distribution system evolution: implications for sensing and ......device failure short circuit labor...

Distribution System Evolution:Implications for Sensing and Measurement

EAC Panel Discussion13 September 2017

Jeffrey D. Taft, PhD

Chief Architect for Electric Grid Transformation

Pacific Northwest National Laboratory

PNNL-SA-128900

Topics

• Distribution Problem Domain Model

• Key Trends for Distribution evolution

• Present Status and Core Issues for Distribution Sensing

• Sensing and Measurement Implications for Grid Modernization

2

DER Problem Domain

Merchant DG

DC/AC

Inverter

OLTC OLTC

OLTC OLTC

Distribution

Operations Center

Primary Distribution Substation

Primary Distribution Substation

Fast

stabilization

Recloser

Recloser

Distribution

Transformers

EV Charging Stations

Utility

Bulk Storage

DC/AC

Inverter

RPower Flow

ControllerR

DC/AC

Inverter

Merchant DG

Distribution

Transformer

Distribution

TransformerDistribution

Transformer

Distribution

Transformer

Line

Sensor

Line

Sensor

Line

SensorLine

Sensor

Transactive

Commercial

Building

Prosumer DG

Prosumer DG

Merchant DG

Merchant DG

Merchant DG

Line

Sensor

Power Flow

Controller

Power Flow

Controller

V/VAr

Control

Agent

V/VAr

Control

Agent V/VAr

Control

Agent

Voltage

Control

Agent

Merchant Storage

The “Energy Bank/

Warehouse”

Ctrl

G

Ctrl

VER

Tertiary

CtrlVolt/Var

regulation

Storage

Secondary

Ctrl

LoadDR

Sync/PCC Microgrid

To other

feeder loads

Shunt

Capacitor

Distribution

Transformer

DER

AggregatorDER

AggregatorDER

Aggregator

Trend: Renewables Integration Changes Grid Operations

Wind Variability

Solar Variability

6

Introduces balancing and stability issues to the grid.

Introduces fast dynamics to the distribution system, affecting voltage regulation and system stability.

Trend: Less Time, More Data, More Endpoints

Increasingly faster device/system dynamics

Moving from slow data sampling to fast streaming data

Massive numbers of sensing and control endpoints

Thousandsof endpoints

Tens of millions of endpoints

WAMS/AMI/DER/DR/networking…

Source: Alexandra von Meier

Dynamics Trending Faster

5

Future Distribution = Load + Gen Tie Point

Underlying diagram source: EPRI

DistributionSystem

Bulk EnergySystem

internet

internet internet

internet

internet

internet

8

Preferred DER Structure

System Operator

Agg/ESO

DSO/DO

DER

inside a single

DO service area

Prosumer

Merchant

Community

< owns

< owns

< owns

electrically connects to >

supplies grid services to or thru >

supplies grid services to or thru>

< aggregates & dispatches

< coordinates & dispatches


< m

ay o

pera

te

< o

pe

rate

s

indirectly supplies grid services thru >

Electric Distribution Grid

is o

pera

ted

by >

Bulk Energy

System

< electrically connect at T/D interface >

is o

pera

ted

by >

DER Grid Services Entity-Relationship DiagramBoxes represent entity classes

Lines represent relationships (read relationship text in direction of the arrow)

< coordinate at T/D interface >

One; any particular

One and only one; a single

Zero or more; optionally many

One or more; at least one

Cardinality Symbol Key

con

tracts

with

>

co

ntra

cts

with

>

Note: all DER-based grid services are

delivered to and through the electric

distribution grids. The term “supplies”

refers to the business arrangement.

DER

Markets

Wholesale

Markets

Other Key Trends

8

Monolithic Partitionable

Vertical siloes Layers and platforms

Rigid/brittle Cushioned/tensile

Weakly observable Strongly observable

Coordination gaps Pervasive coordination

DSOs will need superb distribution grid measurement capability to support demanding distribution grid operational capability.

Where Does Distribution System Sensing Stand Now?

• Many distribution systems have little or no distribution SCADA

• Many distribution substations have no SCADA

• Most existing distribution sensing is low speed and low capability– Mostly polled or exception reporting

– Minimal sensing of grid variables

• Older AMI is not very useful for grid sensing; newer technology is improved but not widely deployed

• Communication networks for distribution are often weak, siloed, and non-converged– Poor performance (low bandwidth, lossy, insecure)

– Not future-proofed

– Cause high integration costs and complexity

• Poor grid topology model accuracy (needed for data context)

• Grid sensor installation is expensive

Common Distribution Grid Sensors

• Line sensors– Voltage, current, or both; compute real/reactive power, THD and harmonics

• Control device sensing– Voltage, current or both

• Faulted circuit indicators

• Premises meters – residential

• Premises meters – C&I

• Feeder meters

• Substation sensing

• Power quality monitors

Lindsey CVI Line Sensor

Fisher Pierce FCI

What Do We Need to Know for Distribution Grid Operations?

Extended Grid Statev2.4

System

Electrical

throughputs

Note: see discussion about

outage in the presence of DG/

DS and/or microgrids (service

outage vs. grid outage).

