File Exchange

image thumbnail

Electricity Load Forecasting for the Australian Market Case Study

version (13.7 MB) by David Willingham
This is a case study of forecasting short-term electricity loads for the Australian market.


Updated 01 Sep 2016

View Version History

View License

This is a case study of how MATLAB can be used to forecast short-term electricity loads for the Australian market using Sydney temperature and NSW histroical load data sets. Nonlinear regression and neural network modeling techniques are used to demonstrate accurate modeling using historical, seasonal, day-of-the week, and temperature data.
Highlights include:

• Forecasting short-term electricity loads and prices

• Accessing data from regional wholesale electricity markets

• “White-box” modeling using customisable algorithms and viewable-source functions

• Automatic Report Publishing

This case study is for practitioners at power generators, utilities or energy trading groups whose focus is transmission planning, distribution operations, derivative valuation, or quantitative analysis. Familiarity with MATLAB is not required.

Cite As

David Willingham (2021). Electricity Load Forecasting for the Australian Market Case Study (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (12)

Abolfazl Nejatian

Nicely Written Code,
This is my forecasting code which allows you to predict your time series data with LSTM, CNN, and MLP Networks.

please visit my Mathworks account and download it from the link below:

ya gao

prajwol kharel

HM01X_Data_066062_32826586401242.txt where is this text file


Kindly share the direct link to dataset used.

SOUY Bunheng

Do you have video explain on how to build this code

Kalyan Malla

Hi David, is a recorded webinar of this case study available? I am actually carrying out a similar study. It would be a great input to me. Thanks.

David Willingham

Hi Philippe,

I have developed another example for long term forecasting using econometrics techniques, check it out at:


Hi has someone modified it for a long term forecasting ?


Faiz Mahdi

Hi David

I am usinf Artificial Neural network in my work to predict a model that can be used same experiments information to predict new results with less error ratio in comparison to experimental and theoretical ones.

I have done some work and got results but the error ratio is still high.

If you can help me or you know any one how is doing this work please send me back on my email,



MATLAB Release Compatibility
Created with R2011a
Compatible with any release
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!