renewable energy integration: the grand challenges...

1

RENEWABLE ENERGY INTEGRATION: THE GRAND CHALLENGES AND

OPPORTUNITIES TOWARDS A SUSTAINABLE ENERGY FUTURE

presentation by George Gross

University of Illinois at Champaign-Urbana at the IEEE PES Distinguished Lecture Program at

the IEEE PES Chapter of the South Australia SectionUniversity of South Australia, Adelaide, Australia

September 29, 2015© 2015 George Gross, All rights Reserved.

2

PRESENTATION OBJECTIVES

Provide an overview of the challenges and opportunities raised by the large–scale integration of renewable resources

Provide an up–to–date assessment of the status of renewable energy implementation

Discuss key challenges and opportunities arising from current practice and research

Suggest directions for future research to address the specific challenges and opportunities in renewable energy integration

3

THE CHANGING ENVIRONMENT There is a growing worldwide interest in the

integration into the grid of renewable resources, as well as, demand response and storage resources, to displace costly and polluting fossil–fuel–fired generation

The integration progress pushes the creation of sustainable paths to meet each nation’s energy needs, veering it towards energy independence

The context within which this progress unrolls is the restructured, competitive electricity industry, taking full advantage of the advances in the Smart Grid implementation throughout the world

4

THE SMART GRIDThe smart grid represents a modernized electricity delivery system that monitors, protects and automatically optimizes the operation of all its interconnected elements – from the central and distributed generators through the high–voltage transmission grid and the distribution network, to industrial users and building automation systems, to energy storage devices and to end–use consumers and their thermostats, electric vehicles, appliances and other devices.

5

THREE SALIENT ASPECTS

Combined digital intelligence and real–time

communications: to improve the operations/con-

trol of the transmission and distribution systems

Advanced metering solutions: to replace the

legacy metering infrastructure

Deployment of appropriate technologies, devices,

and services: to access and leverage consumption

information in smart appliances and system state,

including the integrated renewable resources

6

TODAY’S ENERGY CHALLENGE: GROWING DEMAND

growth in primary energy demand

growth in electricity demand

China

105% 195%

India

126% 282%

Europe & North America

11% 31%

M. East & Africa

73% 131%

SouthAmerica

56% 81%

Source: IEA forecast 2007-30

7

TODAY’S LEGACY GRID

one-way power flows

centralizedgeneration

coal

hydro

nuclear

end use customers

industrial

commercial

residential

substation

high voltage transmission

medium-low voltage

distribution

8

SMART GRID INFRASTRUCTURES

Source: EPRI Intelligrid

the information layer

active consumers

emission reductions

increased efficiencies

the electrical infrastructure

renewable integration

situational awareness

improved productivity

9

THE SMART GRID WILL…

Improve the ability to connect new loads, such as

battery vehicles, and provide new products and

services

Facilitate the integration of distributed energy

resources, storage devices, demand-response

resources at deeper penetration levels

Provide the ability to collect and store information

to effectively address climate change issues

10


Enable customers to become active participants in wholesale/retail electricity markets and able to manage/control/optimize electricity usage

Operate resiliently under cyber and human-produced attacks and in case of natural disasters

Provide situational awareness for bulk power system and distribution operations

Anticipate and effectively respond to a wide range of disturbances

11


Reduce drastically the time for service restoration

after the onset of an outage

Improve reliability of power system operations

Optimize resource and equipment utilization and

enhance efficiency of electricity production and

delivery

12

SMART GRID VISION

Enhance customer value

Empower customers through knowledge

Utilize electricity more efficiently

Implement cost-effective technological

advancements

Reduce future demand growth

Support “green” environmental initiatives

Improve reliability and system security

13

BEYOND THE SMART GRID…

14

hydro 6.3 %

coal39.1 %

nuclear 19.5 %

petroleum1.0 %

other renewable sources 6.9 %

RENEWABLES’ ROLE IN THE 2014 US ELECTRICITY SUPPLY

Source: http://www.eia.gov/totalenergy/data/monthly/pdf/sec7_5.pdf; issued April 2015

natural gas27.4 %

coal38.7 %

15

RENEWABLES’ ROLE IN THE 2014 US ENERGY SUPPLY

Nuclear

98.5 quadrillion BTU 9.6 quadrillion BTU

Nuclear

Source: http://www.eia.gov/totalenergy/data/monthly/pdf/sec1_7.pdf

coal18.3 %

natural gas28 %

petroleum35.3 %

nuclear8.5 %

hydro 25 %

wind 18 %

biomass 50 %

renewables 9.8 %

solar PV 4.4 %

geothermal 2.3 %

16

SOLAR ENERGY

Images From: http://www.scientificamerican.com/article.cfm?id=how-to-use-solar-energy-at-night

