Advanced Distribution Management Systems and … 2, 2017 1 Advanced Distribution Management Systems and Distribution Automation Distribution Systems and Planning Training for New England

October 2, 2017 1October 2, 2017 1

Advanced Distribution Management

Systems and Distribution Automation

Distribution Systems and Planning Training for New England Conference of Public Utility Commissioners, Sept. 27-29, 2017

Murali Baggu (NREL) and Kevin Schneider (PNNL)


DMS: Monitoring and

control of

Distribution level

assets for better

reliability of the

system

Technological Advancements in Management Systems

DERMS: Monitoring

dispatch and control

of DERs

Microgrid Control and

Management: Control,

Schedule and Dispatch

of assets with in the

Microgrid for system

resiliency


• Seamless Integration

among multiple systems

• Sharing common

platform

• IT- OT convergence

• Cybersecurity

• Integration of different

data systems to

enhance decision

making

Enterprise Systems Integration and Interoperability


Short term Planning and Operations

Convergence of systems operations

and short term planning

• Generate possible control

scenarios for optimal system

operation

• Spatial and temporal visualization

of multiple scenarios and possible

mitigation measures

• Cost and Benefit Analysis


• DERMS provide situational

awareness, control/dispatch and

monitoring of DERs in the

distribution system:

– PV with and without smart inverters

– Energy storage

– Electric vehicles

Distributed Energy Resource Management System


• Understand the true value energy

storage can generate when they are

highly utilized by stacking multiple

grid services.

Multi Service Dispatch of Energy Storage


• Tools to model large scale distribution

systems for evaluating ADMS applications

• Integrate distribution system hardware for

PHIL experimentation

• Develop advanced visualization capability

ADMS Testbed

OMS

DERMS

A

Actual ADMS Deployment


What Makes Grid Smart?

• New intelligent devices

• New data analytics and predictive analytics

techniques

• Foundational optimization and control theory

methods

• New Business Models (eg., “PaaS” (Platform

as a Service) or “SaaS” (Software as a Service))

• Tools for modeling, simulation and

visualization


A Decision Support System to assist the control room and field operating

personnel with the monitoring and control of the electric distribution

system in an optimal manner while improving safety and asset protection

Distribution Management System

1) Network Connectivity Analysis (NCA)2) State Estimation(SE)3) Load Flow Applications (LFA)4) Volt/ VAR Optimization (VVO)5) Load Shed Application(LSA)6) Fault Location, Isolation & System

Restoration (FLISR)7) Load Balancing via Feeder Reconfiguration

(LBFR)8) Distribution Load Forecasting

*Slide Source: http://www.smartgridnews.com/artman/uploads/1/Distribution_Management_Systems_-_Robert_Uluski.pdfand http://distributionmanagementsystem.blogspot.com/

http://www.smartgridnews.com/artman/uploads/1/Distribution_Management_Systems_-_Robert_Uluski.pdf

http://distributionmanagementsystem.blogspot.com/

October 2, 2017 10

Distribution Grid Modernization – Advanced

Distribution Management Systems Applications

October 2, 2017 11October 2, 2017 11

Volt-VAR Optimization

October 2, 2017 12October 2, 2017 12

Volt-VAR Optimization (VVO)

► In contrast to the traditional operational methods,

Volt-VAR Optimization controls multiple devices to

achieve a global optimum.

► Volt-VAR Optimization, and the associated global

optimum(s), exists in many forms.

► The general principle is to control the voltage and

reactive power on a distribution feeder so that load

can be managed.

► One example of VVO:

◼ Voltage optimization involves “flattening” the voltage

profile and lowering.

◼ VAR optimization involves controlling the flow of

reactive power, which has an impact on voltage.

October 2, 2017 13October 2, 2017 13

Volt-VAR Optimization

► Voltage Reduction

◼ Voltage is reduced to the low end of ANSI C84.1, customer voltage never go below 114V.

◼ Rule of thumb: for every 1% reduction in voltage, there is a 0.7% reduction in energy consumption.

◼ The majority of energy savings, >90%, are behind the customer meter.

► VAR Optimization

◼ Capacitors are switched to reduce the reactive power flow on the feeder.

◼ This reduces the series losses of the line, increasing efficiency.

October 2, 2017 14October 2, 2017 14

Automated Reconfiguration

October 2, 2017 15October 2, 2017 15

Switching Devices

► Switches

◼ Can be manually or remotely operated.

◼ Are not designed to interrupt load.

◼ Cannot be part of an automated

reconfiguration scheme.

► Breaker/reclosers

◼ Can be manually or remotely operated.

◼ Are designed to interrupt load and fault

currents.

