NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Advanced Grid Technologies Laboratory Workshop Series

Murali Baggu, Ph.D.

Manager, Distributed Power System Section

July 8th 2015

Distribution Management System Testing at NREL

2

Core Research Question(s)

How do high-penetration DGPV operational impacts—both technical and economic—compare among:

1. Baseline: PV active power only

2. Local Autonomous Control: PV reactive power support and voltage regulation, etc.

3. Central control: Alstom IVVC/LVM application

And (for 2&3), what specific settings provide the best performance?

3

Test Feeder

Recloser 1

Recloser 2

Recloser 3

Cap Bank 1

Cap Bank 2

Regulator 1

Regulator 2

Regulator 3

Feeder Head – Breaker/Regulator

4

Operational Impact Studies in DOTS

Idea: Utilize the Distribution Operator Training Simulator (DOTS) to perform time-series power system simulations. Compare baseline and hypothetical scenarios to determine operational impacts or analyze benefits. Methodology: 1. Import historical loading, switching and DG output data 2. Run time-series power flows to baseline simulation 3. Update system for hypothetical case and rerun simulation 4. Compare simulation results and analyze delta.

Advantage over traditional planning tools: 1. Leverage existing DMS model 2. Easy to include LVM control in studies 3. Automate simulation setup, data capture and reporting…

5

Python Scripting Package: Architecture

Web

Service TCP/IP

DMS Study Server

Interface Server

py4etd Module

Python Scripts

Python Scripts

Python Scripts

ETB/ DMS Client

iGrid

SQLite DB CSV Files

6

Picking the best days with statistics

1. Identify multivariate regression model formulation

2. Randomly sample from day types (stratified sample)

3. Parameterize regression model based on 40 days

4. Bootstrap validation that regression model estimate of “annual” (sampled days) capacitor bank switches.

5. Repeat steps 5-7 hundreds of times

6. Select best 40-day sample: low average error & tight confidence intervals

Note: Analysis with R v3.3

7

Summary Block Diagram

500kW Advanced Inverter

600kVAr CapBank And controller

PV Simulator

SCADA Control

Opal-RT Real-time Simulator

8

Lab Setup

9

Cost Benefit Analysis

• ds

10

Looking forward

• Halfway through this project: o Tools built and validated

o Time to run scenarios

o And compare with PHIL

• An Environment for further research o Battery Storage impacts

o Other technologies

o Other feeders

• Expanded Financial Analysis