View
9
Download
0
Category
Preview:
Citation preview
THE ROLE OF SYNCHROPHASORS IN ENSURING RELIABLE POWER SYSTEM CONTROL
Marija Ilic
Professor, Carnegie Mellon University
milic@ece.cmu.edu
Thorp-Phadke Symposium May 2013
Potential of Measurements, Communications and Control
2
PMU Control
Constrained Line
Line-to-Ground Clearance
Transfer Capacity in Real Time
DLR
Combined state estimation and optimization problem
3
Power System Control
State Estimation
Measurements (2 sec) + PMU (millisecond)
Power Flow Analysis
Optimal Power Flow
Yang Weng
• System Load Curve
0 5 10 15 20 25120
140
160
180
200
220Every 10 min Real Time Load of NYISO in Jan 23, 2010
Time (hours)
Real P
ow
er
Load P
(p.u
.)
0 5 10
180
200
10 min Real Time Load of NYISO in Jan 23, 2010
Time (min)Re
al P
ow
er
Lo
ad
P(p
.u.)
Forecasted Load
Actual Load with Disturbance
Lower Bound
Upper Bound
4
From preventive to corrective management of future electric energy systems
5
Predictable load and the disturbance
0 20 40 60 80 100 120 140 160 180
20.3
20.4
20.5
20.6
20.7
20.8
20.9
21
21.1
Time (mins)
Lo
ad
Re
active
Po
we
r E
vo
lutio
n (
p.u
.)
NYISO August 2006 Load Data for 3 Hours, Power Factor = 0.8
0 20 40 60 80 100 120 140 160 18015
16
17
18
19
20
21
22
23
24
Time (mins)
Lo
ad
Re
active
Po
we
r E
vo
lutio
n (
p.u
.)
NYISO August 2006 Load Reactive Power Data for 3 Hours, Measurement Frequency = 0.50 Hz,Power Factor = 0.8
Pre-planed Load Value
Real Load Evolution, with 0.5Hz Sampling
5
PMUs-enabled grid for efficient and reliable scheduling to balance predictable load
• PMUs and SCADA help more accurate state estimate of line flows, voltages and real/reactive power demand
• AC OPF utilizes accurate system inputs and computes settings for controllable grid, generation and demand equipment to help manage the system reliably and efficiently
• Adjustments done every 15 minutes
• Model-predictive generation and demand dispatch to manage ramp rates
Today static dispatch for scheduling
5
Model-predictive scheduling with wind generation---slow time scale
• 20% / 50% penetration to the system
6
Le Xie
9
Conventional
cost over 1 year *
Proposed
cost over the
year
Difference Relative Saving
$ 129.74 Million $ 119.62 Million $ 10.12
Million
7.8%
*: load data from New York Independent System Operator, available online at http://www.nyiso.com/public/market_data/load_data.jsp
0 50 100 150 200 250 3000
50
100
150
Coal Unit 2 (Expensive) Generation
Time Steps (10 minutes interval)
MW
50 60 70 80 90 1000
50
100
150
Coal Unit 2 Generation: Zoomed In
Time Steps (10 minutes interval)
MW
Conventional Dispatch
Centralized Predictive Dispatch
Distributed Predictive Dispatch
Conventional Dispatch
Centralized Predictive Dispatch
Distributed Predictive Dispatch
BOTH EFFICIENCY AND RELIABILITY MET
Model-predictive dispatch with price-responsive demand
8
• Elastic demand that responds to time-varying prices
J.Y. Joo kWh
$
9
Model-predictive dispatch with EVs
10
• Interchange supply / demand mode by time-varying prices
NiklasRotering
Optimal Control of Plug-in-Electric Vehicles: Fast vs. Smart
14
Large-Scale Nonlinear Grid Optimization for Corrective Actions
Imports can be increased by the following:
Optimal generator voltages
Optimal settings of grid equipment (CBs, OLTCs, PARs, DC lines, SVCs)
Studies have shown 20-25% economic efficiency by implementing corrective (not preventive!) actions
Optimal selection of new equipment (type, size, location)
..Remembering the summer of 1983 when Arun and I wrote our reactive power distribution factors currently used in PJM to explain how reactive power affects voltages.. US National Grid Studies still do not go beyond real power distribution factors.
16
On-line resource management can prevent blackouts….
17
PMU-Based Robust Control –fast time scale (automated)
Zhijian Liu
P
P
• Automated Voltage Control (AVC) and Automated Flow Control (AFC) – Design Best Locations
of PMUs – Design Feedback
Control Gains
P
P
18
Building on the long-ago joint work with Jim
0 10 20 30 40 50 60 70 80 90 1000
0.05
0.1
0.15
0.2
0.25
Time (sec)
Syste
m W
ors
t V
oltage D
evia
tion (
p.u
.)
Automatic Voltage Control for ONE Pilot Point Control Case
No Control
One Pilot Point Control
5% Reliability Criteria
Pilot Point: Bus 76663
• Robust AVC Illustration in NPCC System
19
All load buses are
Monitored
0 10 20 30 40 50 60 70 80 90 1000
0.05
0.1
0.15
0.2
0.25
Time (sec)
Syste
m W
ors
t V
oltage D
evia
tion (
p.u
.)
Automatic Voltage Control for Unlimited Information Control Case
No Control
Full Information Control
5% Reliability Criteria
20
0 20 40 60 80 1000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Time (sec)
Syste
m W
ors
t F
low
Devia
tion (
p.u
.)
Automatic Flow Control for ONE Pilot Point Control Case
No Control
One Pilot Point Control
5% Reliability Criteria
Pilot Point: Bus 75403
AVC for the NPCC with PMUs
21
Simulations to show the worst voltage deviations
in response to the reactive power load
fluctuations (3 hours)
0 20 40 60 80 100 120 140 160 1800
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Time (min)
Syste
m W
ors
t V
olta
ge
De
via
tio
n (
p.u
.)
