Integrating social practice and economic rationality principles in household energy

demand modelling

Dimitrios Papadaskalopoulos, Goran Strbac, CAP, Imperial CollegeKavin Narasimhan, Tom Roberts, Nigel Gilbert, CRESS, University of Surrey

wholeSEM 3rd annual conferenceSession 2a: Integrating society within energy systems modelling

05/07/2016, Cambridge, U.K.

Structure of the presentation

• Emerging challenges for power systems

• Role of demand side in emerging setting

• Economic rationality / utility maximization framework

• Social practice framework

• HOPES model

• INPUTS project

2

Emerging challenges for power systems

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

1 69 137

205

273

341

409

477

545

613

681

749

817

885

953

1021

1089

1157

1225

1293

1361

1429

• Under-utilised conventional generation needs to remain in the system as a “back-up” energy source and flexibility provider

COST EFFICIENCY?3

0

20

40

60

80

100

120

140

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

Demand(GW)

Time(h)0%penetration 10%penetration 30%penetration

50%penetration 100%penetration

• Under-utilisedgeneration and network capacity needs to be built in order to cover new demand peaks

Role of demand side in addressing emerging challenges• Energy efficiencyØImproved insulation of buildingsØUse of efficient electrical appliancesØChanges in consumers’ behavior

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

• Demand flexibility / redistributionØShift operation of appliances in timeØUse appliances with storage

components > decouple acquisition and consumption of energy

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

Overall reduction of generation and network requirements

More efficient utilisation of generation and network assets

4

Flexible demand appliances

5

Example: Impact of smart charging of EV

7

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Increase

Energyincrease

PeakincreaseBaU

PeakincreaseSmart

Economic rationality / utility maximisation framework• The consumers determine the schedule of their appliances by

maximizing their perceived utilityØBased on microeconomics foundationsØMost common approach in power systems / energy

economics literature• Maximize U = B – λ*P subject to operational constraints

ØU : perceived utility (expressed in monetary terms e.g. £)ØB : perceived benefit or satisfaction from the use of the

appliances, expressing consumers’ preferences and requirements (£)

Øλ : electricity price at each time period (£/kW)ØP : electrical power consumed at each time period (kW)Øλ*P : electricity payment (£) 6

• Benefit function B is a non-decreasing, concave (usually quadratic) function of the instantaneous power demand

• Marginal benefit B’ function is a linear decreasing function of instantaneous power demandØAs the level of demand increases, the extra satisfaction that

a consumer perceives is reduced

7

Economic rationality / utility maximisation framework

• Insights from sociologyØSatisfaction is not a function of electrical power > the

satisfaction that the consumers perceive depend on the service quality and is appliance-specific

ØSatisfaction cannot be accurately expressed in monetary terms, even by the consumers themselves

ØHuman consumers do not generally behave economically rationally > they do not always consciously think about energy use

ØA number of other factors drive consumers’ behavior > perceptions, habits, social interactions etc

8

Economic rationality / utility maximisation framework

• Practices: repeated activities undertaken by people in their daily lives (cooking, cleaning, washing, etc.)

• Individuals are carriers of social practices

• Energy use is driven by the daily rhythms of practices performed in households

• Not only have appliances changed over the years, but the ways in which appliances are used have also changed over the years. This in turn influences energy use

9

Social practice framework

Practices

Energy Consumption

Individuals (or Households)

Social practice framework

SOCIAL PRACTICE

MEANING

SKILLMATERIAL

Outcome(s) intended from performing a

practice

An instrument (object, body, mind) used to carry out a practice

Ability to carry out practice (knowledge,

know-how)

10

Social practice framework

11

Agent-based modelling

An approach used to situate an initial population of agents (autonomous and heterogenous entities) in a relevant environment; allow them to interact according to simple rules, and thereby generate (or ‘grow’) a macroscopic phenomenon from bottom-up (Epstein 1999:42)

Households Elements Practices12

Interactions among agents

13

HOuseholds and Practices in Energy consumption Scenarios (HOPES) model

Household demographics & Energy appliances

INPUT

Energy consumed

OUTPUT

Pick-elements: Households draw elements from the system

Perform-practices: Households perform practices by combining elements

Adapt-elements: Elements in the system adapt over time

1. Based on weather2. Based on tenure, type and

occupancy3. Based on working patterns4. Based on social interactions5. Based on targeted

information6. Based on the history of

practices performed7. Based on electricity tariff

8. Verify if the elements that households have are appropriate for performing the practices

9. Update the ‘state’ attribute of elements

10. Adapt elements based on their state using Crossover (a Genetic Algorithm approach)

14

INtegrated Practice and Utility based perspectiveson demand Time Shifting (INPUTS)

• How can we integrate the electricity price in the social practice modeling framework?

• Which data to use in order to calibrate HOPES model?• How does this more detailed demand side model affect power system

outcomes at the operational and planning timescale?• How does this difference in outcomes depend on:

ØThe evolution of the power system: Level of renewables, electrification level of heat and transport sectors etc

ØThe electricity market arrangements: how many different prices communicated to the consumers

ØThe level of system considered: national level with a very large number of diverse consumers against local neighborhood level with a small number of consumers

15

d.papadaskalopoulos08@imperial.ac.ukk.narasimhan@surrey.ac.uk

Comments / Questions ?