Enhancing forecasting

capabilities in the IMO:

Presentation to the MAC

Peter Shardlow

Technical Specialist Market Modelling

9 September 2015

Presentation Overview

Technological innovation is changing electricity market fundamentals.

• New technologies are providing exciting opportunities for market participants and customers.

The IMO recognises the need to adapt to the evolving energy landscape.

• Enhancing our forecasting and modelling capabilities to meet the challenges of changing conditions

in the Western Australian energy market.

We’re focused on delivering improvements in three key areas in FY16:

• Long term demand forecasting;

• Analysis of customer behaviour and segmentation; and

• Modelling of small-scale solar PV and storage technology.

2

The electricity market is evolving in response to

technological innovation

3

Traditional consumer/producer model breaking

down due to microgeneration and storage.

Demand influenced by technology that gives

consumers control over how and when they use

energy.

Internet-linked technology allowing for

continuous interaction between consumer,

retailer and wholesale market.

Where we’re heading

Utilities focused on the safe, reliable delivery of

reasonably priced electricity.

Demand driven by population growth, economic

activity, air conditioning sales and efficiency

measures.

Little scope for dynamic interactions between

customer and supplier.

Where we’ve been

Our capacity to undertake detailed analysis of energy

data is also improving

4

Improved analytical capabilities

Satellite

cloud cover

data

Low cost IT

infrastructureSmart

Metering

Live solar +

storage data

New sources of information

New

analytical

techniques

Three focus areas for delivering improved forecasting

and analysis at the IMO

5

Long-term demand

forecasting

Customer behaviour

analysis

Modelling of solar PV and

storage technology

Improving the IMO’s long-term demand forecasting

capabilities

6

Improved forecasting

algorithms.

More robust modelling and ability to

analyse variability over different time

horizons.

Development of synthetic

climate data and simulation

model.

Better approach to understanding

climate-based demand variability.

Integrated data visualisation

tools.Models become easier to interrogate

and more efficient to update.

Incorporation of new data

sources (e.g. solar) into our

modelling.

Improved forecast accuracy.

IMO

Deli

ve

rab

le

Ben

efit to

pa

rticip

an

ts

Market Segmentation will focus on six questions to

better understand energy use in Western Australia

7

Residential Industrial/Commercial

Market Segment

What does the residential

customer demand profile

look like?

Will segmentation help us

understand the impact of

changing demographics?

Which economic variables

impact customer energy

demand?

Have energy use patterns

changed significantly

since market start?

Do household’s respond

to solar by changing

consumption behaviour?

How are customers

responding to wholesale

market signals?

Purpose

Validate critical forecasting

assumptions with data.

Improved understanding of

demand variability.

Inform market evolution and

policy development.

Analysis of metering data is providing insight into how

customers are responding to market mechanisms

8

Forecast temperature over

35°c, actual temperature

was 28°c

3

Downward deviation

aligned with high

temperature

2

Daily maximum

over 35°c

1

Delivering improved small-scale PV and storage

simulation models and analysis

9

Real-time rooftop solar

generation data

(FY16)

Publish estimates of

small-scale solar PV

generation in the SWIS.

Provides participants with

data for (unmetered)

solar generation.

Improved battery

storage modelling

(FY16)

Incorporate simulation

modelling of battery

controller technology in

the next ESOO.

Better understanding of

the opportunities and risks

around small-scale

storage.

Near-term rooftop solar

forecasting

(FY17)

Use cloud cover data to

forecast rooftop solar

generation.

Better forecasts of net

customer demand in the

near future.

Pro

jec

tD

eli

ve

rab

leB

en

efi

t

How we’re using real-time data to gain insight into the

impact of solar PV variability on the SWIS

10

Figure 1: Estimated SWIS Rooftop solar production (% of maximum

output)

75%10:30 2,300 MW

85%12:00 -100

25%11:00 +70

Rooftop Solar + Metered Generation

0%

25%

50%

75%

100%

0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00

Loss

Solar PV

-

500

1,000

1,500

2009-10 2011-12 2013-14 2015-16 2017-18 2019-20 2021-22 2023-24

Projected

Historic

Figure 2: Historic and projected solar PV capacity, 2014 ESOO expected case

Three focus areas for delivering improved forecasting

and analysis at the IMO

11

Long-term demand forecasting

Customer behaviour analysis

Modelling of solar PV and storage technology

Support the IMO and market

participants in delivering on

the Wholesale Electricity

Market objectives

Project next steps

Three ways the MAC can assist the IMO to deliver better analysis.

12

The IMO’s forecasting work is not developed in a vacuum

We would like feedback on how market participants use our analysis and how we can better align our

deliverables with stakeholder requirements.

We believe that transparency is paramount

We believe there are opportunities to work with analysts from other organisations to obtain critical

feedback on our methodology and results.

We need assistance to understand new innovative technologies

The market is evolving rapidly, and we need to adapt our modelling to capture the effects of

innovative new technologies.