Time series forecast with neural network and wavelet techniques

Exhibitor: Ruey Hwa Loh

Supervisor: Zhao Dong

Research Group: Complex and Intelligent Systems

Industry Sector: Energy and Utilities

Forecasting of electricity has always been the essential part of an efficient power system planning and operation, especially short term forecasts as it has becoming increasingly important since the rise of the competitive energy markets. The aim of short term load forecast is to predict future electricity demands based on historical data and other information such as temperature. This thesis proposes designing a model using neural network and wavelet techniques to increase the accuracy of time series load forecast.

The model is created in the form of a program package written with MATLAB®. The time series data used are historical electricity load and pricing data of Queensland obtained from the NEMMCO website. Wavelet technique is implemented to the time series data, decomposing the data into number of wavelet coefficient signals. The decomposed signals are then fed into neural network for training. To obtain the predict forecast, the outputs from the neural network are recombined using the same wavelet technique

The simulation results showed that the model was capable of producing a reasonable forecasting accuracy in short term load forecast.

Thesis Document (PDF)