17

2014 SOLAR ENERGY STATUS The global PV cumulative

capacity reached 177 GWin 2014 with ability to produce 200 TWh of electricity every year

50 % of the world’s total PV cumulative capacity –88.5 GW – is installed in Europe

In 2014, the annual growth rate of PV has reached almost 28 %, and is currently the third largest renewable energy source in terms of global, installed capacity

Lakelands Park Middle School, MD hosts a 111 kW rooftop system

Source: http://www.iea-pvps.org/fileadmin/dam/public/report/technical/PVPS_report_-_A_Snapshot_of_Global_PV_-_1992-2014.pdf

18

2014 INSTALLED SOLAR CAPACITY

Source:http://www.cleanenergybusinesscouncil.com/site/resources/files/reports/EPIA_Global_Market_Outlook_for_Photovoltaics_2014-2018_-_Medium_Res.pdf

China (18,051; 13 %)

Italy (18,051; 13%)

Germany (36,103; 26% )

Rest of the World (13,886; 10 %)

India (2,7172 %)Greece (2,717 2 %)

United Kingdom (2,717; 2 %)

Czech Republic (2,717; 2 %)

Australia (2,717; 2 %)Belgium (2,717; 2 %)

France (4,166; 3 %)Spain (5,554; 4 %)

Japan (13,886; 10 %)

USA (12,497; 9 %)

19

PER CAPITA INSTALLED SOLAR POWER CAPACITY (W )

Source: EPIA Global Market Outlook for Photovoltaics until 2016, http://www.epia.org/publications/epiapublications/globalmarketoutlookforphotovoltaicsuntil2016.html

0

50

100

150

200

250

300

Germany Italy CzechRepublic

Belgium Spain Japan U.S.

W

20

SOLAR ENERGY DENSITY IN THE US, SPAIN AND GERMANY

Sour

ce: S

EIA

-US

Sola

r Ind

ustr

y Yea

r in

Rev

iew,

200

9

21

2010 – 2014 GLOBAL CUMULATIVE CSP CAPACITY BY QUARTER

Sour

ce: h

ttp://

ww

w.n

rel.g

ov/c

sp/s

olar

pace

s/

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

2010 2011 2012 2013 2014year

GW

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q42011 2012 20132010 2014

22

PV SOLAR CAPACITY PRICE DECLINE

80

0

60

40

20

1977 2013200719891983 1995 2001

$ / w

att

0.74

Source: Bloomberg, New Energy Finance, http://www.smartgridnews.com/artman/publish/Technologies_DG_Renewables/Goldman-agrees----solar-will-soon-dominate-electric-power-6465.html#.U2f86_ldUsd

23

WIND ENERGY

24

47,620 59,09173,938

93,889120,624

158,975

198,001

238,126

283,194

318,105

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

thou

sand

s of M

W1997 – 2014 GLOBAL WIND CAPACITY

Source: http://www.gwec.net/global-figures/graphs/

26

Europe36.26 %

North America21.14 %

Asia38.4 %

Pacific 1.2 %

Africa & Middle East 0.70 %

Latin America & Caribbean2.3 %

total installed capacity: 369, 597 MW

2014 INSTALLED WIND CAPACITY

Sour

ce: G

WE

C A

nnua

l Rep

ort 2

015,

htt

p://w

ww

.gw

ec.n

et/g

loba

l-fig

ures

/gra

phs/

27

2014 CUMULATIVE WIND CAPACITY: THE TOP 10 COUNTRIES

114,609MW

65,879 MW

39,165 MW

22,987 MW

22,465 MW

12,440 MW

9,694 MW

9,285 MW

8,663 MW

5,939 MW

0 20000 40000 60000 80000 100000 120000

China

USA

Germany

Spain

India

United Kingdom

Canada

France

Italy

Brazil

Source: http://www.gwec.net/wp-content/uploads/2012/06/Top-10-cumulative-capacity-Dec-2014.jpg

28

2014 WIND CAPACITY ADDITION AND CUMULATIVE TOTAL

added capacity MW

China 23,300Germany 5,119

USA 4,854Brazil 2,783India 2,315UK 1,467France 1,042Turkey 804World total 51,230

cumulative capacity MW

China 114,760USA 65,877

Germany 39,223India 22,904Spain 22,665UK 12,413Canada 9,684France 9,170World total 372,112

Source: http://energy.gov/sites/prod/files/2015/08/f25/2014-Wind-Technologies-Market-Report-8.7.pdf

