SENCO

INTEGRATED SYSTEM MODELLING

January 2005

Mark Barrett

Coping with variability: integrating renewables into the electricity system.

A one day conference at the Open University

24th January 2006

MarkBarrett@sencouk.co.uk SENCO Sustainable Environment Consultantswww.sencouk.co.uk

SENCO

2

Outline of talk

Introduction

Scenario context

– Demand

– Electricity supply context

Energy, space and time

– problem illustrated

– load management

– demand and renewable supply correlation

– annual sample day simulation model

– optimisation

– beyond the UK

Supply security

A comprehensive UK energy scenario (under development) is to be found here:

www.sencouk.co.uk/Energy/UKEnergy/

SENCO

3

The energy system: demand and supply options

PRIMARY ENVIRONMENTAL IMPACT

ENERGY DEMAND ENERGY CONVERSION PRIMARY ENERGY

Finite energyFossil

Income energy

Finite energy

Fission

Geothermal

Sun

Moon

GasOil

Coal

Heat

Kinetic

Heat

Electricity

Kinetic

Light

Chemical

Cool

Generator

Waste heat

Heat pump

Chemical

Heater

Nuclear

Transport

Waste heat

ENVIRONMENTAL HEAT

Tidal

Wave

Wind

HydroHeat

Biomass

LightDomestic

Services

Industry

Heat engine

Impact Impact

Fusion

Motor

Fuel cell E

Energy demands and sources interact and can be combined in limitless ways.

SENCO

4

Integrated planning

Energy planning should be integrated across all segments of demand and supply. If this is not done, the system may be technically dysfunctional or economically suboptimal. Energy supply requirements are dependent on the sizes and variations in demands, and this depends on future social patterns and demand management. For example:

• In 2040, what will electricity demand be at 4 am? If it is small, how will it affect the economics of supply options with large inflexible units, such as nuclear power?

• The output from CHP plants depends on how much heat they provide, so the contribution of micro-CHP in houses to electricity supply depends on the levels of insulation in dwellings.

• Solar collection systems produce most energy at noon, and in the summer. The greater the capacity of these systems, the greater the need for flexible back-up supplies and storage for when solar input is low.

• The scope for electric vehicles depends on demand details such as average trip length. Electric vehicles will add to electricity demand, but they reduce the need for scarce liquid fuels and add to electricity storage capacity which aids renewable integration.

• Electricity supply systems with a large renewable component require flexible demand management, storage, electricity trade and back-up generation; large coal or nuclear stations do not fit well into such systems because their output cannot easily be varied over short time periods.

• The amount of liquid biofuels that might available for air transport depends on how much biomass can be supplied, and demands on it for other uses, such as road transport.

• Is it better to burn biomass in CHP plants and produce electricity for electric vehicles, or inefficiently convert it to biofuels for use in conventional engines?

SENCO

5

Modelling the energy demand and supply system: the challenge

The energy demand and supply system needs to be analysed on different temporal and spatial scales. This system is globally interconnected, but some connections are strong (e.g. intra-UK electricity system) while others are weak (UK gas and African oil demand)

Time scales

Demands and renewable supplies vary with social activity patterns and the weather. There can be significant variations over seconds, minutes, hours, days, day of week, months and years.

Spatial scales

Demands and renewable supplies vary spatially: from within buildings, to local, regional, national and international levels.

The modelling challenge

No model can represent details of the whole system at all temporal and spatial levels. The art is to achieve a balance so that all important processes and energy policy options are covered.

Aspects that need further exploration include:

– the context of the future energy scenario

– future demands

– storage and non-renewable supply

– electricity trade

SENCO

6

Electricity system design with variable sources : elements 1

Electricity demands

• energy services - type, temperature, quantity, time, weather dependency

– non-heat (motive, lighting, equipment) and storable heating/cooling

– interruptible load

• end use technologies; control, storage, interruptibility

– space and water heaters, fridges, freezers, dishwashers, clothes washers

• multi-fuelled energy services

– electric/ solar / gas heating

– electric / liquid fuelled hybrid cars

Energy storage

• Parameters

– energy form input, output and stored (electricity, heat, ..)

– storage capacity (J)

– input and output power (W)

• Type and system location

– end use storage

• embedded in end use technologies (building thermal mass, refrigerators)

• purpose built (hot water tank, solid heat storage heater)

– local /neighbourhood storage

• garage for refuelling electric vehicles

– system storage

• centralised pumped storage

SENCO

7

Electricity system design with variable sources : elements 2

Generation

• Renewable

– resource intensity variation in space and time of hydro, wind, wave, solar, tidal, biomass

– technology output variation a function of resource variation and design

• minimum/ maximum resource intensity for operation

• efficiency variation with resource intensity

– manipulation of technology output variation

• spatial distribution of wind, wave, solar, tidal energy collectors

• technology design; solar collection orientation, tidal scheme configuration

• CHP

– heat load size and temporal variation

– CHP technology; variable heat: electricity ratio

• Back-up; biomass, fossil, nuclear

Regional and international linkage

System operation

– prediction in time and space

• demand

• supply

– system control

• end use

• supply

SENCO

8

Domestic sector: house heat loss factors

Implementation of space heat demand management (insulation, ventilation control) depends on housing needs and stock types, replacement rates, applicability of technologies