◼ Can be connected to a SCADA system.

October 2, 2017 16October 2, 2017 16

Fault Location, Isolation, and Service Restoration

► Auto-loop

◼ Recloser type devices operated in pre-defined teams.

◼ Static set points that can only respond to conditions within a predefined range.

◼ Can operate independently or connected to DMS.

► Adaptive Reconfiguration

◼ Recloser type devices operate in flexible configurations.

◼ Can operate independently or connected to DMS.

◼ Complex to operate and fully functional systems are still emerging.

Substation

CB

CB

RCL

RCL

CB

CB

RCL

RCL

RCL

RCL

RCL

RCL

RCL

RCL

October 2, 2017 17October 2, 2017 17

► 15 ADMS applications considered for review

(DR, Short Circuit Analysis, DERMS, SOM,

etc.)

► Four applications that are being evaluated:

◼ Volt-VAr Optimization (VVO)

◼ Fault Location Isolation & Service

Restoration (FLISR)

◼ Online Power Flow (OLPF)/ State

Estimation (DSSE)

◼ Market Participation

ADMS Applications

October 2, 2017 18October 2, 2017 18

► Model Improvement

◼ data needs for feeder models; specs and locations for adding new telemetry points; evaluate

impact on ADMS applications

► Calibrate OLPF/DSSE functions

◼ Compare the states testbed measurements, tune algorithms

► Evaluate performance of hierarchical distributed sensing

◼ Integrating sensing technologies like AMI, OpenFMB, OpenADR, grid-edge smart controls,

distribution PMUs

► Modeling loss of PV

◼ Behavior of behind-the-meter components (PV), net load allocation, integrate forecasting,

customer facility data, load models, etc.

OLPF/DSSE Use Cases

October 2, 2017 19October 2, 2017 19

► Voltage Regulation

◼ legacy voltage control assets, smart inverters,

energy storage, autonomous controllers

► Peak Load Management

◼ CVR for peak load management and

interaction with “aggregators” like DERMS

and DRAS

► Performance evaluation

◼ Multi-objective VVO, different control

architectures

► Interaction with Active Grid Edge Devices

◼ Centralized VVO with grid-edge controllers

Volt-VAr Optimization (VVO) Use Cases

October 2, 2017 20October 2, 2017 20

► High Penetration of DERs

◼ Upstream & downstream DERs; line loading before and after fault; intermittency &

visibility challenges

► Interaction with Microgrids

◼ Impact of temporary fault, black start, need for direct comm

► Very High Loading Conditions

◼ Unnecessary backup feeder trip, Use of load forecasting

► Multiple Simultaneous Faults

◼ Thunderstorms leading to multiple faults, feeder re-tripping & lockouts

► Widespread Outages

◼ Uncertain distribution configurations, comm status and feeder outages

Fault Location Isolation & Service Restoration

(FLISR) Use Cases

October 2, 2017 21October 2, 2017 21

► Maintaining power quality while

providing bulk grid services

► Distribution System Operations

(DSOs) providing market functions

► Estimating available capacity for

bidding in energy markets

Market Participation Use Cases

October 2, 2017 22

ADMS Case Studies

October 2, 2017 23October 2, 2017 23

Case Study 1: Feeder Voltage using Advanced

Inverters and a DMS

Objective:

Understand advanced inverter and distribution management system (DMS) control options for large (1–5 MW) distributed solar photovoltaics (PV) and their impact on distribution system operations for:

Active power only (baseline);

Local autonomous inverter control: power factor (PF) ≠1 and volt/VAR (Q(V)); and

Integrated volt/VAR control (IVVC)

Approaches:

Quasi-steady-state time-series (QSTS)

Statistics-based methods to reduce simulation times

Cost-benefit analysis to compare financial impacts of each control approach.

Energy Systems

Integration

Recloser 1

Recloser 2

Recloser 3

Cap Bank 1

Cap Bank 2

Regulator 1

Regulator 2

Regulator 3

Feeder Head – Breaker/Regulator DMS

Opal-RTReal-time Simulator

Grid Simulator

Grid Simulator

Visualization

PV Inverter

Cap/VR

Palmintier, B., Giraldez, J., Gruchalla, K., Gotseff, P., Nagarajan, A., Harris, T., ... Baggu, M. (2016). Feeder Voltage Regulation With High Penetration PV Using Advanced Inverters and a Distribution Management System: A Duke Energy Case Study (NREL Technical Report No. NREL/TP-5D00-65551). Golden, CO: National Renewable Energy Laboratory.

October 2, 2017 24October 2, 2017 24

The Problem!