29

2014 INSTALLED WIND CAPACITY IN REGIONS IN CHINA

MW

3,630

3,216

2,081

1,694 1,5901,373

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

Gansu Xinjiang InnerMongolia

Ningxia Shanxi Hebei

Source: GWEC Annual Report 2014, http://www.gwec.net/wp-content/uploads/2015/03/GWEC_Global_Wind_2014_Report_LR.pdf

30

US CUMMULATIVE WIND CAPACITY

Source: http://energy.gov/sites/prod/files/2015/08/f25/2014-Wind-Technologies-Market-Report-8.7.pdf

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

MW

31

US – GERMANY COMPARISON

attribute US Germany ratio

population (million) 321 82 3.9

area (mi2 ) 3, 119, 884 137, 882 22.6

peak load (GW ) 777 80 9.7

annual energy (billion kWh) 3, 963 544 7.3

installed wind capacity (MW ) 65, 877 39,223 1.7

32

US WIND SITE CHARACTERIZATION

Source: http://www.nrel.gov/wind/systemsintegration/images/home_usmap.jpg

33

2014 Q4 WIND ENERGY STATUS

Source: AWEA, Fourth Quarter 2014 Market Report, p. 5

34

The US installed over 4,854 MW of wind power

capacity in 2014

The total installed wind power capacity in the US

has approached 66,000 MW by the end of 2014

Wind power represented 24 % of US electric-

generating capacity additions in 2014

2014 WIND ENERGY STATUS

Source: U.S. Department of Energy 2014 Wind Technologies Market Report

35

RENEWABLE PORTFOLIO STANDARD (RPS) STATUS IN 2015

Source: www.dsireusa.org , June 2015

WA: 15% x 2020*

OR: 25%x 2025* (large utilities)

CA: 33% x 2020

MT: 15% x 2015

NV: 25% x2025* UT: 20% x

2025*†

AZ: 15% x 2025*

ND: 10% x 2015

NM: 20%x 2020 (IOUs)

HI: 100% x 2045

CO: 30% by 2020 (IOUs) *†

OK: 15% x 2015

MN:26.5% x 2025 (IOUs)

31.5% x 2020 (Xcel)

MI: 10% x 2015*†WI: 10%

2015

MO:15% x 2021

IA: 105 MW IN:10% x 2025†

IL: 25% x 2026

OH: 12.5% x 2026

NC: 12.5% x 2021 (IOUs)

VA: 15% x 2025†KS: 20% x 2020

29 states + Washington DC + 3 territories have a RPS (8 states and 1 territory have renewable portfolio goals)

Renewable portfolio standard

Renewable portfolio goal Includes non-renewable alternative resources* Extra credit for solar or customer-sited renewables†

U.S. Territories

DC

TX: 5,880 MW x 2015*

SD: 10% x 2015

SC: 2% 2021

NMI: 20% x 2016

PR: 20% x 2035

Guam: 25% x 2035

USVI: 30% x 2025

NY: 29% x 2015

NH: 24.8 x 2025VT: 75% x 2032

MA: 15% x 2020(new resources) 6.03% x 2016 (existing resources)

RI: 14.5% x 2019CT: 27% x 2020

PA: 18% x 2021†

NJ: 20.38% RE x 2020 + 4.1% solar by 2027

DE: 25% x 2026*

MD: 20% x 2022DC: 20% x 2020

ME: 40% x 2017

36

DOE FUNDING TARGETS FOR ARRAPROJECTS

$ 16.8 billion

$ 4.5

$ 6.0

$ 3.4$ 6.6

$ 8.1

loan guarantees

climate R&DWAPA/BPA

other

Smart Grid

energy efficiency/renewable energy

(incl. transportation)total: $ 45 billion

37

ARRA SMART GRID FUNDING

$3.4 billion

0.4850.615

Smart Grid investment grants

Smart Grid demonstration

projects other

38

2008 GREEN PORTIONS OF STIMULUS%

of G

DP

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

China U.S.Source: Center for American Progress analysis of HSBC research

39

GRAND CHALLENGES/OPPORTUNITIESqu

antif

icat

ion/

tool

s forecasting

dynamic simulation

controllable resource impacts

economic/reliability impacts

environmental impacts

asse

ssm

ent

data analytics

cyber security

transmission expansion

challenges/opportunities

40

SALIENT CHARACTERISTICS OF SOLAR AND WIND POWER

Highly time-dependent nature of solar and wind power: uncertain variability characteristics intermittency effects

Solar power is dependent upon both solar irradia-tion and air temperature

Unlike conventional generation sources that have the ability to store the primary energy prior to conversion into electricity, wind and solar energy resources are unable to store their primary energy

41

SALIENT CHARACTERISTICS OF WIND AND SOLAR POWER

Both solar radiation and temperature levels and

wind speeds are subject to a broad range of

sources of uncertainty of both an atmospheric

and a geographic nature

Such uncertainty is, typically, very difficult to

analytically characterize; it also exhibits highly

seasonal dependence

42

PV OUTPUT AND LOAD

Sources: ERCOT and NREL

20

30

40

50

60

load

(GW

)