0

50

100

150

200

250

300

350

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

W/o

C

Vent loss

Roof

Window

Wall opaque

Floor

GBR: TechBeh: W/oC : Elements

SENCO

9

House: monthly space heating and cooling loads

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

1 2 3 4 5 6 7 8 9 10 11 12

GJ/

mo

nth

0

5

10

15

20

25

30

35

40

oC

Gross

Incidental gain

Solar

Heat

Cool

Ambient temperature

Equilibrium temperature, noheating/cooling

Thermostat temperature

United Kingdom 2005 : TechLifestyle Scenario : House temperatures and heat flows

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1 2 3 4 5 6 7 8 9 10 11 12

GJ/

mo

nth

0

5

10

15

20

25

30

35

40

oC

Gross

Incidental gain

Solar

Heat

Cool

Ambient temperature

Equilibrium temperature, noheating/cooling

Thermostat temperature

United Kingdom 2050 : TechLifestyle Scenario : House temperatures and heat flows

• Future space heating need greatly reduced

• Potential growth in air conditioning depends on detailed house design and temperature control

• Less seasonal variation in demand

SENCO

10

Domestic sector: electricity use

Electricity demand is reduced because of more efficient appliances, including heat pumps for space heating.

0

50

100

150

200

250

300

350

400

450

500

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

PJ

AirCon_EH

HeatOff_EH

Heater_EH

Heater_EH

Cooker_EH

CWash_EW

Freezer_EH

Refrig_EH

Refrig_EH

DishW_EW

CWash_EW

Light_EL

Equip_E

GBR: TechBeh: Residential : Electricity

SENCO

11

Transport: passenger vehicle distance

A large reduction in road traffic reduces congestion which gives benefits of less energy, pollution and travel time.

0

50

100

150

200

250

300

350

400

450

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

Gv.

km

Int:Pas:Plane_LB

Int:Pas:Plane_K

Int:Pas:Ship_D

Nat:Pas:Ship_D

Nat:Pas:Plane_K

Nat:Pas:Rail_E

Nat:Pas:Rail_LB

Nat:Pas:Rail_D

Nat:Pas:Bus_E

Nat:Pas:Bus_H2

Nat:Pas:Bus_CNG

Nat:Pas:Bus_LB

Nat:Pas:Bus_D

Nat:Pas:Car_E

Nat:Pas:Car_H2

Nat:Pas:Car_LB

Nat:Pas:Car_LPG

Nat:Pas:Car_D

Nat:Pas:Car_G

Nat:Pas:MCyc_G

Nat:Pas:Bike_S

GBR: TechBeh: Passenger : Vehicle distance

SENCO

12

End use sectors: energy delivered by fuel

Reduction in fossil fuel use through efficiency and shift to alternatives.

0

1000

2000

3000

4000

5000

6000

7000

800019

90

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

PJ

H_Solar

S_Bio

L_Bio

G_Bio

S_CHP

L_CHP

G_CHP

H_Pipe (DH)

E_Ele

S_Fos

L_AviKje

L_MotGas

L_GasDie

L_LiqPeG

L_Fos

G_Fos

GBR: TechLifestyle: Delivered : by fuel

SENCO

13

Energy supply: electricity

Options exercised:

• Phase out of nuclear and coal generation

– some fossil (coal, oil, gas) capacity may be retained for security

• Extensive installation of CHP, mainly gas, in all sectors

• Utilisation of biomass waste and biomass crops

• Large scale introduction of renewable electricity

– wind, solar, tidal, wave

Electricity supply in the scenarios requires more analysis of demand and supply technicalities and economics, particularly:

• future technology costs, particularly of solar-electric systems such as photovoltaic

• demand characteristics including load management and storage

• renewable supply mix and integration

SENCO

14

Energy supply: electricity : generating capacity

Capacity increases because renewables (especially solar) and CHP have low capacity factors. Some fossil capacity retained for back-up and security.

0

20

40

60

80

100

120

140

160

180

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

GW

e

S_Fos

L_FueOil

G_Fos

N_Nuc

S_Bio

L_Bio

G_Bio

S_MunRef

E_Hydro

H_Geothe

H_Solar

E_Wave

E_Tide

E_Wind

Pump_E

S_Fos

L_Fos

G_Fos

S_

L_

G_

GBR: TechBeh: Electricity : Capacity : GWe

SENCO

15

Electricity: generation

Finite fuelled electricity-only generation replaced by renewables and CHP. Proportion of fossil back-up generation depends on complex of factors not analysed with SEEScen.