October 2, 2017 25October 2, 2017 25

Study System Characteristics

Cap1: A 450-kVAR (150 kVAR per phase) VAR-controlled capacitor with temperature override. Cap2: A three-phase 450-kVAR capacitor (always disconnected unless controlled otherwise by IVVC)

Reg1: A set of three single-phase 167-kVA regulators with a voltage target of 123

Reg2: A set of two single-phase 114-kVA regulators on phase B and phase C with a voltage target of 123 V;

Reg3: A second set of two single-phase 76.2-kVA regulators on phase B and phase C with a voltage target of 124 V;

October 2, 2017 26October 2, 2017 26

PF limit

P generated

P curtailedQ

(k

VA

R)

S (k

VA

)

Q

P

PF = 0 night mode

Operating point

with active power curtailed

Operating point with

oversized inverter

0.97 1.03

Voltage (p.u.)R

eact

ive

Po

wer

(per

cen

t of

kV

A r

ati

ng)

0

0.95 1.051.0

-50%(absorbing)

50%(injecting)

Constant Power Factor Set Point

Volt/VAR Curves

Local Control Modes for the PV Inverter

October 2, 2017 27October 2, 2017 27

Data and Statistical Processing

To capture the high ramp rates associated with the PV plant variability on the feeder and to generate accurate feeder statistics, a complete 1-minute data set (i.e., 525,600 measurements) for 2014 were provided

Statistical smoothing to create native load

October 2, 2017 28October 2, 2017 28

Data Analysis: 1 Year ➔ 40 Days 1 Year

Clear(>35MWh)Or

Cloudy(<10MWh)

Spring Summer

(old) Summer

(new) Fall Winter (old)

Winter (new)

Extreme Heating

Clear weekday

14

9

11 11 6

6 Clear weekend

5 5 7 9

Cloudy weekday

5 4 7

7 12

8 Cloudy weekend

8

Moderate weekday

13

6 14

17 17 6

5

Moderate weekend

4 5

High weekday

16 19 15 15 9

High weekend

13 8 9 6 6 16

Extreme weekday

5 for all

shoulder seasons

6

(Combine

with Spring)

Extreme weekend

Histogram of t

t*

De

nsity

−300 −100 0 100 300

0.0

00

0.0

02

0.0

04

−3 −2 −1 0 1 2 3

−2

00

02

00

Quantiles of Standard Nor mal

t*

SCADAData1Year@1min

RandomSetofDays

RegressionModel

Simula onOf40-days

Regression:Reg/CapOpera onsVoltageChallenges

AnnualizedResults

Representa veDayDuplica onforTime

Series

PVVariability

Assessquality

LoadSeason

40typesofdays

October 2, 2017 29October 2, 2017 29

Simulations Cases

► Baseline

► Local PV Control (PF = 0.95)

► Local PV Control (Volt/VAR)

► Legacy IVVC (Exclude PV)

► IVVC with PV @ PF 0.95

► IVVC (Central PV Control)

-5000

-4500

-4000

-3500

-3000

-2500

-2000

-1500

-1000

-500

0

100 200 300 400 500 600 700P

V g

enera

tio

n (

kW

)

Time (Minutes)

