Embed Size (px)
Citation preview
1 1
Interconnected- Highly Utilized Grid WorkshopSeptember 29-30, 2016
Interconnected- Highly Utilized Grid WorkshopSeptember 29-30, 2016
Planning and Operational Challenges in Interconnected DER-based Grids
Planning and Operational Challenges in Interconnected DER-based Grids
Santiago GrijalvaGeorgia Institute of Technology
Santiago GrijalvaGeorgia Institute of Technology
2 2
Interconnected GridInterconnected Grid
• Millions of spatially distributed, variable DERs are being connected to the grid.
• Operation needs much better coordination across all subsystems: ISO, utilities, microgrids, buildings, homes.
• Grid is cyber-controlled and more Integrated with other systems (gas, transportation, etc.)
3 3
OverviewOverview
1. Transmission– Spatial Distribution– Utilization, Contingencies, Metrics– Scalable Decentralized Optimization
2. Distribution– Multi-layer Cyber-Physical System– Distribution System Operator (DSO) Simulator– Distribution PV Hosting Capacity
4 4
Transmission UtilizationTransmission Utilization
• Line Capacity Utilization– Eastern Interconnection, Sumer Peak 2016, Normal Operation
4
70,00060,00050,00040,00030,00020,00010,000
% o
f MVA
Lim
it (M
ax)
90
80
70
60
50
40
30
20
10
0
Line Number
5 5
UtilizationUtilization
5
% of MVA Limit (Max)
70,00060,00050,00040,00030,00020,00010,000
% o
f MVA
Lim
it (M
ax)
90
80
70
60
50
40
30
20
10
0
% of MVA Limit (Max)Effect of Variability
Desired Higher Utilization
Line Number .
Today
6 6
Weak Elements/Severe ContingenciesWeak Elements/Severe Contingencies
• Eastern Interconnection, N-1 Contingencies (80k+)– Weak Elements are overloaded under many contingencies– Sever Contingencies overload many elements.
• Insecurity Metric: Aggregate MW Contingency Overload. • Has been increasing year after year.
61,000800600400200
Aggr
MVA
Ove
rload
5,0004,5004,0003,5003,0002,5002,0001,5001,000
5000 1,00900800700600500400300200100
Aggr
MVA
Ove
rload 6,000
5,000
4,000
3,000
2,000
1,000
0
1000 Weakest Elements 1000 Most Severe Contingencies
7 7
Transmission PlanningTransmission Planning
• Currently resource and transmission planning are not well integrated. Difficulty handling:– Full AC models, voltage constraints, and limits due to
voltage and transient stability.– Emerging technologies: routers, switching, DC.– Large number of scenarios.
• Challenges:– Integrated gen + demand + DER, + transmission planning. – Must handle order of magnitude increase in control variables.– Intertemporal behavior, integer variables, non-convexities.– New controls, devices, operational structures. – Interactions with other infrastructures.
8 8
Scalable Decentralized OptimizationScalable Decentralized Optimization
• Example: Unit Commitment– Large-Scale ISO realistic data– Full UC model: reserve, CSU, CSD, ….17k+ constraints.
– Problem solved orders of magnitude faster.
3
1
2
4
5
6
7
8
9
10
11
1213
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
3435
36
3738
3940
41
42
43 44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61 62
63
64
6566
67
6869
70
71
72
73
7475
76
77
7879
80
81
82
83
84
85
86
87
8
89
90
91
92
93
94
95
96
97
98
99
100
9 9
OverviewOverview
1. Transmission– Spatial Distribution– Utilization, Contingencies, Metrics– Scalable Decentralized Optimization
2. Distribution– Multi-layer Cyber-Physical System– Distribution System Operator (DSO) Simulator– Distribution PV Hosting Capacity
0 10
Emerging Complexity in DistributionEmerging Complexity in Distribution
• DER integration: – PV, Wind, EV, Demand Response, CHP. – Microgrids, Virtual Power Plants
• Active customers/prosumers• Microgrids• Smart distribution devices/smart appliances• AMI creates a wealth of data that can be exploited• Evolving marketplaces• Utility business models
1 11
Layered Energy Cyber-Physical SystemLayered Energy Cyber-Physical System
12 12
DSO Simulation Studio (ARPA-E Open 2015)DSO Simulation Studio (ARPA-E Open 2015)
12
• A Multi-Layer SimulatorDSO Simulation Studio
Devices Controllers: Intelligent ControllersPower Electronics, Protections, Sensors
Market
System
Time Series DER Quasi-Steady State Solver
Power Grid: Wires, Transformers, Capacitors, etc. DER: Flexible Load, Solar PV, Storage, CHP, Wind, EV
Locational-Temporal Pricing Module
Security Modeling
Forecasting Multi-Agent Prosumer Model
DSO Design and Policy Engine
DSO Rules
Regulators
Market Participants
UtilityEngineers
Developers
Decentralized Energy Scheduling
DER Services
DER Services Valuation
3 13
DSO Simulator Expected FeaturesDSO Simulator Expected Features
13
• Decentralized energy scheduling of DER-rich systems of arbitrary size.
• Explicit modeling of energy services transacted in the DSO.
• Locational and time-vector pricing of P/Q, ancillary, and security services.
• 3D Interactive Visualization• Analytics and valuation of DER
services, DSO rules, and business models.
• Simulation of multi-scale interactions of DSO with up-stream ISO, same level DSOs, and downstream (microgrid, building, and home) prosumer subsystems.
3D Visualization Concept
4 14
PV Hosting CapacityPV Hosting Capacity
• How much PV can we put on a distribution circuit?• Constraints:
– Under and over-voltages– Thermal limits– Protections– Back-feeding
• Complexity– Variability (seasonal, daily, second to second). – Need quasi-static time series (QSTS) Simulation
• One-year, at 1 sec granularity (31M solutions). – Local and distributed PV– Effect of controllers and smart inverters
5 15
PV Impact and Hosting CapacityPV Impact and Hosting Capacity
6 16
QuantizationQuantization
• Various methods for computation improvement: circuit reduction, parallelization.
• Recent improvements: quantization– Accurate results, ~ 5% of solutions compared to brute-force.
7 17
Towards DER Hosting CapacityTowards DER Hosting Capacity
• Challenges:– System impact and hosting capacity determination for solutions
with all types of DERs, (e.g. solar + storage). – Hosting capacity for circuits that already host complex DERs.