0

200

400

600

800

1000

1200

140019

90

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

PJe

S_Fos

L_FueOil

G_Fos

N_Nuc

S_Bio

L_Bio

G_Bio

S_MunRef

E_Hydro

H_Geothe

H_Solar

E_Wave

E_Tide

E_Wind

Pump_E

S_Fos

L_Fos

G_Fos

S_

L_

G_

GBR: TechBeh: Electricity : Output : PJe

SENCO

16

Energy flow charts

The flow charts show basic flows in 1990 and 2050, and an animation of 1990-2050. These illustrate the relative magnitude of the flows.

Note that the scale of these charts varies.

Observations:

• Energy services:

– space heating decreases

– other demands increase, especially motive power and transport

• Fuel supply

– increase in efficiency (CHP)

– increase in renewable heating, biomass and electricity

– imports of gas and oil are required

– electricity is exported

SENCO

17

UK Energy flow chart: 1990SENCO GBR : TechBeh : Y1990

Trade Extraction Fuel processing Electricity and heat Delivered Sectors Useful energyEnvironment

Waste energy

Trd_E

Trd_N

Ext_G

Ext_S

Ext_L

Solid

Nuclear

Refinery Liq

Solid

Nuclear

L_FueOil

ElOnly

Gas

Solid

Elec

Liq

Biomass Food

Res_G_

Res_S_Res_E_Res_L_

Ser_G_Ser_S_Ser_E_Ser_L_

Ind_G_

Ind_S_Ind_E_

Ind_L_

Oth_G_Oth_L_

Tra(nat) E

Tra(nat) L

Tra(int) L

Mot W

Proc W

H>120C

H<12-C

Water H

Space H

Space ACCool

CO2 CO2

SENCO

18

UK Energy flow chart: Animation 1990 to 2050

SENCO

19

UK Energy flow chart: 2050SENCO GBR : TechBeh : Y2050

Trade Extraction Fuel processing Electricity and heat Delivered Sectors Useful energyEnvironment

Waste energy

Trd_G

Trd_E

Trd_L

Ext_G

Ext_S

Biomass

Solid

Wind

TideWave

Solar

Biowaste

BiomassBiomass

proc

Refinery

S_BioL_Bio

Liq

Wind

TideWaveSolar

Waste

CHPDHFuI

ElOnly

Auto

CHPDH_H

Auto_H

Gas

G_CHP

H_Solar

Solid

Elec

Heat

L_CHP

Liq

Biomass Food

Res_G_CHPRes_H_Solar

Res_E_

Ser_G_CHPSer_H_SolarSer_E_

Ind_G_Ind_G_CHP

Ind_H_SolarInd_S_Ind_E_

Ind_L_Ind_L_CHP

Oth_G_

Tra(nat) ETra(nat) L

Tra(int) L

Mot W

El equip

Proc W

Light

H>120C

H<12-C

Cooking

Water H

Space H

Space AC

Cool

CO2

SENCO

20

Energy systems in space and time

SEEScen has a main focus on annual flows, although it can simulate seasonal and hourly flows.

Other models are required to analyse issues arising with short term variations in demand and supply, and with the spatial location of demands and supplies.

Questions arising:• Can energy service demands be met hour by hour?• What spatial issues might arise? Increasing the geographical range of electricity systems

increases the temporal diversity of demand and supply.

• What is the best balance between :– local supply and long distance transmission?– demand management, variable supply, optional or back up generation and system or

local storage?

These questions can be posed for different time scales (hour by hour, by day of week, seasonal) and spatial scales (community, national, international).

The EleServe, EST and InterTrade models have been used to illustrate the issues and indicate possible solutions for integrating spatially separate energy demands and sources, each with different temporal characteristics.

SENCO

21

UK energy, space and time illustrated with EST

SENCO

22

UK energy, space and time illustrated with EST : animated

SENCO

23

Technology configuration

What is the best configuration?

Demand Supply

Heat

Solar thermal

Gas

Boiler

Wave

Wind turbine

Light

Wave machin

Solar PV

CHP

Elec

Heat engine

Boiler

TideTide machin

Heat engine

Oil

Coal

Wind

Water

Space

CHP

Applian

Light

Cook

Sun

Heat engine

SENCO

24

Building store CHPWave

22 Acacia

Local storeCoal

23 Acacia

Wind

SystemDemand store

Gas CCConverter

Store

Node CityWind

FRANCEUnit cost

Time varyingDemand

SupplyStorage

Energy, space and time problem

What is the best configuration?

What capacities?

Where to locate converters and stores?

Where to place transmission nodes?

SENCO

25

EleServe : matching demand to supply with load management

The EleServe model has these components:Electricity demand • disaggregated into segments across sectors and end uses• each segment with

– a temporal profile– load management characteristics

Electricity supply• each renewable source with own temporal profile• heat related generation with its own temporal profile• optional thermal generators characterised by energy costs at full and part load, and for starting up• trade and system storage

Operational control• load management: demands are moved if the net cost is of a move is negative, accounting for

differences in marginal supply costs, energy losses and other operational costs• optional units brought on line to minimise diurnal costs

The following graphs demonstrates the role that load management can play in integrating variable renewables and CHP into electricity supply. Heat and electricity storage (hot water tanks, storage heaters, vehicle batteries) can be used to store renewable energy when it is available, so that the energy can later be used when needed.