1 Pilot Point Secondary Voltage Control with Measurement Frequency = 0.50 Hz,Power Factor = 0.8
No Control
One Pilot Point per Area
5% Criteria
0 20 40 60 80 100 120 140 160 1800
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Time (min)
Syste
m W
ors
t V
olta
ge
De
via
tio
n (
p.u
.)
2 Pilot Points Secondary Voltage Control with Measurement Frequency = 0.50 Hz,Power Factor = 0.8
No Control
Two Pilot Point per Area
5% Criteria
2 Pilot Points Control Performs Better Than 1 Pilot Point!
21
AFC Using PMUs- NPCC System
0 5 10 15 20 25 30 350
0.02
0.04
0.06
0.08
0.1
0.12
Location of Disturbance
Lin
es R
ea
l P
ow
er
Flo
w D
evia
tio
n (
p.u
.)
System Worst Line Real Power Flow Deviation Under Disturbance From Different Buses
No Control
One Pilot Point per Area
0 20 40 60 80 100 1200
0.02
0.04
0.06
0.08
0.1
0.12
Line NumberL
ine
s R
ea
l P
ow
er
Flo
w D
evia
tio
n (
p.u
.)
Worst Real Power Flow Deviation of Each Line Under Disturbance From Different Buses
No Control
One Pilot Point per Area
Control real power disturbance
….Versions of AVC implemented in EdF Italy, China.. It may be time to consider by the US utilities
Liu and Ilic, “Toward PMU-Based Robust Automatic Voltage Control (AVC) and Automatic Flow Control (AFC),” IEEE PES, 2008
23
Pushing the limits to what is doable –transient stabilization ( Selkrik fault with conventional controller)
24
Voltage response with conventional controllers-base case Selkrik fault
Concepts proposed to manage fast phenomena using
non-interacting control; no need for fast communications (Chapman,J; Allen, E)
25
Bus voltages with new controllers
26
Rotor angle response with local nonlinear controllers--an early example of flat control design
Issues with standards for preventing SSR-related safety problems
27
Nonlinear control for storage devices (FACTS,flywheels)
[1] The test system: J. W. Chapman, “Power System Control for Large Disturbance Stability: Security, Robustness and Transient Energy”, Ph.D. Thesis: Massachusetts Institute of
Technology, 1996.
[2] Linear controller: L. Angquist, C. Gama, “Damping Algorithm Based on Phasor Estimation”, IEEE Power Engineering Society Winter Meeting, 2001
[3] Nonlinear controller: M. Ghandhari, G. Andersson, I. Hiskens, “Control Lyapunov Function for Controllable Series Devices”, IEEE Transactions on Power Systems, 2001, vol. 16, no. 4,
pp. 689-694
Linear PI power controller[2]
Nonlinear Lyapunov controller[3]
No controller on TCSC
Use of interaction variables in strongly coupled systems
Interaction variable choice 1:
Interaction variable choice 2:
Must proceed carefully… • The very real danger of new complexity.
• Technical problems at various time scales lend themselves to the fundamentally different specifications for on-line data
• No longer possible to separate measurements, communications and control specifications
• Major open question: WHAT CAN BE DONE IN A DISTRIBUTED WAY AND WHAT MUST HAVE FAST COMMUNICATIONS –Jim—I think this is doable when the grid has localized response…
The persistent challenge: SE to support on-line scheduling implementation (Yang Wang)
Current Power System State Estimation Problems
Nonlinearity Non-convexity
Historical Data are not really used
New devices (i.e. PMU) placement problem
Convexification Semi-definite Programming
Graph-based distributed SDP
SE
Computational Burden
Non-parametric Static state Estimation
Parametric Dynamic state
Estimation
Information Theory based algorithm for
State Estimation
Parallel Computing Algorithm
Load serving entities (LSEs)
Backbone Power Grid
and its
Local Networks (LSEs)
LSE LSE
LSE
LSE
Information flow: MISO
Local Distribution Network (Radio Network)
Multilayer Information for State Estimation
PQ Diesel PQ Wind PQ PQ
Distributed SE
Computation
Physical Layer Online Diagram Information Layer Diagram
LSE
State information exchange
State information
Exchange on the
boundary nodes
Local State Estimation (LSE)
Backbone
Distributed SE
Computation
LSE LSE
Local serving entities (LSEs)
LSE
LSE LSE
LSE
LSE
LSE
LSE
LSE
LSE
LSE LSE
LSE
LSE
LSE
LSE
LSE
LSE
Ideal Placement of PMUs
14 bus example graphical representation
Qiao Li, Tao Cui, Yang Weng, Rohit Negi, Franz Franchetti and Marija D. Ilic, “An information theoretic approach to PMU placement in electric power systems,
IEEE Transactions on Smart Grid, Special Issue on Computational Intelligence Applications in Smart Grids. (Accepted, to appear) 2013
PMU Information Gain Index
Qiao Li, Tao Cui, Yang Weng, Rohit Negi, Franz Franchetti and Marija D. Ilic, “An information theoretic approach to PMU placement in electric power systems,
IEEE Transactions on Smart Grid, Special Issue on Computational Intelligence Applications in Smart Grids. (Accepted, to appear) 2013
Looking ahead- Framework for integrating combination of technologies at value
• Value is a system-dependent concept (time over which decision is made; spatial; contextual)
• Cannot apply capacity-based thinking; cannot apply short-run marginal cost thinking
• Reconciling economies of scope and economies of scale
• Value of flexibility (JIT,JIP, JIC)
• Hardware, information, decision-making software; distributed, coordinated –all have their place and value
35
Recommended