Embed Size (px)
Citation preview
2014 Smart Grid R&D Program
Peer Review Meeting
(Microgrids as a Resiliency Resource)
(Kevin Schneider - PNNL)
(Chen-Ching Liu - WSU)
(6/11/2014)
PNNL-SA-103186
December 2008
Microgrids as a Resiliency Resource
Objective
Life-cycle Funding
Summary ($K) Prior to
FY 14
FY14,
authorized
FY15,
requested
Out-year(s)
PNNL* 250 380 300 600
WSU 188 150 166 320
• In FY13 GridLAB-D Analysis contains three focus areas,
“Micro Grids as a Resiliency Resource” is one area.
Technical Scope
To increase the resiliency of the nation’s electric
infrastructure via the integration of microgrids.
Increased resiliency can be achieved by using
microgrids to support their own critical loads,
community critical loads, and/or to act as a black start
resource.
To use distribution level dynamic simulations to
determine the capability of microgrids to act as a
resiliency resource. Simulations are conducted
on the Washington State University system, as
well as portions of the Avista Utilities system.
Operational generator tests, and an AMI
diagnostics system, are used to validate
simulation results. The validated simulations are
then used to determine to what extent microgrids
can be used as a resiliency resource. 2
December 2008
Problems and Needs
• Wide area outages due to extreme weather events are becoming more common in the United States, and because of their scale, building more central generators and transmission lines will not address the associated problem.
• Microgrids have proven themselves capable of supplying load when utility service is lost. Despite this benefit, the high cost of micro grid deployment, and operations, has resulted in only a modest number of deployments.
• To increase the adoption rate of microgrids, by increasing their value, it is necessary to examine the full spectrum of benefits that can be realized. This project focuses on values derived from their use as a resiliency resource.
• Customer Resource • Community Resource • Black Start Resource
• To facilitate the utilization of microgrids as a resiliency resource, analysis capabilities are needed that will allow for designers and integrators to have greater flexibility in exercising their engineering judgment.
• This work focuses on the development of dynamic simulation capabilities to support both the design, and the development, of operational strategies.
3
December 2008
Significance and Impact
• Improved resiliency can be broken into three phases:
• Prevention: best practices in advance to “harden” the system • Survivability: the ability of the system to react to real-time events • Recovery: the ability to rapidly restore end-use load after disruptive events
• Microgrids have the ability to improve the survivability of distribution systems, and to support recovery.
• Resiliency resource potential of microgrids
• Customer Resource: the traditional application of microgrids where only the local loads are supplied.
• Community Resource: an extended application of microgrids where critical loads outside of the normal microgrids boundaries are supplied.
• Black Start Resource: a new application where microgrids are used to support the auxiliary loads of a thermal plant that does not have native black start capabilities.
4
December 2008
Technical Accomplishments
• Developed a generalized method of conducting dynamic simulations at the distribution level.
• Open Source: Implemented in the GridLAB-D simulation environment.
• Simplicity: Models are easier to build than a full electromagnetic model.
• Detailed: A level of detail consistent with transmission dynamic solvers.
• Developed a dynamic model of the WSU and Pullman system that has been used to examine the use of microgrids as a community resource.
• The model contains the WSU system, and portions of four Avista feeders.
• This model will be used to evaluate the dynamic impacts on switching, protection, and motor controllers.
• Two portions of the dynamic model must be validated:
• Generators: Currently validating generator models based on operational testing on two of the three WSU generators.
• System: In the process of developing an AMI-based diagnostic
system to evaluate the system model. (This will be discussed in a later presentation)
5
December 2008
FY14 Performance and Results
• Dynamic solvers have been implemented in GridLAB-D that allow for full unbalanced simulations at the distribution level.
• Multi-Time Scale: Seamless integration between quasi-steady state and dynamic simulation time-scales.
• Multiple Devices: Models for generators, governors, and regulators have been developed.
• Speed: Significantly faster than full electromagnetic simulations.
• A completed dynamic model of the WSU and Avista feeders has been completed. (>2,200 nodes)
• Size:
• Validation-Generator: Validation of the generator model data is currently being conducted.