SENCO

26

Electricity : diurnal operation without load management

SENCO

27

Electricity : animated load management optimisation

SENCO

28

Electricity : diurnal operation with load management

EleServe Scenario: Efficiency + CHP + renewables 2025 Winter day : Summer day SENCO

System

-5

0

5

10

15

20

25

30

1 25

hrs

GW

System demand

Essentialgeneration

Dem Net E

Trade

Dem Net ET

Store

Dem Net ETS

Optionalgeneration

Reserverequirement

Reservestore+hydro

Res req. Net Store

Demand (LM)

0

5

10

15

20

25

30

0 0

hrs

GW

I:Fue:Gen I:Ind:Lig

I:Ind:Mot I:Ind:Pro

O:Far:Gen O:Pub:Lig

O:Tra:Mot S:Com:Ref

S:Com:Spa S:Com:Coo

S:Com:Lig S:Com:Mis

S:Com:Spa S:Com:Wat

S:Pub:Ref S:Pub:Spa

S:Pub:Coo S:Pub:Lig

S:Pub:Mis S:Pub:Spa

S:Pub:Wat R:Fri:Ref

R:Fri:Ref R:Fre:Ref

R:Coo:Coo R:Was:Was

R:Clo:Was R:Dis:Was

R:Tel:App R:Mis:App

R:Lig:Lig R:Hot:Wat

R:Unr:Spa R:Off:Spa

R:Coo:Spa

Marginal costs

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

0 0

hrs

p/kW

h

Distribution

Startup energy

Generation energy

Generation

0

5

10

15

20

25

30

0 0

hrs

GW

1 RTid 2 RAer3 RWav 4 RSolPV5 RHydhh 8 Gchp6 Gchp 7 Ochp9 N 10 N12 C 19 C17 C 11 C18 C 13 C16 C 15 C14 C 53 Gcc42 Gcc 39 Gcc28 Gcc 43 Gcc54 Gcc 58 Gcc32 Gcc 30 Gcc60 Gcc 38 Gcc48 Gcc 33 Gcc40 Gcc 41 Gcc59 Gcc 50 Gcc45 Gcc 61 Gcc46 Gcc 52 Gcc31 Gcc 29 Gcc36 Gcc 56 Gcc51 Gcc 27 Gcc44 Gcc 57 Gcc49 Gcc 47 Gcc35 Gcc 37 Gcc55 Gcc 24 G23 G 25 O26 O 22 G21 G 20 G62 Ogt 63 Ogt65 Ogt

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0.0 10.0 20.0 30.0 40.0 50.0 60.0

GW

-5

0

5

10

15

20

25

30

1 25 0

5

10

15

20

25

30

SENCO

29

EleServe : matching demand to supply with load management

The simulation with the EleServe model demonstrates how a variable electricity supply of about 50% of peak demand, using heat storage alone, can be absorbed such that the net demand met by optional generation is flat.

This indicates that large fractions of variable electricity supply can be absorbed into the electricity system without special measures.

The question then is how the bulk of supply might be met with variable sources.

SENCO

30

Demand and supply correlation: general considerations

• Weather variables are correlated• Energy demands vary with time because of

social activity patterns and weather• Renewable energy supply is weather dependent.

The firm capacity of renewables is that amount of switchable (biomass/fossil/nuclear) capacity that doesn’t have to be built to meet demands with a certain probability.

The firm capacity of a renewable depends on the correlated variations in demands and that renewable supply, and the variations in some demands depend on the same weather parameters as the outputs from some renewables.

To illustrate:• if solar PV were to meet space heating demand,

its firm capacity would probably be close to 0%; if it were to meet air-conditioning demand it might be 50% as large A/C loads occur at times of high insolation.

• a significant fraction of space heating is positively correlated to wind speed and wind power, because it increases ventilation losses; whereas A/C load is negatively correlated with wind.

Insolation Ambient temp

Wind speed

Demand Space heat - - +

Lighting -

Air cond + + -

Cooling + + -

Transport + + -

Supply Wind +

Solar thermal

+ +

Solar PV + -

Heat pump +

Wave +

The table summarises these relationships, where ‘+’ means an increase of demand or supply flow with weather parameter. Note that the locations of demand and supply have to be accounted for; in general, the greater their separation, the less correlation between weather parameters.

SENCO

31

Demand and supply correlation : example; building heat loss and wind

The heat loss factors (W/C) from buildings due to ventilation and convection increase with wind speed. The chart illustrates these changes in a well insulated house, along with wind power, which increases as the cube of wind speed within the aerogenerator operating range. Assuming 10 M houses with temperature difference (internal-ambient) of 10 C, then the total heat loss from these houses rises from 13 GW to 16 GW as wind speed increases from 0 to 10 m/s. Note this ignores time lags due to building thermal mass and difference in demand and supply location,

0

20

40

60

80

100

120

140

160

180

0 1 2 3 4 5 6 7 8 9 10

Wind m/s

Hea

t lo

ss W

/C

0

20

40

60

80

100

120

Win

d p

ow

er

Vent

Glazing

Total heat loss

Wind power

SENCO

32

Demand and supply day sampling

The VarInt module of the EST model simulates weather, demand and supply hour by hour.

