keeping it together - international institute for energy...

1

© 2017, General Electric Company. Proprietary information. All rights reserved.© 2016 General Electric Company - All rights reserved

Keeping it Together

International Conference on Energy Systems Integration; Denver | December 6, 2017

Nicholas W Miller, Senior Technical Director, Energy Consulting, GE Power

Transient stability in a world fully electrified by wind & solar generation

Confidential. Not to be copied, reproduced, or distributed without prior approval.

CAUTION CONCERNING FORWARD-LOOKING STATEMENTS:

This document contains "forward-looking statements" – that is, statements related to

future events that by their nature address matters that are, to different degrees,

uncertain. For details on the uncertainties that may cause our actual future results to

be materially different than those expressed in our forward-looking statements, see

http://www.ge.com/investor-relations/disclaimer-caution-concerning-

forwardlooking-statements as well as our annual reports on Form 10-K and quarterly

reports on Form 10-Q. We do not undertake to update our forward-looking

statements. This document also includes certain forward-looking projected financial

information that is based on current estimates and forecasts. Actual results could

differ materially. to total risk-weighted assets.]

NON-GAAP FINANCIAL MEASURES:

In this document, we sometimes use information derived from consolidated financial data but not presented in our financial statements

prepared in accordance with U.S. generally accepted accounting principles (GAAP). Certain of these data are considered “non-GAAP

financial measures” under the U.S. Securities and Exchange Commission rules. These non-GAAP financial measures supplement our GAAP

disclosures and should not be considered an alternative to the GAAP measure. The reasons we use these non-GAAP financial measures

and the reconciliations to their most directly comparable GAAP financial measures are posted to the investor relations section of our

website at www.ge.com. [We use non-GAAP financial measures including the following:

• Operating earnings and EPS, which is earnings from continuing operations excluding non-service-related pension costs of our principal pension plans.

• GE Industrial operating & Verticals earnings and EPS, which is operating earnings of our industrial businesses and the GE Capital businesses that we expect to retain.

• GE Industrial & Verticals revenues, which is revenue of our industrial businesses and the GE Capital businesses that we expect to retain.

• Industrial segment organic revenue, which is the sum of revenue from all of our industrial segments less the effects of acquisitions/dispositions and currency exchange.

• Industrial segment organic operating profit, which is the sum of segment profit from all of our industrial segments less the effects of acquisitions/dispositions and currency exchange.

• Industrial cash flows from operating activities (Industrial CFOA), which is GE’s cash flow from operating activities excluding dividends received from GE Capital.

• Capital ending net investment (ENI), excluding liquidity, which is a measure we use to measure the size of our Capital segment.

• GE Capital Tier 1 Common ratio estimate is a ratio of equity

Keeping it Together

6 December, 2017

3

© 2017, General Electric Company. Proprietary information. All rights reserved.

What the policy people see3

Source: NREL

4


But, the grid people need to keep the lights on ALL the time4

Source: GE NSPI Study

Enough wind tomeet 30%

annual energypenetration

5


At periods of high instantaneous penetration, things get interesting

5

Wind 4.5

PV 7.5

CSP 2.2DG5.7

Others 12.4

Wind 8.3

PV 0.2CSP 0.0DG 0.1

Others 9.3

Wind 7.0

PV 0.7CSP 0.0DG 0.7

Others 4.1

Wind 10.9

PV 5.1

CSP6.1

DG 3.8

Others 9.1

Production/Dispatch in GWSource: NREL/GE WWSIS 3a

6


What the grid people see6

• ~10000 nodes• 38 balancing authorities• ~120,000 circuit miles of

transmission• ~4000 generators-

~265GW • ~5000 loads• ~100,000 state variables

7


Is it stable?

Derived from original figure by ElgerdElgerd, Olle, I. 1971. “Electric Energy Systems Theory: An Introduction” McGraw-Hill, pg 478.

8


8

What happens when the flow patterns change?

Heavy summer Base

Heavy summer Base with high COI flows

Heavy summer High Mix with high COI flows

12

3

2

3

1

California, Oregon interface power flow (MW)

4,800 MW

Time (Seconds)

P (M

M)

A disturbance on the north-south

corridor between Oregon and California

(Trip of the Pacific DC Intertie)

Source: NREL/GE WWSIS 3a

Same generation

& new flows -failure

More wind & solar AND

new flows – fails too, but less badly!

9


“Strong Grid” “Weak Grid” “Impending Fault”

What about “weak grids”? What is Grid Strength?• Grid strength is like a “stiffness” of a power system

• It is specifically for voltage (not frequency)

• Unlike frequency stability, location matters

• In a strong grid, bus voltages do not change much when the system is ‘whacked’ by a disturbance like a fault

• In a weak grid, bus voltages change a lot during disturbances like faults

10


0

5000

10000

15000

20000

25000

30000

35000

Base Hi-Mix Extreme Base Hi-Mix Extreme

Dis

pat

ch/P

rod

uct

ion

(M

W)

WIND

PV

CSP

GAS+

HYDRO

NUCLEAR

COAL

Ge

ne

rati

on

pro

du

cti

on

(GW

)

Desert Southwest

Northeast (of the West)