24 48 72 96 120 144 1680

100

200

300

PVou

tput

(MW

)

time (h)

43

PV POWER OUTPUT AT THE NEVADA 70 kW POLYCRYSTALLINE ARRAY

data collected on a 10 – second basis

kW

80

70

60

50

40

30

20

10

0

44

CAISO DAILY WIND POWER PATTERNS IN MARCH 2005

Source: CAISO

MW700

600

500

400

300

200

100

0

hour

45

ONTARIO DAILY WIND POWER OUTPUT

MW700

600

500

400

300

200

100

0

hour

46

SALIENT CHARACTERISTICS OF WIND AND SOLAR POWER

Wind power production has limited controllability

and dispatchability, thereby imposing additional

burden on conventional controllable resources

The predictability of wind speed is rather proble-

matic and has limited accuracy; consequently, so

is the predictability of wind power outputs

47

MISALIGNMENT OF WIND POWER OUTPUT AND LOAD

0

50

100

150

200

250

win

d po

wer

out

put (

MW

)

24 48 72 96 120 144 168 3000

4000

5000

6000

7000

8000

load

(MW

)

hour

wind power

load

48

quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security


GRAND CHALLENGES/OPPORTUNITIES


49

ERCOT FEBRUARY 26, 2008RELIABILITY EVENT CONTRIBUTORS

Source: NREL, http://www.nrel.gov/docs/fy08osti/43373.pdf

load

actual wind output

a faster than expected

evening load up-ramp

a large down-ramp of

wind generation which

started at 15:00

the unexpected loss

of conventional

generation

50

FORECASTING NEEDS More accurate forecasts are needed over shorter

and longer-term forecasting periods explicitly taking into account the meteorological sources of uncertainty

Forecasting accuracy decreases as the look-ahead time increases

Tools to forecast ramps and renewable energy generator output over various time intervals are critical to ensure the reliable and the economic integration of renewable energy

51

quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security




52

0

2

4

6

8

10

12

14

0

5

10

15

20

25

30

win

d sp

eed

(m/s

)at 1

00m

hei

ght

win

d ou

tput

(MW

)10-MINUTE WIND SPEED/OUTPUT

DATA FOR 1/2/2006 AT LUBBOCK, TX

Source: NREL, http://www.nrel.gov/wind/integrationdatasets/western/data.html

wind output

wind speed

53

CHALLENGE: DYNAMIC SIMULATION TOOLS

Renewable energy sources may experience wide

and sudden changes during their operations

System planners and operators need appropriate

tools to gain better understanding of, and ability

to deal with, the intermittency and variability

impacts on the dynamic performance of power

systems over time periods as long as 15 minutes

54

These needs encompass the development of

models appropriate for the representation of the

dynamic behavior of renewable resources and

their interactions with conventional units and

control elements and their incorporation into

existing dynamic simulation tools

Numerical tractability is a key consideration in

view of the longer period of simulation

CHALLENGE: DYNAMIC SIMULATION TOOLS

55

DYNAMIC SIMULATION TIME SCALESelectromagnetic transients

machine-networkinteraction

first swing transient stability

10-6 10-5 10-4 10-3 10-2 0.1 1 10

time, sec

induction motor dynamics

generator/excitation dynamics

prime mover control

mechanically switched capacitors/reactors

SVC

generator inertial dynamics

DC

protective relaying

Adapted from: C. Taylor, Voltage Stability Analysis

56

load/power transfer increase

line/transformer overload

10 100 1,000 10,000time, sec

gas turbine start-up

boiler dynamics

OLTC transformer & distribution voltage regulator

powerplant operator actions

generation change/AGC

excitation limiting

system operator actions

prime mover control

load diversity thermostat control

protective relaying including overload protection

DC converter OLTC

DYNAMIC SIMULATION TIME SCALES

Adapted from: C. Taylor, Voltage Stability Analysis

57

MODELING NEEDS FOR VERDYNAMIC STUDIES

Under-load tap changers

Switchable shunts

Reactive power compensation devices

Load dynamics, including induction motors

Peaking units

Utility–scale energy storage units

Protection/control system coordination

58

quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security




59

The deepening penetration of variable energy

resources (VERs), such as wind and solar

resources, reduces the need for energy from

controllable resources

However, the highly variable and uncertain VER

power outputs, which may be misaligned with the

load, result in steeper ramping requirements in

the net load – the difference between the system

load and the total VER generation output

VER INTEGRATION CHALLENGES

60

The integration of deepening penetration of VERs

results in large ramp magnitude, steep ramp rate

and long continuous ramp duration requirements

in the net load, which are considerably more

marked than those in the system load itself

In addition, deepening VER penetration exacer-

bates the number of up and down ramping events


61

day

frac

tion

of p

eak

load

1.0

0.5

0

load

net load

steep ramp up required

Source: M. Lange & U. Focken, “Physical Approach to Short-Term Wind Power Prediction”, Springer, 2006

steep ramp down required

DEEPER WIND PENETRATIONSTRONGLY IMPACTS THE NET LOAD

62Source: http://www.caiso.com/Documents/FlexibleResourcesHelpRenewables_FastFacts.pdf, issued October 2013