31 August, 2014 (cloudy day)5 July, 2014 (clear day)

October 2, 2017 30October 2, 2017 30

Baseline Results

−4000

−2000

0

2000

4000

06AM 08AM 10AM 12PM 02PM 04PM 06PM 08PM

Saturday, April 5 − Baseline

Fe

ede

r K

W

Fe

ed

er

KV

AR

KV

AR

−500

0

500

1000

1500

2000

2500

Feeder KWFeeder KVAR

Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

KW GenerationKVAR

Solar PV Farm

0

1

0

1

2

3

4Cap1Cap2Total Cap Changes

Capacitor Status

Ca

p S

tatu

s

To

tal C

ap

Ch

ang

es

−15

−10

−5

0

5

10

15

Total Tap ChangesABC Tap Position

0

1

2

3

4

5Substation LTC

Ta

p P

ositio

n

To

tal Tap C

han

ge

s

−15

−10

−5

0

5

10

15

Total Tap ChangesReg1 AReg1 BReg1 CReg2 BReg2 CReg3 BReg3 C

0

5

10

15

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Cha

ng

es

−1.0

−0.5

0.0

0.5

1.0Over voltageUnder voltageTotal Violations

−1.0

−0.5

0.0

0.5

1.0


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

−4000

−2000

0

2000

4000


Friday, August 15 − Baseline

Fe

ede

r K

W

Fe

ed

er

KV

AR

KV

AR

−2000

−1000

0

1000

2000Feeder KWFeeder KVAR

Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000KW GenerationKVAR

Solar PV Farm

0

1

0

2

4

6

8

10

12Cap1Cap2Total Cap Changes

Capacitor Status

Ca

p S

tatu

s

To

tal C

ap

Ch

ang

es

−15

−10

−5

0

5

10

15

Total Tap ChangesABC Tap Position

0

5

10

15Substation LTC

Ta

p P

ositio

n

To

tal Ta

p C

ha

nge

s

−15

−10

−5

0

5

10

15

Total Tap ChangesReg1 AReg1 BReg1 CReg2 BReg2 CReg3 BReg3 C

0

20

40

60

Regulator Status

Ta

p P

ositio

n

Tota

l Ta

p C

ha

ng

es

0

50

100

150Over voltageUnder voltageTotal Violations

0

200

400

600

800

1000


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

October 2, 2017 31October 2, 2017 31

Autonomous Local Control

−4000

−2000

0

2000

4000


Saturday, April 5 − Local PV Control (PF=0.95)

Fee

de

r K

W

Fe

ed

er

KV

AR

KV

AR

−500

0

500

1000

1500

2000

2500

Feeder KWFeeder KVARKW BaselineKVAR BaselineKW Delta from BaselineKVAR Delta from Baseline

Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR

KW GenerationKVARKVAR BaselineDelta from Baseline

Solar PV Farm

0

1

0

1

2

3

4Cap1Cap2Total Cap ChangesCap Changes BaselineDelta from Baseline

Capacitor Status

Ca

p S

tatu

s

To

tal C

ap C

ha

ng

es

−15

−10

−5

0

5

10

15

Total Tap ChangesTotal Tap Changes BaselineDelta from BaselineABC Tap Position

0

1

2

3

4

5Substation LTC

Ta

p P

ositio

n

To

tal Tap C

han

ge

s

−15

−10

−5

0

5

10

15

Total Tap ChangesTotal Tap Changes BaselineDelta from BaselineReg1 AReg1 BReg1 CReg2 BReg2 CReg3 BReg3 C

0

5

10

15

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Chan

ges

−1.0

−0.5

0.0

0.5

1.0Over voltageUnder voltageTotal ViolationsTotal Violations BaselineDelta from Baseline

−1.0

−0.5

0.0

0.5

1.0


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

−4000

−2000

0

2000

4000


Friday, August 15 − Local PV Control (PF=0.95)

Fe

ede

r K

W

Fe

ed

er

KV

AR

KV

AR

−2000

−1000

0

1000

2000Feeder KWFeeder KVARKW BaselineKVAR BaselineKW Delta from BaselineKVAR Delta from Baseline

Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

2

4

6

8

10


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap

Ch

ang

es

−15

−10

−5

0

5

10

15


0

5

10

15Substation LTC

Ta

p P

ositio

n

To

tal Ta

p C

ha

nge

s

−15

−10

−5

0

5

10

15


0

20

40

60

Regulator Status

Ta

p P

ositio

n

Tota

l Ta

p C

ha

ng

es

0

50

100

150Over voltageUnder voltageTotal ViolationsTotal Violations BaselineDelta from Baseline

0

200

400

600

800

1000


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

October 2, 2017 32October 2, 2017 32

−4000

−2000

0

2000

4000


Saturday, April 5 − Local PV Control (Volt−Var)

Fee

de

r K

W

Fe

ed

er

KV

AR

KV

AR

−500

0

500

1000

1500

2000

2500


Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

1

2

3


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap C

ha

ng

es

−15

−10

−5

0

5

10

15


0

1

2

3

4

5Substation LTC

Ta

p P

ositio

n

To

tal Tap C

han

ge

s

−15

−10

−5

0

5

10

15


0

5

10

15

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Chan

ges

−1.0

−0.5

0.0

0.5


−1.0

−0.5

0.0

0.5

1.0


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

−4000

−2000

0

2000

4000


Friday, August 15 − Local PV Control (Volt−Var)

Fee

de

r K

W

Fe

ed

er

KV

AR

KV

AR

−2000

−1000

0

1000


Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

2

4

6

8

10


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap C

ha

ng

es

−15

−10

−5

0

5

10

15


0

5

10

15Substation LTC

Ta

p P

ositio

n

To

tal Tap C

han

ge

s

−15

−10

−5

0

5

10

15


0

20

40

60

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Chan

ges

0

50

100


0

200

400

600

800

1000


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

Local Volt/VAR Control

October 2, 2017 33October 2, 2017 33

−4000

−2000

0

2000

4000


Saturday, April 5 − Legacy IVVC (exclude PV)