It comprises these calculations:

• Randomised weather is generated

• Hourly demands, dependent on weather and use patterns, are calculated

• CHP supply is calculated from heat loads

• Renewable supply is calculated from weather parameters

• Excesses and deficits of variable energy as compared to demand are calculated

• Excesses and deficits are input or output to trade, electricity stores and heat stores

• Any remaining deficit is met with optional generation (e.g. biomass or coal)

It works in two modes:

• Day sampling to estimate the storage capacity (GWh, GW) required to meet deficits and ensure no renewable energy wastage

• Optimisation: the minimum cost of meeting a given fraction of demand from variable sources is found.

These current results are for illustration only. They show a system in which electricity demands are, over the sampled year, met entirely from variable renewable and CHP electricity.

SENCO

33

VarInt : Sample dayThe Weather chart shows:

– ambient temperature, solar intensity and wind speed at demand location

– wind speeds, wave intensity and height of tide

The Demands and supply chart shows :

– weather dependent demands: air conditioning, lighting, space heat

– weather independent, electricity specific demands

– ‘additional demands’: inputs to stores, outputs to trade

The Supplies and demand chart shows :

– renewable generation: hydro, tidal, wave, aerogeneration , solar generation

– CHP

– ‘additional supplies’: outputs from stores, inputs from trade

– optional generation

The Cumulative demand and supply chart shows cumulative total demand and supply; the maximum difference between these determines storage needs.

The Data table shows:

– The difference between the day’s energy demand and supply (deficit shown positive)

– The maximum instantaneous shortfall in supply in any hour (GW)

– The storage requirement if no supply is to be wasted

SENCO

34

VarInt : Sample day : day of variable supply excessSENCO Energy, space, time model Demand and supply day sampling Month 1 Dummy data Days sampled 37

Current Max AverageDay's supply energy deficit -1018 273 266 GWh

Maximum supply power deficit 12 11 11 GWDiurnal storage requirement 134 49 48 GWh

Weather

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

1 5 9 13 17 21

Tem

p (C

) / w

ind

spee

d (m

/s)

0

50

100

150

200

250

Inso

lati

on (

W/m

2)

Tamb1

Wv_d1

Wv_g1

Wv_g2

Tide_g1

Sol_d1

Wave_g1

Demands and supply

0

10

20

30

40

50

1 5 9 13 17 21

GW

StHea_In

StEle_In

TradeOut

D_AC1

D_Lig1

D_Sp1

D_Hea1

D_ElSpe1

D_ElSpe1

D_AC1

D_Sp1

D_Lig1

SVar

SupTot

Supplies and demand

0

5

10

15

20

25

30

35

40

45

4 8 12 16 20

GW

Optional

StEle_Out

TradeIn

StHea_Out

Sol_g1

Aer_g2

Aer_g1

Hydro

Wave_g1

Tid_g

CHP

SVar

StEle_In

Tid_g

Sol_g1

Aer_g2

Hydro

Dem_Tot

Cumulative demand and supply

0

100

200

300

400

500

600

700

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

GW

h

Dem cum

Sup cum

SVarCum

SENCO

35

VarInt : Sample day : day of variable supply deficitSENCO Energy, space, time model Demand and supply day sampling Month 1 Dummy data Days sampled 37

Current Max AverageDay's supply energy deficit -55 273 266 GWh

Maximum supply power deficit 10 11 11 GWDiurnal storage requirement 44 49 48 GWh

Weather

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1 5 9 13 17 21

Tem

p (C

) / w

ind

spee

d (m

/s)

0

50

100

150

200

250

Inso

lati

on (

W/m

2)

Tamb1

Wv_d1

Wv_g1

Wv_g2

Tide_g1

Sol_d1

Wave_g1

Demands and supply

0

10

20

30

40

50

1 5 9 13 17 21

GW

StHea_In

StEle_In

TradeOut

D_AC1

D_Lig1

D_Sp1

D_Hea1

D_ElSpe1

D_ElSpe1

D_AC1

D_Sp1

D_Lig1

SVar

SupTot

Supplies and demand

0

5

10

15

20

25

30

4 8 12 16 20

GW

Optional

StEle_Out

TradeIn

StHea_Out

Sol_g1

Aer_g2

Aer_g1

Hydro

Wave_g1

Tid_g

CHP

SVar

StEle_In

Tid_g

Sol_g1

Aer_g2

Hydro

Dem_Tot

Cumulative demand and supply

0

50

100

150

200

250

300

350

400

450

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

GW

h

Dem cum

Sup cum

SVarCum

SENCO

36

VarInt : Day sampling : animation

SENCO

37

Demand and supply day sampling

This illustrative modelling indicates how large fractions of variable renewable and CHP electricity might be absorbed using system and ‘sub-system’ storage.

