Grid Computing Solutions for Artificial Neural Network-based Electricity Market Forecasts

This paper presents a grid computing approach to parallel-process a neural network time-series model for forecasting electricity market prices. A grid computing environment introduced in a university computing laboratory provides an access to otherwise unused computing resources. The grid computing of the neural network model not only processes several times faster than a single iterative process but also provides chances of improving forecasting accuracy. Results of numerical tests using the real market data by over twenty grid-connected PCs are reported.