Co

al

35000

30000

25000

20000

15000

10000

5000

0

Co

al

Lightspringbase

Lightspringhigh mix

Lightspringextremesensitivity

Lightspringbase

Lightspringextremesensitivity

Lightspringhigh mix

Wind and Solar Displacement of Thermal (East portion of the Western Interconnection)

11


Weak Grid Example: Transient Stability in Northeastern Western Interconnection

L

Aeolus

500kV

Large Coal PlantsGateway South 500kV transmission project

Gateway West 500kV transmission project

12


It depends on how stressed the system is:

0

0.2

0.4

0.6

0.8

1

1.2

0 0.5 1 1.5 2

Wyo

dak

23

0kV

Vo

ltag

e (p

u)

Time (Seconds)

High Mix Extreme

Highly stressed,

but stable

Too stressed, unstable: failure is very fast


13


• Synchronous machines

(generators, condensers)

• Motor Loads

• Most of today’s power electronic-

based generation (PV, Wind,

HVDC)

• Battery storage

• Other Loads

What Contributes to Grid Strength?

And what does not (today)?

14


Dave Johnson Voltage

Light Spring BaseLight Spring High Mix Light Spring ExtremeLight Spring Extreme with synchronous condenser conversion

Synchronous Condenser Conversion Results in Acceptable Performance even under extreme stress

Disturbance: Aeolus bus fault and line trip

1

3

2

4

1

2

3

4

Reinforcements for Extreme sensitivity makes the system stable again: 3 condensers total ~1700MVA plus ~500 MVAr shunt banks.

Repurpose the old stuff!


15


Weak Grid – why the wind & solar need to keep improving

Small wind plant relative to grid Large wind plant relative to grid

1 Ton


16


Better controls can make a huge difference:

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 0.5 1 1.5 2

Wyo

dak

23

0kV

Vo

ltag

e (p

u)

Time (Seconds)

High Mix

Extreme without pre-disturbance overloads with less aggressive LVPL

Extreme without pre-disturbance overloads with OEM weak grid controlsFail with “standard”

controls

“saved” with weak grid controls on the

wind plants


17


“Load Netting” isn’t adequate anymore

P = P0((ap +bpV +cpV2) x (1+kpfF))

Q = Q0((aq +bqV +cqV2) x (1+kqfF))

Today: we simplify to

this!!!!P + jQ

General practice is for “load netting”:

What is lost when distributed PV is “buried” in net load:

• Load shedding is wrong.

• Vulnerability to common mode tripping of dPV is lost.

• Weather impacts can’t be captured.

• Dynamics of dPV are lost.

“ZIP” load model

18


Composite Load Model: “CMPLWG”*Structure with Distributed Generation

Electronic

M

M

M

Loadflow

Bus

Static

M

UVLS

UFLS

PLagg

PV gen.

PLnet

Pdg

Pma

Pmb

Pmc

Pmd

Pel

Pst

QLnet

Qdg

*WECC’s Composite Load Model Working Group”

19


The behavior of the load can dominate the system dynamics

Same disturbance* - very different

outcome

Std WECC load model (blue)

192GW total load modeled as induction motor (18%

overall) + static – fast voltage recovery, very stable

response

Composite load model (red)

143.9 GW modeled as composite load – fails to recover

There’s a shocking dearth of actual supporting data!!!

Load behavior dominates system response – what will ESI “loads” look like?

*

• 3Φ fault at Vincent 500 kV

• Clear fault and trip 2 Midway-Vincent lines in 6 cycles

• No load shedding in either case

Source: NREL/GE WWSIS 3

20


20

Fault ride-through is critical:Impact of common-mode tripping of DG without low-voltage ride-through

Pessimistic

DG with LVRTDG without LVRT

1

2

1

2

Distributed PV tripped on fault: took entire

system down.

Distributed PV behaved well on fault: recovery

satisfactory

Source: NREL/GE WWSIS 3

21


Is it stable with ESI?

Derived from original figure by Elgerd

Goran’s factories

22


What happens in an ESI world?

It seems inevitable that this simplification will be

wildly inadequateP + jQ

The “load” becomes an even more complex and active element

What might bite us?

• “Prosumer” fault ride-through

• Vehicle to Grid; Grid to Vehicle dynamics.

• Active thermal loads; new load dynamics

• Embedded energy storage.

• Load dynamics driven by weather or externalities (that we

ignore today).

23


Main grid

Subgrid with only load and

inverter-based

generation

Tie-line(s)

What happens when the grid breaks up? Consequences in an ESI World are even more dire…..

24


We’ll need grid forming inverters…and a lot of other brains

Metlakatla 1MW/1.4MWhr BESS. C 1996

25


Take aways


The Grid and ESI future is only as reliable as it is stable• Reliability must be and can be as good or better, with well designed systems. • Flexibility, flexibility, flexibility• Transmission is (still) needed.

System Dynamics are already complex, and getting more so.• Understanding and tools must evolve• ESI brings a whole new layer of complexity and opportunity

Using the actual power system as a simulator is uneconomic and

irresponsible• We’ve got research to do, and we’d better get to it!

26


Thanks…

[email protected]

Source: GE Power Marketing, GE Energy Consulting

keeping it together - international institute for energy...

Documents