Fee

de

r K

W

Fe

ed

er

KV

AR

KV

AR

−500

0

500

1000

1500

2000

2500


Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

1

2

3


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap C

ha

ng

es

−15

−10

−5

0

5

10

15


0

1

2

3

4

5Substation LTC

Ta

p P

ositio

n

To

tal Tap C

han

ge

s

−15

−10

−5

0

5

10

15


0

5

10

15

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Chan

ges

−1.0

−0.5

0.0

0.5


−1.0

−0.5

0.0

0.5

1.0


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

−4000

−2000

0

2000

4000


Friday, August 15 − Legacy IVVC (exclude PV)

Fe

ede

r K

W

Fe

ed

er

KV

AR

KV

AR

−2000

−1000

0

1000


Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

2

4

6

8

10


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap C

ha

ng

es

−15

−10

−5

0

5

10

15


0

5

10

15Substation LTC

Ta

p P

ositio

n

To

tal Ta

p C

ha

nge

s

−15

−10

−5

0

5

10

15


0

20

40

60

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Cha

ng

es

0

50

100


0

200

400

600

800

1000


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

Legacy DMS Integrated Volt/VAR Control

with LTC, VR and Caps Only

October 2, 2017 34October 2, 2017 34

−4000

−2000

0

2000

4000


Saturday, April 5 − IVVC (central PV control)

Fee

de

r K

W

Fe

ed

er

KV

AR

KV

AR

−500

0

500

1000

1500

2000

2500


Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

1

2

3


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap C

ha

ng

es

−15

−10

−5

0

5

10

15


0

1

2

3

4

5Substation LTC

Ta

p P

ositio

n

To

tal Tap C

han

ge

s−15

−10

−5

0

5

10

15


0

5

10

15

Regulator Status

Ta

p P

ositio

n

Tota

l Tap

Chan

ges

−1.0

−0.5

0.0

0.5


−1.0

−0.5

0.0

0.5

1.0


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

−4000

−2000

0

2000

4000


Friday, August 15 − IVVC (central PV control)

Fe

ede

r K

W

Fe

ed

er

KV

AR

KV

AR

−2000

−1000

0

1000


Feeder

0

1000

2000

3000

4000

5000

PV

KW

PV

KV

AR

KV

AR

−2000

−1000

0

1000

2000

PV

KV

AR


Solar PV Farm

0

1

0

2

4

6

8

10


Capacitor Status

Ca

p S

tatu

s

To

tal C

ap

Ch

ang

es

−15

−10

−5

0

5

10

15


0

5

10

15Substation LTC

Ta

p P

ositio

n

To

tal Ta

p C

ha

nge

s

−15

−10

−5

0

5

10

15


0

20

40

60

Regulator Status

Ta

p P

ositio

n

Tota

l Ta

p C

ha

ng

es

0

50

100


0

200

400

600

800

1000


Volta

ge

Vio

lation

s

To

tal V

olta

ge V

iola

tion

s

Integrating Advanced Inverters into IVVC

October 2, 2017 35October 2, 2017 35

Feeder 40-day results of number of

operations of voltage regulation equipment

October 2, 2017 36October 2, 2017 36

Feeder 40-day results of number of load-

voltage violations

October 2, 2017 37October 2, 2017 37

Summary Comparison of Annualized

Scenarios

Scenario

BaselineLocalPVControl

(PF=0.95)

LocalPVControl(Volt/VAR)LegacyIVVC(ExcludePV)

IVVCwithPV(PF=0.95)

IVVC(CentralPVControl)

PVMode LTC

Regulators

Cap

acitors

PV LTC Regulators Capacitors Total Over Under

Default - - - - 5,043 19,160 125 24,328 1.47% 0.00%

PF=0.95 - - - - 5,063 19,943 505 25,511 1.48% 0.00%

Q(V) - - - - 5,087 19,857 541 25,485 1.44% 0.00%

Default Y Y Y - 2,869 2,943 1,863 7,675 0.02% 0.00%

PF=0.95 Y Y Y - 2,498 1,888 1,409 5,795 0.01% 0.00%

IVVCforreactivepower

Y Y Y Y 2,312 2,698 1,151 6,161 0.05% 0.02%

IVVCControl AnnualizedEquipmentOperations VoltageChallenges

October 2, 2017 38October 2, 2017 38

Substation P/Q Plots

October 2, 2017 39October 2, 2017 39

Cost-Benefit Analysis Assumptions

Calculate Commodity

Value Streams

Extrapolate Last Simulated

Year Values

Apply Annual Capital and Non-Capital Unit Costs

Apply Inflation

Apply Discounting

Cross-tabulate (pivot) and

Present Results

Accumulate Discounted Cash Flows

Value Streams: PV energy production, feeder losses, loads, and the frequency of operation of switching equipment with the expected resulting requisite maintenance and replacement expenditures