Communications

latencies packet losses port reachability

performance

port state

status

Asset

Status

set point servicefault age

Condition

MarketsStructure

topologies

circuits comms

parametrics

Ambient

insolationwater flow

gen cooling

Other

airborne

particulates

procured resources

Component, Circuit, and

System Models

Solar Flux, Wind, and

Storm ForecastsDemand and DER forecasts

Constraints, Utilization,

Locational Net Value

accum. stress

path

distances

vulnerability

securityexposures

cyber physicalhuman

vehicles

digging

precip

humidity

wind

temp

fog

Weather

space weather

GMD

GIC

model params

admittance

matrices

component

impedances

DER hosting conducting

capacities

Estimators, Modelers, Analytics, Forecasters

as-built

as-operated

Type/Config

Note: assets also include work

force and field crews, control room

systems, servers, databases, and

software applications

Thermal

hot spots profiles

utilization integrity

chemstruct elect

quake

flood

storm

volcano

sinkhole

nat. disaster

fire

enviro

icing

veg

salt fog

lightning

animal

Group

Category

Root

Kind

Set

Type

Exa

mp

les

Ta

xo

no

my C

lasse

s

sag DGA

op

count

storage

capacity SoC

conduction

V, I P, Qf, f

generation

capacity reserves fuel mix bulk/DG mix power quality

grid outage

system performance

reliability

op effectiveness distortion violations

unbalance

consumption

losses

gross load rampingnet load

DR (flexibility)

capacity actual (invoked)

demand

energyessential reliab svcs capacity

T NWA

LMPs

prices

D NWA

flexibility

contract prices

flexibility

posture

planned unplannedavail

capacity

spot

NERC

definedelectrical

comms

data

training

Geospatial Information

location

imagery

terrain descriptiondevice failure

short circuit

labor strike

security breach

open/closed

in/out

under attack

Note that examples (purple

boxes) are indicative of

class elements but are not

intended to be exhaustive

lists.

Distribution Grid Measurement Issues

• Control System PoV (grid state level)– Dynamic system snapshot

• Instrumentation PoV (AC waveform level)– Volt/VAr control; feeder phase unbalance

– Real power flow (incl. direction)

– Synchronization (DG, microgrids, power switching)

– Fault detection, characterization & localization (3-phase AC methods)(e.g. methods of Bollen & Gu, Naidoo & Pillay, Krishnathovar & Ngu)

– Asset utilization & asset condition• Many methods use complex impedance measurement

• Data Sharing PoV (application level)– Low latency multi-user access needed

– High back end integration cost/complexity

Implications for Modernized Distribution GridSensing & Measurement

• Electrical Measurement– Fast waveform sampling

– Magnitudes and phases (unbalanced)

– High precision (small phase differences)

– Timing/synchronization

• Architecture– Infrastructure layer

• Networking– Redundancy

– Multicast streaming (SSM)

• Sensor technology– Grid topology sensing

– Inexpensive installation

• Design– Observability methodology

– Allocation tool

Physical Infrastructure

Communication

Network

R

DG

DG

Device Operations

(Feeder)

Circuit Operations

(Substation)

System Operations

(Control Center)

Sensor Network

Network

Services

Micro

virtualizer

Legacy

sensorSmart

sensors

Application 2

Application 1

Application 3

SensorNetwork

SensorAllocation

Observability Strategy

Final Comments• 20th Century distribution grids did not need much

observability

• Modernized grids with high DER penetration and advanced capabilities will need advanced sensing and measurement and data transport– Fast synchronized sensing & measurement

– Adequate coverage (observability strategy-> sensor network design)

– Inexpensive installation

– Grid topology and system models

– Volatility at the edge

• Sensing and networks as core infrastructure layer– Streaming sensor data

– Timing distribution

– Synchronized measurements

• Heavy lift for many distribution utilities to get to this level– Many advanced grid management schemes will have to wait until this level of

upgrade happens

15

Thank You

Jeffrey D. Taft, PhD

[email protected]

Existing DER Structure

System Operator

Agg/ESO

DSO/DO

DER

inside a single

DO service area

Prosumer

Merchant

Community

< owns

< owns

< owns



< dispatches

supplies grid services to >





< dispatches


< o

pe

rate

s



is o

pera

ted

by >

Bulk Energy

System


is o

pera

ted

by >



One; any particular




Cardinality Symbol Keyco

ntra

cts

with

>

con

tracts

with

>

co

ntra

cts

with

>





Wholesale

Markets

DO

Likely Near Term DER Structure

System Operator

Agg/ESO

DSO/DO

DER

inside a single

DO service area

Prosumer

Merchant

Community

< owns

< owns

< owns



< dispatches






< dispatches


< m

ay o

pera

te

< o

pe

rate

s



is o

pera

ted b

y >

Bulk Energy

System


is o

pera

ted b

y >



< coordinate at T/D interface >

One; any particular




Cardinality Symbol Keyco

ntra

cts

with

>

con

tracts

with

>

con

tracts

with

>





DER

Markets

Wholesale

Markets

Separate Components from Infrastructure via Layering

18

• Siloed, coupled apps

• Long latency

• Poor flexibility

• Expensive integration

• Independent apps

• Low latency

• High functional flexibility

• Low cost integration

• Future-proofed investment

Structure Transformation

Hawaii PUC Docket No. 2016-0087, Order No. 34281, Dismissing Application Without Prejudice and Providing Guidance for Developing a Grid Modernization Strategy, Jan. 4, 2017 pp. 54-57

Electricity and Energy Systems are Changing Organically

• Business and operating models are being

forced into change

• Grid economics is headed into new territory

• Consumer expectation and technology are

bigger drivers than regulation or policy

Source: P De Martini



distribution system evolution: implications for sensing and ......device failure short circuit labor...

Documents