• Validation-System: A validation method for the system model is being developed. (This will be discussed in a later presentation)
• By the end of FY14, it will be possible to examine the dynamic impact of transformer and line in-rush currents; which is essential for examining the use of micro grids as a resiliency resource. This will allow
6
December 2008
Future Plans FY15
• Verify strategies for using microgrids as a community resource
• Operations: Using the existing dynamic model, the effects of transformer in-rush and line charging will be examined with respect to generator sizing and voltage deviations.
• Protection: Impacts on the existing protection scheme will be examined.
• Microgrids as a black start resource
• Using the existing dynamic model, the effects of transformer in-rush and line charging will be examined to determine the reach of this resource.
• Evaluate the ability of microgrids to support units that are not traditionally black start.
• WSU will continue work on restoration/recovery schemes to enhance resiliency.
7
December 2008
Microgrids to Enhance Fast Recovery of Distribution Systems
(Washington State University)
• Resiliency: “..ability to prepare for and adapt to changing conditions and withstand and recover rapidly from disruptions..”*
• For distribution systems, resiliency means the ability to withstand major disturbances. Fast recovery is essential for a resilient system.
• Generation resources and control capabilities of microgrids enhance fast recovery of distribution systems.
• To demonstrate feasibility for microgrids to enhance distribution systems by serving critical load and strengthening fast recovery capability following a major outage.
• Challenge: designing a practical and efficient strategy to make full use of power sources within microgrids to support distribution system restoration.
8
* Office of the Press Secretary of the White House, Presidential Policy Directive – Critical Infrastructure Security and Resilience [Online]. Available:
http://www.whitehouse.gov/the-press-office/2013/02/12/presidential-policy-directive-critical-infrastructure-security-and-resil
December 2008
Main Results So Far
• Graph-theoretic distribution system restoration (DSR) incorporating microgrids - “spanning tree search” algorithm. Paper accepted by IEEE Trans. Power Systems.*
• Restoration software tool developed based on MATLAB and GridLAB-D.
• Pullman-WSU distribution system has been simulated and analyzed.
• A transient model for WSU Microgrid in Simulink.
• Reliability analysis to evaluate contribution of WSU microgrid.
• Restoration scheme under major outages.
• Algorithm designed for optimal planning to increase remote-controlled switches in a distribution system to achieve maximal restoration ability.
9
* J. Li, X.-Y. Ma, C.-C. Liu, and K. P. Schneider, "Distribution system restoration with microgrids using spanning tree search," IEEE Trans. Power Syst.,
to be published.
December 2008 10
DSR Algorithm Based on Spanning Tree Search
• Spanning tree search proposed to identify a post-outage distribution system topology that will achieve a minimum number of switching operations and take full advantage of the resources of microgrids.
December 2008 11
Restoration with/without Microgrids
41 64
74
62 44 70
71
61
77
48 58
75 76
45 63
57
72
69
46 47
60
65
5568
59
79
42 54
78
56 49 50 66
5143
52
73
67
53
80
1 24
34
22 4 30
31
21
37
8 18
35 36
5 23
17
32
29
6 7
20
25
1528
19
39
2 14
38
169
10 26
11 3
12
33
27
13
40
121 144
154
142 124 150
151
141
157
128 138
155 156
125143
137
152
149
126 127
140
145
135148
139
159
122 134
158
136 129 130146
131 123
132
153
147
133
160
81 104
114
102 84 110
111
101
117
88 98
115 116
85 103
97
112
109
86 87
100
105
95108
99
119
82 94
118
96 89 90 106
9183
92
113
107
93
120
T1T2
T6T3
T4
T5
T7
161
Sub-
Transmission
Node
F-a
F-b
F-c
F-d
M Microgrid 1
M Microgrid 2
M
Microgrid 3
Microgrid 4 M
Scenario Fault Location Switching Operations without Microgrids Switching Operations with Microgrids
1 6-7 Open: 49-50, 90-92
Close: 78-9, 53-96, 136-120 Close: 39-Microgrid1
2 127-139
Open: 46-47, 96-89
Close: 136-120, 53-96, 45-90
Partial Restoration, 315.04 kVA load
should be shed at F-b
Open:50-43, 90-92
Close: 45-90, 73-Microgrid2,
136-120
1. Microgrids reduce the
number of switching
operations during
restoration (Scenario 1) ;
2. Using the capability of
microgrids to pick up
more interrupted load
(Scenario 2).