October 2, 2017 40October 2, 2017 40

-$3.50 M

-$3.00 M

-$2.50 M

-$2.00 M

-$1.50 M

-$1.00 M

-$0.50 M

$0.00 M

2014 2019 2024 2029 2034 2039 2044 2049 2054 2064

Cu

mu

lati

ve

NP

V

Year

Baseline

Local PV Control (PF=0.95)

Local PV Control (Volt/VAR)

Legacy IVVC (Exclude PV)

IVVC with PV @ PF=0.95

IVVC (Central PV Control)

Cumulative NPV’s and Aggregated cumulative

discounted cash flows for equipment replacement

-$60,000

-$50,000

-$40,000

-$30,000

-$20,000

-$10,000

$0

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.95

)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.95

)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Cen

tral

PV

Co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.95

)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Cen

tral

PV

Co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.95

)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Cen

tral

PV

Co

ntr

ol)

2034 2044 2054 2064

Cu

mu

lati

ve N

PV

Year

SubLTC, replace

Reg1 A, replace

Reg1 B, replace

Reg1 C, replace

Reg2 B, replace

Reg2 C, replace

Cap 1, replace switches

October 2, 2017 41October 2, 2017 41

Categorical Cost and Savings

-$200,000

-$100,000

$0

$100,000

$200,000

$300,000

$400,000

$500,000

$600,000

$700,000

$800,000

$900,000B

ase

line

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Cen

tral

PV

Co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Ce

ntr

al P

V C

on

tro

l)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Ce

ntr

al P

V C

on

tro

l)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(Ex

clu

de

PV

)

IVV

C w

ith

PV

@ P

F=0.

95

IVV

C (

Ce

ntr

al P

V C

on

tro

l)

2034 2044 2054 2064

Cat

ego

rica

lly A

ggre

gate

d C

um

ula

tive

Dis

cou

nte

d C

ash

Flo

w

Year

Per Scenario Savings Relative to Baseline (scenario minus baseline)

Equipment Replacement

Losses

Loads

PV Production

-$3.50 M

-$3.00 M

-$2.50 M

-$2.00 M

-$1.50 M

-$1.00 M

-$0.50 M

$0.00 M

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(e

xclu

de

PV

)

IVV

C w

ith

PV

@ P

F=0

.95

IVV

C (

cen

tral

PV

co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(e

xclu

de

PV

)

IVV

C w

ith

PV

@ P

F=0

.95

IVV

C (

cen

tral

PV

co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(e

xclu

de

PV

)

IVV

C w

ith

PV

@ P

F=0

.95

IVV

C (

cen

tral

PV

co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(e

xclu

de

PV

)

IVV

C w

ith

PV

@ P

F=0

.95

IVV

C (

cen

tral

PV

co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(e

xclu

de

PV

)

IVV

C w

ith

PV

@ P

F=0

.95

IVV

C (

cen

tral

PV

co

ntr

ol)

Bas

elin

e

Loca

l PV

Co

ntr

ol (

PF=

0.9

5)

Loca

l PV

Co

ntr

ol (

Vo

lt/V

AR

)

Lega

cy IV

VC

(e

xclu

de

PV

)

IVV

C w

ith

PV

@ P

F=0

.95

IVV

C (

cen

tral

PV

co

ntr

ol)

2014 2024 2034 2044 2054 2064

Cat

ego

rica

lly A

ggre

gate

d C

um

ula

tive

Dis

cou

nte

d C

ash

Flo

w

Year

Equipment Replacement

Losses

Categorically aggregated cumulative discounted cash flows for costs of feeder losses and equipment replacement

Categorical cost and savings compared to baseline

October 2, 2017 42October 2, 2017 42

Case Study 1: Conclusion

► This work Illustrates the potential for coordinated control of voltage management equipment,

such as the central DMS-controlled IVVC by:

◼ Providing substantial improvement in distribution operations with large-scale PV systems

◼ reducing regulator operations

◼ Decreasing the number of voltage challenges

► The preliminary cost-benefit analysis showed operational cost savings for the IVVC

scenarios that were:

◼ partially driven by reduced wear and tear on utility regulating equipment,

◼ but dominated by the use of CVR/Demand reduction objective

► Work needed in the area of integrating advanced inverters as controllable resources into

IVVC optimization strategies

◼ Event triggered operation of DMS IVVC

◼ Power factor set point in place of reactive power set point

October 2, 2017 43

Thank you

Questions?

Backup slides

October 2, 2017 44October 2, 2017 44

Case Study 2: Voltage Support Using Smart

PV Inverters

• 2040 nodes, serving ~4.15 square miles.