CAISO DAILY NET LOAD CURVE UNDER DEEPER PENETRATION

26,000

net load (MW)

time (hour)

0

22,000

18,000

14,000

10,000

0:00 4:00 8:00 12:00 16:00 20:00 24:00

63

0

2,000

4,000

6,000

8,000

14,000

20,000

26,000

32,000

0:00 4:00 8:00 12:00 16:00 20:00

net load = load - wind - solar

Source: Mark Rothleder, Clyde Loutan, Renewable Energy Integration, Chapter 6 Case Study–Renewable Integration. Elsevier, 2014.

solargeneration (MW)load (MW)

loadnet load

wind

time

STEEP RAMPING REQUIREMENTS IN THE 2020 CAISO NET LOAD CURVE

6,700 MW over 3 hours



64

The steeper ramping events occur at various time

scales from seconds to hours

The rapid net load changes that occur on a time

scale from seconds to minutes markedly increase

the requirements for load following, regulation

and reserves services, which are generally

provided by the controllable resources and rely

heavily on the units with large ramping capabilities


65

CAPACITY–BASED ANCILLARYSERVICES

response time in minutes30100

regulation

supplemental/non-spinning reserves

load following; spinning reserves

120

66

At deeper VER penetrations, while the power gridis less dependent on the controllable resourcesfor energy production, operators must rely moreintensely on the controllable resources tomanage the steeper ramping requirements in thenet load, due principally to the VER outputvariability and intermittency

The controllable resources must have adequateramping capability to compensate the rapidchanges in VER outputs so that supply–demandbalance is kept around the clock

REQUIREMENTS ON THE CONTROLLABLE RESOURCES

To be more specific, the controllable resources must have steep ramping rates; longer ramping durations; rapid direction change ability in both the up

and the down directions ability to respond rapidly to a steep increase or

decrease in the output level request; and ability to perform start-ups and shutdowns

multiple times per day

REQUIREMENTS ON THE CONTROLLABLE RESOURCES

CRITICALITY OF RAMPING

In the absence of accurate forecasts for VER

output, operators mitigate the uncertainty of a

sudden rise/drop in VER outputs by requiring the

committed controllable units, including storage,

to maintain adequate ramping up/down capabilities

The incorporation of such ramping requirements

into the resource dispatch results in newly added

constraints and incurs additional expenses

69

OPPORTUNITY COSTS

Conventional generators incur opportunity costs in

the provision of ramp capability services

Rather than generate at as high a capacity as each

conventional unit can clear for an entire hour to

sell its energy at the market clearing price, a seller

may need to reserve a portion of the capacity to

be used for ramping capability service provision

The foregone profits from the energy not sold are

the opportunity costs of a conventional unit

70

CYCLING COSTS

In addition to the opportunity costs, the extensive

cycling of the generation units that provide ramp

capability service causes considerable wear and

tear, whose impacts may include additional

emissions and increased maintenance costs

71

CYCLING EFFECTS

age in years

3025

20

15

10

5

05 10 15 20 25 30 35 40

equi

vale

nt fo

rced

outa

ge ra

te (%

)

45

cycling begins

cycling operation

cycling related generation loss

base-loaded operation

reduced plant life

Adapted from: Intertek APTECH, “Power Plant Asset Management”

72

CO2 EMISSION INCREASES FOR NON-FULLY LOADED OPERATIONS (%)

technology startup partialloading ramping

coal 110 5.1 0.4

gas CC 32 15.6 0.3

gas CT 32 14.4 0.3

Source: NREL, “Emission Impact of Fossil Fuel Unit Cycling”

73

CYCLING EFFECTS AND RESEARCH NEEDS

incr

ease

d du

ty

cycl

e

cost quantification ?

cost allocation policy ?

availability impact

assessment ?

emission impact

evaluation ?

