid System performance 100% 90% Time (hour) A fault occurs at zone 139 Mean time to operate a manual switch is assumed to be 60 minutes LI J, MA X-Y, LIU C-C, et al. Distribution system restoration with microgrids using spanning tree search. IEEE Trans. Power Syst.,

18 3-D Substation Scene 18

19 Physical Security: Video Surveillance System Video cameras Video streams On-site servers for intelligent surveillance Detection module Locating module Marked video stream Storage of video records Humanmachine interface 2-D grid map data Database Visualization tool Data fusion Tracking module Patrolling module Entire data grid 19

20 Image Processing - Detecting Image processing in each frame of the video stream 20

21 Image Processing - Tracking Even for a well-trained operator, controlling two PTZ cameras simultaneously is not a practical task without computer assistance; Automatically move the PTZ cameras to keep the intruder in the middle of image plane. Tracking intruder with computer assistance 21

22 Drill in CEPS July 16-17, 2013 (Prague, Czech Republic) 22

23 Drill in CEPS - Scenario 23

24 Further Information Y. Xu, C. C. Liu, K. Schneider, F. Tuffner, and D. Ton, "Microgrids for Service Restoration to Critical Load in a Resilient Distribution System," IEEE Trans. Smart Grid. Y. Jiang, C. C. Liu, M. Diedesch, E. Lee, A. Srivastava, Outage Management of Distribution Systems Incorporating Information from Smart Meters, IEEE Trans. Power Systems. J. Xie, C. C. Liu, M. Sforna, M. Bilek, and R. Hamza, "On Line Physical Security Monitoring of Power Substations, Int. Trans. Electrical Energy Systems, June Y. Xu, C. C. Liu, K. Schneider, and D. Ton, Toward a Resilient Distribution Systems, IEEE PES General Meeting, Jul J. Li, X. Y. Ma, C. C. Liu, K. Schneider, Distribution System Restoration with Microgrid Using Spanning Tree Search, IEEE Trans. Power Systems, Nov. 2014, pp S. I. Lim, S. J. Lee, M. S. Choi, D. J. Lim, and B. N. Ha, Service Restoration Methodology for Multiple Fault Case in Distribution Systems, IEEE Trans. Power Systems, Nov C. C. Liu, S. J. Lee, K. Vu, "Loss Minimization of Distribution Feeders: Optimality and Algorithms" IEEE Trans. Power Delivery, April 1989, pp C.C. Liu, S.J. Lee, S.S. Venkata, An Expert System Operational Aid for Restoration and Loss Reduction of Distribution Systems IEEE Trans. Power Systems, May 1988, pp