• Peak load = 9.89MW

• One controllable 1200 kVAr capacitor: switch on

at 6 am, switch off at 10 pm

• ~1 MW existing PV power with standard inverters

(~10% penetration)

• ~700 kW planned PV with smart inverters (~7%

penetration)

• ~500 kW planned PV with standard inverters

(~5% penetration)

• Planned PV Systems are distributed at 16 service

transformer locations.

F. Ding, A. Pratt, T. Bialek, F. Bell, M. McCarty, K. Atef, A. Nagarajan and P. Gotseff, “Voltage support study of smart PV inverters on a high photovoltaic penetration utility distribution feeder,” in IEEE 43rd Photovoltaic Specialists Conference (PVSC), Portland, OR, June 2016.

October 2, 2017 45October 2, 2017 45

Voltage Impact Analysis

• Maximum Load Day• Minimum Load Day• Clear Day• Cloudy-Intermittent Day• PV Intermittent Day

For each type of day, simulated:• Unity power factor• Six fixed power factors:

o 0.95, 0.9, 0.85 leadingo 0.95, 0.9, 0.85 lagging

• Three Volt/VAr curves

October 2, 2017 46October 2, 2017 46

Voltage Impact Analysis: Fixed Power Factors

Maximum voltage change throughout the feeder under different fixed power factors for each type of days.

October 2, 2017 47October 2, 2017 47

Voltage Impact Analysis: Volt/VAr Control

AvailableQ = Sinverter2 -Pout

2

October 2, 2017 48October 2, 2017 48

Voltage Impact Analysis: Volt/VAr Control

Maximum load days: Voltage Results at Two AMI locations

The voltage range (maxV-minV) was reduced the most with Volt/VAr curve-3: by 0.007 p.u. (22%) at AMI-1 and by 0.006 p.u. (28%) at AMI-2.

October 2, 2017 49October 2, 2017 49

Voltage Impact Analysis: Reactive Power

Output

• At Phase A: PV systems 1, 2, 3, 4, 6, 9, 12, 13, 14, 15, 16.

• At Phase B: PV systems 5, 10, 11.• At Phase C: PV systems 7, 8.

Reactive Power Output: Volt/VAr Curve-1Max Qtotal = 134 kVAr

Terminal voltages of three representative smart inverters, one for each phase.

October 2, 2017 50October 2, 2017 50


Output

Reactive Power Output: Volt/VAr Curve-2

Max Qtotal = 150 kVAr

October 2, 2017 51October 2, 2017 51


Output

Reactive Power Output: Volt/VAr Curve-3

Max Qout = 250 kVAr(33% of rated kVA rating)

October 2, 2017 52October 2, 2017 52

Voltage Impact Analysis: Statistics

Summary

• Voltage range reduced by up to 0.009 p.u. (11%) on maximum load days.

• Standard deviation reduced by up to 0.002 p.u. (23%) on minimum load days.

Voltage range and standard deviation for 1) unity power factor, 2) Volt/VAr curve-1, 3) Volt/VAr curve-2, and 4) Volt/VAr curve-3. Data in red indicate values beyond 1.5 times the interquartile range.

“Tighter” Voltage Profile

October 2, 2017 53October 2, 2017 53


• Results from a specific utility feeder:

• Fixed 0.85 power factor change in primary voltage just over 1%

• 700 kW of PV with smart inverters had comparable voltage impact

to 1,200 kVAr switched capacitor

• Volt/VAr reduced range and standard deviation of voltages

• No deadband had most significant voltage profile improvement

• Reduced voltage range up to 11%; standard deviation up to 23 %

October 2, 2017 54October 2, 2017 54

Conservation of Voltage Reduction

Conservation of Voltage Reduction (CVR): A voltage reduction scheme that flattens and

lowers the distribution system voltage profile to reduce overall energy consumption.

• Works best with resistive and constant impedance loads

• Normally performed by flattening the system voltage

using capacitor banks and/or voltage regulators and

lowering the voltage by controlling a substation Load Tap

Changer

• Also performed by using a central volt/VAR optimization

performed by a distribution management system (DMS)

Recloser 1

Recloser 2

Recloser 3

Cap Bank 1

Cap Bank 2

Regulator 1

Regulator 2

Regulator 3

Feeder Head – Breaker/Regulator DMS

PV Inverter

Capacitor Bank

Voltage Regulator

October 2, 2017 55October 2, 2017 55

Case Study 3: Smart Inverter Volt-VAR

Function for CVR and Power Quality

LTC

CapacitorPV/Smart

Inverter

Randomly allocate PV locations and smart inverters

Various PV penetration levels, Various Smart inverter densities

Voltage Optimization Methodology

Annual Energy Saving Power Quality

QST

S Si

mu

lati

on

Objective: Evaluate the effects ofdistributed PV with smart inverters onthe conservation of voltage reduction(CVR) energy savings and power qualityin distribution systems.