maintenance

capital spending

power costs due to forced outages

startup fuel

labor

heat rate

74

CONTROLLABLE UNIT IMPACT CHALLENGES

The quantification of the additional controllable

unit costs, maintenance expenses, reduced

lifetime and emission increases that result from

renewable energy–induced ramping requirements

is a key need

The development of cost allocation mechanisms

for the additional controllable unit costs needs to

be addressed

75

CONTROLLABLE UNIT IMPACT CHALLENGES

The market for ramp capability design needs to incentivize investment in flexible resources

including storage through payments for the capacity provided;

incentivize provision of ramping capability services in the operations; and

incorporate transparency in the compensation of the conventional generation units to ensure reimbursement of opportunity costs for the ramping capability service provision

76

RESOURCE SCHEDULING TIME SCALES

schedulingfunction typical objective

typical time

horizonbasic time

unit

maintenance levelize reserves year month or week

hydro usage

maximize the value of the hydro generation to replace the energy

generated by high-priced fuel resources

year month or week

unitcommitment minimize total costs week or

day hour

economicdispatch

minimize production costs including system losses day minutes

77

SCHEDULING AND MARKET DESIGN NEEDS

There is a need for the development of resource

scheduling tools that explicitly represent the

uncertain renewable resource outputs

In addition, there is a need for new market design,

such as sub-hourly markets, which explicitly

consider the VER uncertain and intermittent nature

78

PV POWER OUTPUT OF 1 - MW CdTeARRAY IN GERMANY

kW

1000

0

samples collected on a 5 – minute basis

900

800

700600

500400

300

200

100

79

SURPLUS BASE-LOADED GENERATION

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

MW

Source: IESO

range of minimum weekly demand

base-loaded generation

average weekly minimum demand

80

STOCHASTIC SCHEDULING TOOLS

The additional sources of uncertainty associated

with renewable resources necessitate the

deployment of stochastic methods for unit

commitment scheduling

Such methods need to take into account the

sources of uncertainty in resource variability/

intermittency, correlation with load variability and

correlation among the renewable resources

81

STOCHASTIC SCHEDULING TOOLS

The development of stochastic and/or robust

optimization methods for the solution of unit

commitment problems for large-scale systems is

a high priority

Key considerations are the issue of

computational tractability and the investigation of

the applicability of parallel computing

82

ADDITIONAL RESOURCE DEPLOYMENT

Storage and demand response resources (DRRs)

provide the system operator additional flexibility

in reliably and economically managing the supply–

demand balance

Storage and DRRs may also provide system

services such as regulation, load following and

reserves

83

LACK OF STORAGE CAPACITY

The lack of utility–scale storage in today’s power system drives electricity production to be the prototypical

just-in-time manufacturing system; and electricity to be a highly perishable commodity

Storage capacity is primarily in the pumped storage plants; there are only two operational CAESunits in the world

Pace of energy storage development has been very slow in the past

84

STORAGE TECHNOLOGIESdi

scha

rge

time

(h)

rated power (MW)

85

STORAGE POLICY ISSUES

The regulatory treatment of storage units is an

open question and needs to be addressed by

policy analysts and regulators

The regulatory framework will guide the operation

and control of storage issues and set the stage to

ensure storage resource profitability

The treatment of short–term storage may differ

from that of longer–term energy arbitrage–based

storage

86

DRRs

DRRs are demand–side entities which actively

participate in the markets as both buyers of

electricity and sellers of load curtailment services

DRRs effectively reduce the load during peak

hours and/or shift the demand, in part or in whole,

from peak hours to low–load hours

87

DEMAND RESPONSE RESOURCES (DRRs)

market clearing transmission scheduling

resourcespassive loadsDRRs

88

supply-side payments

$/M

Wh

MWh/h

ρ′new market clearing price

ρ

shift in demand

curve

suppose a demand-side bid gets accepted to curtail MW

demand to be met =

MARKET WITH DRRs

89

SYMBIOSIS OF RENEWABLE RESOURCES WITH STORAGE/DRRs

The flexibility of storage and DRRs may be

deployed to take advantage of the renewable

output whenever available

When effectively deployed, storage and DRRs

reduce the operational costs and emissions,

arising from increased unit cycling, by displacing

controllable units

90

0 2 4 6 8 10 12 14 16 18 20 22 2411,000

12,000

13,000

14,000

15,000

16,000

17,000

CYCLING UNITS WITHOUT DRR

Source: ISO-NE

dem

and

(M

W)

time of day

start-up the cycling units

peaking units

cycling units

91

0 2 4 6 8 10 12 14 16 18 20 22 2411,000

12,000

13,000

14,000

15,000

16,000

17,000

CYCLING UNITS WITH DRR

Source: ISO-NE

dem

and

(MW

)

start-up the cycling units

delay

time of day

peaking units

cycling units

DRR

impact

92

quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security




93

QUANTIFICATION OF INTEGRATION IMPACTS

The most effective approach to determine the

economic, reliability and environmental impacts

of the integration of renewable resources over

longer–term periods is probabilistic simulation

The appropriate representation of resources such

as wind, solar, demand response and storage, in

the simulation of the power system over longer–

term periods requires the explicit consideration of

the:

94

QUANTIFICATION OF INTEGRATION IMPACTS

interactions of the existing resource mix and the transmission grid with the integrated units

the sources of uncertainty associated with the integration of the time–dependent resources, together with those typically represented in conventional probabilistic simulation

the time–dependent transmission–constrainedmarket outcomes

There is a need for a computationally tractable simulation schemes

95

TYPICAL APPLICATIONS

Resource planning studies

Production costing issues

Transmission utilization issues

Environmental assessments

Reliability analysis

Investment analysis

96

quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security




97

PHASOR MEASUREMENT UNITS (PMUs) GPS–based monitoring devices sample voltages

and currents 30–60 times per second and serve as data acquisition tools

PMUs require analytics to provide the capability to store, manage, retrieve and protect the huge sets of data collected and information generated

In addition, PMUs require new tools for the reliable and economic operation of the electrical grid, as well as applications software for visualization and analysis

98

PHASOR MEASUREMENT UNITS IN NORTH AMERICAN POWER GRID

Source: NASPI, PMU status, updated March 9, 2015: https://www.naspi.org/sitesearch#searchResultSection

99

PMU vs. SCADA DATA: APRIL 5, 2011 EVENT

Source: EIA ,Match 30, 2012, http://205.254.135.7/todayinenergy/detail.cfm?id=5630

.991.001.011.021.031.041.051.061.07

0

8:22 8:29 8:36 8:43 8:50 8:58

time

volta

ge in

p.u

. SCADA

PMU

100

PMU vs. SCADA DATA: FEBRUARY 7, 2010 EVENT

Sour

ce:

NE

RC

, Rep

ort o

n R

eal-T

ime

App

licat

ions

of S

ynch

roph

asor

sfor

Impr

ovin

gR

elia

bilit

y ,O

ctob

er 1

7, 2

010

101

SYNCHROPHASOR MEASUREMENT APPLICATIONS

The PMU installations are opening up a broad

range of useful applications in real–time opera-

tions, control and analysis, as well as planning

The PMU measurements provide a rich trove of

data and create huge challenges in data storage,

processing, security and management/utilization

The advances in addressing these challenges will

facilitate the smoother integration of renewables

102

BIG DATA ANALYTICS CHALLENGES

PMUs are simply one source of data collection, but throughout the power system there are myriad sensors that collect massive volumes of data and create huge data silos in various formats

The performance of analytics on such huge data sets may be impossible on a central processor; distributed processing on parallelized multi-pro-cessors may be an effective alternative

As various sources continually generate data in real time, analytics may frequently be carried out heuristically and without an opportunity to revisit past entries of generated data

103

BIG DATA ANALYTICS CHALLENGES

In light of their disparate nature and measurement, massive datasets are often incomplete, prone to outliers, noisy and vulnerable to cyber-attacks

Techniques are needed to solve vast data storage, analysis, management and distribution issues, as well as the schemes to handle the practical issues: mega data analytics

The collected data and generated information can be used, when integrated with domain expertise, to address identified gaps in today’s planning, operations and analysis tools

104

CHALLENGES IN PMU APPLICATIONS

Power system operations

VER integration

islanding schemes

missing data

replacement

remedial action

schemes

special protection

schemes

power oscillation

detection

power system

restoration

protection and control

schemes for

distributed generation

resources

105


Real–time system operations

situational awareness

visualization

transient stability

margin monitoring

voltage stability

operation margin

monitoring

wide–area monitoring

frequency stability

monitoring

static estimation and

PMU data

dynamic line ratings

congestion management

106


Power system planning applications include

event analysis

static model

validation

dynamic model

validation

new metric formulation

new tool development

inter–area oscillation

damping control

disturbance classifier

107

quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security




108

CYBER SECURITY

Cyber security is essential in the development of the Smart Grid to protect customer information and the physical grid from cyber vulnerability

There is a critical need to understand the potential grid vulnerabilities and the impacts of cyber attacks on the grid

There is also a need to develop infrastructure and policies which shield the grid from, and reduce the impacts of, cyber attacks

109

CYBER SECURITY CONCERNS

An analysis by USA Today of the reports

submitted by federal officials and contractors

since late 2010 to the Department of Energy's Joint

Cybersecurity Coordination Center shows a stream of

attempts to breach the security of critical

information systems that contain sensitive data

about the nation's power grid, nuclear weapons

stockpile and energy labsSource: http://www.usatoday.com/story/news/2015/09/09/cyber-attacks-doe-energy/71929786/