In practice, not all demand-supply mismatch would be absorbed with storage as it would not be the least-cost option.

System planning would take into account the statistics of the variations in demand and supply.

For the dummy data, the histograms shows some statistics from 500 random sample days based on dummy data:

– On about 10% of the days, storage requirement is greater than about half the maximum.

– Although the average energy deficit is zero, the distribution is skewed

Storage requirement

0%

20%

40%

60%

80%

100%

12 72 132 191 251 311 371 430 490 550 610

GWh

% d

ays

ex

cee

ded

Day's energy deficit

0%

20%

40%

60%

80%

100%

-694 -509 -323 -138 48 233

GWh

% D

ays

exce

eded

Day's maximum power deficit

0%

20%

40%

60%

80%

100%

0 2 4 6 8 10 12 14 16 18 20

GW

% d

ays

exceed

ed

SENCO

38

Demand and supply day sampling: diurnal storage requirements

For sample dummy data and weather functions modelled, approximate storage needs were calculated for the worst days in which storage capacity is maximum in terms of GW and GWh.

This is a maximum day’s storage scenario in that it assumes:

– no interruptible or degradable demands

– no ‘firm’ generation from biomass, fossil or nuclear fuels

– no electricity trade, either import or export

The current UK system level storage capacity provided by pumped storage is about 2 GW power and 10 GWh energy.

Residual needs can be met by ‘sub-system’ storage at the neighbourhood or building/vehicle level. For illustration, it is assumed that about half of the residual need is provided each by:

– 25 M hot water tanks in buildings of capacity 450 litres and maximum temperature change 20 C (e.g. 40 C to 60 C) providing 11 kWh energy storage per tank, and a heat input power of 0.4 kW

– 10 M electric vehicle battery sets in cars, local garages, etc. of energy capacity 43 kWh and power capacity 1.2 kW. Current vehicle batteries are around 30 kWh. Note that a fraction of in-vehicle batteries can not be used when vehicles are not grid connected.

Storage calculationStorage requirement (max) Max

Maximum supply power deficit GW 23Diurnal storage requirement GWh 613

System storage Pumped storage UK GW 2.0GWh 10.0

End use storage Buildings VehiclesStore output Heat Elec

Number M 25 10Fraction of total storage 45% 55%

Storage efficiency 99% 80%Capacity per unit kW 0.4 1.2

kWh 11.0 41.4

SENCO

39

VarInt : Sample yearSENCO Energy, space, time model Demand and supply year model Dummy data Months: 1,4,7,10 5 Days/month

Weather

0.0

5.0

10.0

15.0

20.0

25.0

1 1 1 1

Tem

p/W

ind/

Tid

e

0

100

200

300

400

500

600

700

Tamb1

Wv_d1

Wv_g1

Wv_g2

Tide_g1

Sol_d1

Wave_g1

Demands

0

5

10

15

20

25

30

1 1 1 1

GW

D_AC1

D_Lig1

D_Sp1

D_Hea1

D_ElSpe1

Supplies

0

5

10

15

20

25

30

35

40

1 1 1 1

GW

Optional

Trade

Sol_g1

Aer_g2

Aer_g1

Wave_g1

Tid_g

Hydro

CHP

Stores

0

20

40

60

80

1 1 1 1

GW

h

-20-15-10-5051015

GW

StEle_Sto StHea_Sto StEle_In StEle_Out StHea_In StHea_Out

SENCO

40

VarInt : Optimisation

The optimisation has these objectives and constraints:

Objective: Find minimum cost of system, where costs currently include;

• Capital costs of generation and storage

• Energy costs of optional generation (biomass, fossil) and trade

Decision variables:

• Capacities of variable generators, optional generation and stores

Constraints:

• Demands met

• Renewable output less than ‘economic’ maximum

• Fraction of optional generation less than some specified fraction

The optimisation is run for sample days representing a year of weather.

SENCO

41

VarInt : Optimisation :annual summaryOPTIMISATION

Months 4Days/month 5

SysPersNum 480EneNorm 0.0

Days sampled 500

SYSTEM SUMMARYFraction optional 5%

Energy CostTWh G£

Demand 131.9Supply Variable 137.9

Optional 0.0 0 GenerationNet storage loss 0.2 0Country supply 140.9

Country surplus 9.0Trade -9.0 -5 Trade value

Country supply 131.9139 Capital134 Total

Penalty Deficit/excess 00

Objective function 134

Demands Supply StorageRenewables Electricity Heat

D_L

ig1

D_H

ea1

D_S

p1

D_A

C1

D_E

lSpe

1

Hyd

ro

Sol

_g1

Aer

_g1

Aer

_g2

Wav

e_g1

Tid

_g

CH

P

Tra

de

Opt

iona

l

StE

le_I

n

StE

le_S

to

StE

le_O

ut

StH

ea_I

n

StH

ea_S

to

StH

ea_O

ut

GW GW GW GW GW GW GW GW GW GW GWh GW GW GWh GWCapacity 0 2 0 0 7 1 2 21 14 5 0 5 3 3 64 10 10 23 620 100