December 2008 12
Distribution Restoration Software
• A restoration program is developed based on MATLAB and GridLAB-D
MATLAB
Modeling distribution
systems with microgrids
as a connected graph
Searching for
possible spanning tree
topologies
GridLAB-D
Three Phase
Unbalanced Power
Flow
Monitoring voltages
and currents of
distribution systems
with microgrids
CALL
RETURN
RESULTS
December 2008 13
Dynamic Simulation of WSU Microgrid
Pullman-WSU
Distribution System
December 2008 14
Restoration Actions for Pullman-WSU System Under a Major Outage
• A restoration scheme is found by spanning tree search to restore critical loads in a major outage in Pullman.
• Restoration scheme validated by GridLAB-D power flow step by step.
• Dynamic constraints will be considered and simulated with GridLAB-D in FY14.
14
13
11
10
9
SPU121
SPU122
SPU123
SPU124
SPU125
49
48
52
51
50
34 37
29
35
41
30
32
39
21 2322 24 27
43 15 16
171840
38
31
36 42
20 19
46 45 33 44 47 28 25 26
WSU Microgrid
Hospital
City Hall,
Courthouse
& Police
Station
G3
G1
G2
1.1 MW
1.1 MW
2.1 MW
G3 17 19 20 34 37 41 32 39 42 36 38 40
City Hall, Courthouse, Police Station Hospital
December 2008 15
Enhancing Resiliency by Further Automation
• Distribution automation enhances resiliency of distribution systems through remote monitoring and control techniques.
• Upgrading existing manual switches to remote-controlled switches (RCSs) enables faster response to disturbances.
• A greedy rule-based algorithm is proposed to determine the optimal number and locations of RCSs for a distribution system.
• Maximum restoration ability is achieved
• Minimum number of RCSs is determined
F1 F2
F3
Z1 Z2 Z3 Z7 Z6 Z5
Z8Z9Z4
FB1 FB2
FB3
S1 S2 S7 S5 S4
S6S8
S3 S9
Feeder Breaker Normally Closed Sectionalizing Switch
Load Zone Normally Open Tie Switch
• Load Group and Switch Group • Weighted Set Cover • Greedy Strategy
December 2008 16
Assuming
• Permanent (short circuit) failure rate (𝜆P) is 0.02
• Mean time to repair (MTTR) is 4 hours
• Mean time to switch (MTTS) is 90 minutes
Reliability Analysis for Pullman-WSU Case
Three restoration strategies are considered
• Strategy 1 (S1): Restoration after repair.
• Strategy 2 (S2): Restoration after fault isolation. Spanning tree search algorithm is
used and DERs in WSU Microgrid are considered
• Strategy 3 (S3): Restoration with further automation. Three manual switches are
upgraded to remote-controlled switches.
Reliability Indices (Not included if outage duration < 5 min)
• SAIFI (/year): 0.11197 (S1) 0.11197 (S2) 0.10434 (S3) 6.81%
• SAIDI (minute/year): 36.949 (S1) 20.533 (S2) 44.4% 17.774 (S3) 13.4%
December 2008 17
Plan for FY15
• Previous work shows the feasibility for microgrids to enhance resiliency of distribution systems. For practical applications, more research need to be conducted.
• 1. Coordinated Restoration Strategies for Integrated Microgrid-Distribution Systems
• Coordination between DMS and Microgrid EMS
• Consider benefits for both distribution systems and microgrids
• 2. Develop a system restoration algorithm that incorporates microgrids as a source of black start capability.
• Provide cranking power to other power plants that require external help to restart.
• System stability and operational constraints must be evaluated before the sequence of actions is implemented.
December 2008
Questions
Kevin Schneider, Ph.D., P.E.
Staff Research Engineer
Pacific Northwest National Laboratory
1100 Dexter Ave N. Suite 400
Seattle, WA 98109 USA
Tel: 206-528-3351
18
Chen-Ching Liu, Ph.D.
Boeing Distinguished Professor of Electrical Engineering
Washington State University
School of Electrical Engineering and Computer Science
PO Box 642752
Pullman, WA 99164 USA
Tel: 509-335-1150
