Short-Term Electricity Price Forecast Using Neural Network

Deregulation of the electric power industry worldwide raises a series of challenging issues. Forecasting the market clearing price (MCP) is the most essential task and the basis for any decision-making. The basic idea is to use the system spin reservation and the historical load and price to forecast the future price. This paper presents a successful application of a neural network to MCP forecast for Queensland day-ahead energy market in Australia. The structure of the neural network is a three-layer back-propagation (BP) network. The historical loads and prices of 1998 in Queensland day-ahead energy market are used for training and forecasting Results show that the proposed method is effective.