Max 1 15 25 25 5 5 6 5 20 100 999 100 100 999 100Min 0.6 0 0 0 0 0 4 2 0 2 10 2 20 10 20

Efficiency 86% 15% 25% 25% 15% 60% 0.88 0.88 1 0.88TWh 13 35 19 3 61 6 3 53 34 13 4 24 -9 0

Capacity factor 86% 14% 29% 28% 33% 325% 58% -39% 0% 0% 193% 0% -8% 57% 2%Net -12.18 175.3

Capital cost 1.0 1.0 1.0 1.0 1.0 1.2 5.0 1.0 1.0 1.8 2.0 0.4 1.0 0.5 0.1 0.5 0.1 0.1 0.1 0.0Total cost 1.0 12.1 20.8 14.0 8.2 0.3 1.9 2.6 1.4 6.4 5.0 1.0 2.3 62.0 0.0

Variable generators

0 5 10 15 20 25

Hydro

Aer_g1

Wave_g1

CHP

Optional

GW

Storage

0 100 200 300 400 500 600 700

StEle_In

StEle_Sto

StEle_Out

StHea_In

StHea_Sto

StHea_Out

GW/GWh

Optimiser SENCO

Model sheet EneModSpaceTime

Termination criteriaTolerance 0

Evaluations 9999Time (mins) 55

Decision variablesSteps in range 3

Minima -100Maxima 100Output

First method Genetic

Genetic method Recipe

Min found

1

10

100

1000

10000

100000

1000000

0 100 200

Evaluations

0

200

400

600

800

1000

1200

1 10 100 1000 10000 100000 1000000

SENCO

42

VarInt : Sample year : Optimisation animated

SENCO

43

VarInt : Optimisation :annual summary animated

SENCO

44

Energy systems : multi-fuelling and operational issues

Multi-fuelling of demandsSome demands can be met by more than one fuel which can have benefits or disbenefits.For example, water heating might be met by a mix of solar, gas and electric heat.• Solar heating and solar electricity generation are highly correlated and so cannot be used for

substitution• Gas or some storable fuel could be used when insolation is low, but this supply would have a

low capacity factor.

Multi-fuelling has these disbenefits:• capital costs are high because several supply technologies are required• the net demand on the auxiliary fuel (e.g. gas) becomes peaky and problematic

Operational issuesThe design and operation of systems depends on the reliability with which demands and supplies can

be predicted over different time scales, from minutes to months, and the sophistication of control over demand and supply technologies.

• Predictability– energy efficiency and demand management such as insulation reduces less predictable,

weather dependent loads • Technology control

– demand technologies; conversion – system technologies

SENCO

45

International electricity : demand

Further connecting the UK system to other countries increases the benefits of diversity, at the cost of transmission.

The first chart shows the pattern of monthly demands for different European countries.

The second chart shows the normalised diurnal demand patterns for some countries. Note that these are all for ‘local’ time; time zone differences would shift the curves and make the differences larger.

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

J an Feb Mar Apr May J un J ul Aug Sep Oct Nov Dec

AUT

BEL

CHE

DEU

DEN

ESP

FIN

FRA

GRE

EIR

ITA

LUX

NET

NOR

POR

SWE

GBR

0.50

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1.30

1.40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

AUT

BEL

CHE

DEU

DEN

ESP

FRA

GRE

ITA

NET

SWE

GBR

SENCO

46

International electricity: supply; monthly hydro output

Hydro will remain the dominant renewable in Europe for some time. It has a marked seasonality in output as shown in the chart; note that hydro output can vary significantly from year to year. Hydro embodies some energy storage and can be used to balance demand and supply, to a degree determined by system design and other factors such as environment.

Normalised hydro output

0%

20%

40%

60%

80%

100%

120%

140%

160%

180%

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

HydAUT

HydBEL

HydCHE

HydDNK

HydESP

HydFIN

HydFRA

HydDEU

HydITA

HydNOR

HydPOR

HydSWE

HydGBR

SENCO

47

Electricity trade

• An extensive continental grid already exists

• Diversity of demand and supply variations across geographical regions

• What is the best balance between local and remote supply?

InterEnergy model

• Trade of energy over links of finite capacity

• Time varying demands and supply

• Minimise avoidable marginal cost

• Marginal cost curves for supply generated by model such as EleServe

SENCO

48

InterEnergy – trade optimisation animated

SENCO

49

Security: preliminary generalities 1

Energy security can be defined as the maintenance of safe, economic energy services for social wellbeing and economic development, without excessive environmental degradation. All forms of energy supply (renewable, fossil, nuclear) present some form of insecurity.