F. Ding et al., "Application of Autonomous Smart Inverter Volt-VAR Function for Voltage Reduction Energy Savings and Power Quality in Electric Distribution Systems,” IEEE Innovative Smart Grid Technologies , Washington DC, 2017.

October 2, 2017 56October 2, 2017 56

Voltage Reduction Optimization Method

Energy saving is used to measure thevoltage reduction effect:

VRsaving =EnergyVR -BaseEnergynoVR

BaseEnergynoVR×100%

CVR factor (the percentage change in energy consumption per percent change in the load’svoltage): 0.8 (real) and 4.0 (reactive)

Start

Disable smart inverters

Capacitor Optimization LTC Optimization

Enable smart inverters

Solve Power Flow with Smart Inverters

Vmin>=0.95? Tap up the LTC by 1 step

Tap down the LTC by 1 step

Capacitor Optimization

Solve the power flow, Vmin>=0.95?

Restore to the previous capacitor states and LTC tap position

Record the present LTC tap position and capacitor statesYES

NO

Stop

YES

NO

Voltage Reduction Without Smart

Inverter

• Capacitor Optimization: flattest voltage • LTC Optimization: adjust LTC tap position to

achieve the lowest voltage • Autonomous Smart Inverter Volt-VAR Control

October 2, 2017 57October 2, 2017 57

Power Quality Scoring Methodology

Power QualityMetrics

System Average Voltage Magnitude Violation Index(SAVMVI): average value of all voltage magnitudeviolations for all buses at all time steps.

System Average Voltage Fluctuation Index (SAVFI): average valueof voltage fluctuations for all buses at all time steps.

System Average Voltage Unbalance Index (SAVUI): average valueof voltage unbalances for all three-phase buses at all time steps.

System Control Device Operation Index (SCDOI): total number ofcapacitor switching operations and LTC tap changes.

SAVMVI =1

N

1

TVIOVmag

i t( )t=1

T

åi=1

N

åæ

è

çç

ö

ø

÷÷

Location

-2

Location-N

Location

-1

QSTS

System Reactive Power Demand Index (SRPDI): average value of reactive power demands from thesubstation during the entire simulated period.

System Energy Loss Index (SELI): the ratio of total energy loss and total load demand during the entiresimulated period.

October 2, 2017 58October 2, 2017 58

Power Quality Scoring Methodology

Power Quality Score (PQS) Both min and max of each metric are calculated from the definition or determined from IEEE standards and practical limits.

Model Property Value

Substation Bank Size 45 MVA

Circuit Primary Voltage 21 kV

Number of Circuits 3

Peak Bank Load 37.09 MW

Max Circuit Distance 6.63 miles

Total Primary Circuit

Miles85.65 miles

Capacitor Banks 7 (12 MVAr total)

SAVMVI/SAVFI/SAVUI/SCDOI/SRPDI/SELI

Scenario

PV

Penetration

0, 10%, 30%, 50%, 100% of peak

load

Smart Inverter

Penetration0, 25%, 50%, 100%

Voltage

OptimizationWith, Without

Simulation 1 year with 1 h time step

Case Study

[Min(Metric), 10]

[Max(Metric), 0]

Metric Value

Individual Score

S(.)

October 2, 2017 59October 2, 2017 59

One-Year Energy Saving

0.95

0.97

0.99

1.01

1.03

0 5 10V

olt

age

(p

.u.)

Distance from the substation (km)

Base CaseWithoutCVR

Voltage profile at one time step obtained for three cases

Voltage reduction energy savings at different PV penetrations and smart inverter densities.

October 2, 2017 60October 2, 2017 60


• Voltage reduction energy savings increased with autonomous smart inverter Volt-VAR

control. Smart inverters with a lower VVC band center allowed the tap position of the

substation LTC to be lower, compared to cases without smart inverters. This resulted

in a lower distribution system voltage profile and increased voltage reduction energy

savings.

• Since voltage reduction energy savings were prioritized over the PQS, the

implementation of the proposed voltage reduction scheme lowered certain power

quality scoring metrics, including SCDOI and SRPDI, leading to an overall lower PQS.

• Overall without CVR VO, smart inverters had a positive impact on the PQS, and helped

to reduce energy losses and voltage fluctuations.

Documents

Advanced Distribution Management Systems and … 2, 2017 1 Advanced Distribution Management Systems and Distribution Automation Distribution Systems and Planning Training for New England