110


These reports indicate DoE components reported

a total of 1,131 cyber attacks over a 4-year period

ending in October 2014, with 159 of them successful

The US Congress Committee on Science, Space and

Technology noted “As the electric grid continues to be

modernized and become more interconnected, the threat

of a potential cybersecurity breach significantly increases"

111


Cyber security is a top priority for the electricity

industry, given the need for compliance with

NERC standards and the utility obligation to serve

There is a serious need for a “seal of approval” to

ensure compliance with cyber security standards

Effective coordination of stakeholders – NERC,

NIST, FERC, states and industry – can address

cyber security vulnerabilities and threats by

judiciously harnessing the smart grid capabilities

112



quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security


113

“ANTIQUATED” TRANSMISSION SYSTEM

The existing power grids in the US incorporate, by and large, technology developed before the 1950s

About 70 % of all transmission lines and power transformers are 25 years old or older, and 60 % of all circuit breakers are more than 30 years old

The small number of new transmission projects implies that the many technology advancements and the smart grid implementation have not been deployed by the legacy grid

114

DEMAND AND TRANSMISSION CAPACITY GROWTH

0

5

10

15

20

25

30

1988 – 98 1999 – 09

electricitydemand

transmissioncapacityexpansion

%

Source: EPRI

115

CHARACTERISTICS OF TODAY’S GRID The grid was planned and built by utilities to

connect the distant generation sources to the load

centers so as to meet their native customers’

demands reliably and cost effectively

As the grid was designed for different purposes,

its adequacy for its new role in the market environ-

ment requires its expansion and modernization

116

broker /marketer

broker /marketer

“COMMON CARRIER” TRANSMISSION

utilitygeneration

selfgeneration

QF

IPP

EWG

transmissionsystem

117

CARTOONIST’S VIEW

118

CONGESTION

119

TRANSMISSION IN THE COMPETITIVE ENVIRONMENT

Congestion situations result whenever the

demand for transmission services exceeds the

capability the grid has to provide; such situations

occur frequently in electricity markets

Despite the more intense utilization of the grid by

the many established and new players, develop-

ments in transmission expansion have failed to

keep pace with the increasing demand

120

CONGESTION RENTS

0

500

1,000

1,500

2,000

2,500

2006 2007 2008 2009 2010

mill

iion

$

PJM

MISO

121

TRANSMISSION EXPANSION: KEY DRIVERS


deepening penetration

of renewables

environmentalconcerns

congestion

reliability

smart grid technologies

122

U.S. POPULATION DENSITY AND RENEWABLE RESOURCE LOCATIONS

wind

solar

wind

Source: http://www.census.gov/popest/data/maps/2009/PopDensity_09.jpg

123

U.S. ELECTRIC POWER GRID

Source: FEMA

wind

solar

125

INTERCONNECTION QUEUES

0

20

40

60

80

100

120

140

160

Natural Gas Wind Solar Nuclear Coal Other

GW

total wind in queue at the end of 2014

solar capacity that entered queue in 2014

wind represents about 30%of all electricity capacity within these queues in 2014

Sour

ce: h

ttp://

ener

gy.g

ov/s

ites/

prod

/file

s/20

15/0

8/f2

5/20

14-W

ind-

Tech

nolo

gies

-Mar

ket-R

epor

t-8.

7.pd

f

127

TRANSMISSION EXPANSION: THE AWEA VISION

Large-scale integration of renewable energy will

require substantial transmission expansion

Cost allocation mechanisms are needed to

incentivize transmission expansion

There is a need to develop appropriate planning

tools, which take explicit account of renewable

energy uncertainty

128

POLICY CHALLENGES Policy considerations are important for all

aspects of grid planning and operations

Sound technical knowledge is needed to

formulate good policy and provides the basis for

creating the blueprint for policy making

Policy directives are explicitly considered in the

analysis

129

EMERGING AND STANDING RELIABILITY ISSUES, 1-5 TO 6-10 YEARS

likel

ihoo

d

impactslow high

high

energy storage

cybersecurity

transfer capability

variable generation

workforce issues

GHGlegislation

smart grid & AMI

reactive power

Sour

ce: N

ER

C, h

ttp://

ww

w.n

erc.

com

/file

s/20

09_L

TRA

.pdf

130



quan

tific

atio

n/to

ols forecasting

dynamic simulation




asse

ssm

ent

data analytics

cyber security


131

CONCLUDING REMARKS

The growing worldwide interest in renewable

resource integration presents rich opportunities

and challenges for system operators, planners

and researchers

The effective solutions to address these challen-

ges/opportunities will bring about sustainable

paths to meet the world’s future energy needs