A hierarchy of importance for energy services can be constructed:

• Core services which it is immediately dangerous to interrupt

– food supply

– domestic space heating, lighting

– emergency services; health, fire, police

• Intermediate importance. Provision of social services and short-lived essential commodities

• Lower importance. Long-lived and inessential commodities

Part of security planning is for these energy services to degrade gracefully to the core.

The various energy supply sources and technologies pose different kinds of insecurity:

• renewable sources are, to a degree, variable and/or unpredictable, except for biomass

• finite fossil and nuclear fuels suffer volatile increases in prices and ultimate unavailability

• some technologies present potentially large risks or irreversibility

SENCO

50

Security : preliminary generalities 2

Supply security over different time scales

• Gross availability of supply over future years. The main security is to reduce dependence on the imports of gas, oil and nuclear fuels and electricity through demand management and the development of renewable energy.

• Meeting seasonal and diurnal variations. This mainly causes difficulty with electricity, gas, and renewables except for biomass. Demand management reduces the seasonal variation in demand and thence the supply capacity problem for finite fuels and electricity. Storage and geographical extension of the system alleviates the problem.

Security of economic supply.

• Demand management reduces the costs of supply.

– The gross quantities of fuel imports are less, and therefore the marginal and average prices

– The reduced variations in demand bring reduced peak demands needs and therefore lower capacity costs and utilisation of the marginal high cost supplies

• The greater the fraction of renewable supply, the less the impact of imported fossil or nuclear fuel price rise

• A diverse mix of safe supplies each with small unit size will reduce the risks of a generic technology failure

Security from technology failure or attack. In the UK, the main risk is nuclear power.

Security from irreversible technology risk. In the UK, nuclear power and carbon sequestration

Environment impacts. All energy sources and technologies have impacts, but the main concern here are long term, effectively irreversible, regional and global impacts. The greater the use of demand management and renewable energy, the less fossil and nuclear, the less such large impacts.

SENCO

51

Electricity security

Demand management will reduce generation and peak capacity requirements as it :• reduces total demand• reduces the seasonal variation in demand, and thence maximum capacity requirements

It has been illustrated how load management might contribute to the matching of demand with variable supply. This can be further extended with storage, control and interruptible demand.

During the transition to CHP and renewable electricity, supply security measures could be exercised:• Retain some fossil fuel stations as reserves. Currently in the UK, there are these capacities:

– Coal 19 GW large domestic coal reserve– Oil 4.5 GW oil held in strategic reserves– Dual fired 5.6 GW– Gas 25 GW gas availability depends on other gas demands

• Utilisation, if necessary of some end use sector generation. Currently in excess of 7 GW, but these plants are less flexible because they are tied to end use production, services and emergency back-up

• The building of new flexible plant such as gas turbines if large stations are not suitable

Electricity trade with other countries can be used for balancing. There are geographical differences in the hourly variations of demands and renewable supply because of time zones, weather, etc. The strengthening of the link between France and the UK, and creation of links with other countries would enhance this option.

SENCO

52

A wider view of the longer term future

Wealthy countries like the UK can reduce their energy demands and emissions with cost-effective measures implemented in isolation from other counties, and in so doing improve their security. However, at some point it is more practical and cost-effective to consider how the UK can best solve energy and environment problems in concert with other countries.

As global fossil consumption declines because of availability, cost and the need to control climate change, then energy systems will need to be reinforced, extended and integrated over larger spatial scales.

This would be a continuation of the historical development of energy supply that has seen the geographical extension and integration of systems from local through to national and international systems.

The development and operation of these extended systems will have to be more sophisticated than currently. Presently, the bulk of variable demands in rich countries is met with reserves of fossil and nuclear fuels, the output of which can be changed by ‘turning a tap.’ When renewable energy constitutes a large fraction of supply, the matching of demands and supplies is a more complex problem both for planning and constructing a larger scale system, and in operating it.

SENCO

53

Europe and western Asia – large point sources

The environmental impact of energy is a global issue: what is the best strategy for reducing emissions within a larger region?

SENCO

54

World

There are global patterns in demands and renewable supplies:

• Regular diurnal and seasonal variations in demands, some climate dependent

• Regular diurnal and seasonal incomes of solar energy

• Predictable tidal energy income

SENCO

55

World: a global electricity transmission grid?

• Should transmission be global to achieve an optimum balance between supply, transmission and storage?

• Which investments are most cost efficient in reducing GHG emission? Should the UK invest in photovoltaic systems in Africa, rather than the UK? This could be done through the Clean Development Mechanism

SENCO

56

Conclusions and further investigation

Conclusions

– The general energy scenario context is critical

– CHP and biomass potentials are critical assumptions

– Load management alows large fractions of variable sources to be accommodated without additional back-up capacity

– End use and local storage could play a significant role

Further investigation

Further detailed assumptions, modelling and analysis required of all aspects of system integration:

– Demands: size, weather dependence, interuptibility, storage, multi-fuelling

– Supplies: electricity trade, renewable output prediction, CHP control

– Demand and supply prediction and control

Data and modelling

• Real time weather data for demand and supply

• Externd optimisation include demand reduction and spatial aspects