A Neuro-Fuzzy Approach for Modelling Electricity Demand in Victoria

TL;DR

This paper compares neuro-fuzzy, neural network, and statistical models for predicting electricity demand in Victoria, showing neuro-fuzzy systems outperform other methods in forecast accuracy.

Contribution

It introduces and evaluates an evolving fuzzy neural network for electricity demand forecasting, demonstrating its superior performance over traditional and neural network models.

Findings

Neuro-fuzzy system outperforms neural networks, ARIMA, and VPX forecasts.

Evolving fuzzy neural network provides more accurate demand predictions.

Study uses 10 months of 30-minute interval data for model training and testing.

Abstract

Neuro-fuzzy systems have attracted growing interest of researchers in various scientific and engineering areas due to the increasing need of intelligent systems. This paper evaluates the use of two popular soft computing techniques and conventional statistical approach based on Box--Jenkins autoregressive integrated moving average (ARIMA) model to predict electricity demand in the State of Victoria, Australia. The soft computing methods considered are an evolving fuzzy neural network (EFuNN) and an artificial neural network (ANN) trained using scaled conjugate gradient algorithm (CGA) and backpropagation (BP) algorithm. The forecast accuracy is compared with the forecasts used by Victorian Power Exchange (VPX) and the actual energy demand. To evaluate, we considered load demand patterns for 10 consecutive months taken every 30 min for training the different prediction models. Test results show that the neuro-fuzzy system performed better than neural networks, ARIMA model and the VPX forecasts.