2014 pv distribution system modeling workshop: high penetration pv control comparisons and...
DESCRIPTION
2014 PV Distribution System Modeling Workshop: High Penetration PV Control Comparisons and Model-Centric Smart Grid CBA: Robert Broadwater, EDDTRANSCRIPT
High Penetration PV Control Comparisons and Model-
Centric Smart Grid CBA
May 7, 2014
1
Robert Broadwater [email protected] www.edd-us.com
© Copyright Electrical Distribution Design, Inc. 2014
© Copyright Electrical Distribution Design, Inc. 2014 Confidential 2
Model-Centric Smart Grid CBA
Model-Centric Smart Grid
Performance Analysis +
Economic Analysis +
Lab Testing +
Field Validation =
Model-Centric Smart Grid
Reliability, Efficiency, Capacity, Protection, Controllability
© Copyright Electrical Distribution Design, Inc. 2014
Incremental Grid Modernization CBA
Phase-Balance (no capacitors)
Phase Balance for Time
Varying Load
Capacitor Design (capacitors on local control)
Cap Design for Time
Varying Load
Auto Reconfiguration,
Monte Carlo
CBA1
CBA2
CBA3
CBA4
Base System (not optimized,
some capacitors)
“Dependency Ordering” of Investments
Coordinated
Control
Coordinated Control
Distribution Automation
(blue sky days) (storm conditions) Reliability
Efficiency Energy
Time Series Analysis
© Copyright Electrical Distribution Design, Inc. 2014
Time Series Analysis Example
© Copyright Electrical Distribution Design, Inc. 2014
% Errors between Load Factor and Time Series Analysis
-20
-10
0
10
20
30
40
50
60
70
80
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
% E
rro
rs B
etw
ee
n T
ime
Se
rie
s an
d
Load
Fac
tor
Cal
cula
tio
ns
Feeders
© Copyright Electrical Distribution Design, Inc. 2014
Present Value Savings for 10 Years
Case
Cost (Inc/Total)
($000)
Savings Type
Case Savings ($000)
$ Saved / $ Invested (Inc/Total)
Efficiency (Inc/Total)
($000)
Energy (Inc/Total)
($000)
Capital ($000)
Operation
($000) CI
($000)
CBA1 163/163 94/94 29/29 NA NA NA 123 0.75/0.75
CBA2 564/727 227/321 2,234/2,263 NA NA NA 2461 4.36/3.55
CBA3 68/795 88/409 2,064/4,328 NA NA NA 2,132 31.65/5.74
CBA4 1,953/2,748 NA NA 7,014 7,646 9,566 24,226 12.4/10.54
$1.4 ~ Residential $230 ~ Commercial $650 ~ Industrial
Societal Benefits Estimated CO2 reduction = 76,330 tons
© Copyright Electrical Distribution Design, Inc. 2014
Validation of Phase Balancing
Crew Phase Balancing Operation Amps
Days 1 2 3 4 5 6
© Copyright Electrical Distribution Design, Inc. 2014
© Copyright Electrical Distribution Design, Inc. 2014 Confidential 9
High Penetration PV Control Comparisons
Control Approach
• Configurable, Hierarchical, Model-based, Scheduling Control = Forecast – Monitor – Schedule - Adjust
• Collects circuit-wide information and uses model to calculate set-points for control devices
• Sends control set-points to both utility control devices and PV controllers
• Strives to maintain the voltage profile that exists without PV generation while minimizing circuit losses and reducing the motion of utility control devices
© Copyright Electrical Distribution Design, Inc. 2014
CHMSC Control Architecture
CHMSC
PV Controller
PV Controller
PV Generator
PV Generator
...
...
Local PV Controller
PV Generator
Base Controller
Base Controller
Voltage Regulator
Switched Capacitor
...
...
Local Base Controller
Voltage Regulator, Switched Capacitor
Controllable PV Uncontrollable PV Controllable Automated DeviceUncontrollable
Automated Device
© Copyright Electrical Distribution Design, Inc. 2014
Local Voltage Controller
Grid
+ GainVset-point
-
V
Q Limit
++
Q
Q
Vset-point
QMax
QMax
Reactive Power
Voltage
Slope = GainSupplyReactivePower
ConsumeReactivePower
(a) voltage-reactive power control block diagram (b) voltage-reactive power characteristics
© Copyright Electrical Distribution Design, Inc. 2014
CHMSC Algorithm
Updates periodically, where every five minutes is currently used
Updates set-point schedules for base and/or PV control only if schedules change significantly
If a communication failure occurs, the local controllers continue to work against the previously provided schedule as long as local constraint violations do not occur
Load Forecast
Base Controller Schedules
PV Controller Schedules
Set-point Scheduling
Solar Forecast
© Copyright Electrical Distribution Design, Inc. 2014
Circuit Model
© Copyright Electrical Distribution Design, Inc. 2014
123% PV penetration
1 Second PV Generation Data
© Copyright Electrical Distribution Design, Inc. 2014
Controls Evaluated
CHMSC: Feeder losses and utility controller motion are minimized and voltage set-points are used for PV generators
CHMSC – 116V: Average customer voltage set-point at 116V
CHMSC – 124V: Average customer voltage set-point at 124V
CHMSC (PF set): Power factor set-points used for PV generators
CHMSC – 116V (PF set): Average customer voltage set-point at 116V with power factor set-points provided to PV generators
CHMSC – 124V (PF set): Average customer voltage set-point at 124V with power factor set-points provided to PV generators
Local control only (116V): 116V set-point used by all PV generators
Local control only (124V): 124V set-point used by all of PV generators
© Copyright Electrical Distribution Design, Inc. 2014 Confidential 16
CHMSC Results: Sub Q Flow
© Copyright Electrical Distribution Design, Inc. 2014
CHMSC with 116V/124V SPs: Sub Q Flow
© Copyright Electrical Distribution Design, Inc. 2014
CHMSC with 116V/124V SPs
© Copyright Electrical Distribution Design, Inc. 2014
Sub Q Flow Comparison
© Copyright Electrical Distribution Design, Inc. 2014
Average PF Results Comparison
© Copyright Electrical Distribution Design, Inc. 2014
Average Q Generation Comparison
© Copyright Electrical Distribution Design, Inc. 2014
Circuit Loss Comparison
Real power loss (kW) comparison between local control and CHMSC
© Copyright Electrical Distribution Design, Inc. 2014
Circuit Loss Comparison
Local Control CHMSC Improvement
Real Power Loss (kW-hr) 198.98 kW-hr 123.25 kW-hr 38.06%
Reactive Power Loss (kVAR-hr) 240.69 kVar-hr 130.38 kVar-hr 45.83%
© Copyright Electrical Distribution Design, Inc. 2014
© Copyright Electrical Distribution Design, Inc. 2014 Confidential 25
Comparing CHMSC with Local Control with Increasing PV Penetration – 2 Hour Study
Control Device Motion Comparison
Reduction in control device movement with CHMSC with increasing PV penetration
© Copyright Electrical Distribution Design, Inc. 2014
Voltage Violation Comparison
Number of voltage violations during 2 hour period with increasing PV penetration
© Copyright Electrical Distribution Design, Inc. 2014
Conclusions
CHMSC requires less reactive power flow at substation
CHMSC provides higher power factor at PV
• CHMSC has less reactive power generation at PV generator
CHMSC results in lower circuit loss
CHMSC results in fewer utility device controller steps
CHMSC results in fewer voltage and overload violations
© Copyright Electrical Distribution Design, Inc. 2014