Hybrid Load Forecasting Method With Analysis of Temperature Sensitivities

Abstract

Load forecasting is essential in the electricity market for the participants to manage the market efficiently and stably. Many techniques for load forecasting, which are, for example, multiple linear regression, stochastic time series, Kalman filter, expert system, and computational intelligences such as fuzzy systems and artificial neural networks, have been investigated so far. This paper proposes a novel hybrid load forecasting algorithm, which combines the fuzzy linear regression method and the general exponential smoothing method with the analysis of temperature sensitivities. The fuzzy linear regression method is used to consider the lower load-demands in weekends and Monday than on weekdays. The normal load of weekdays is forecasted by the general exponential smoothing method. Moreover, the temperature sensitivities are used to improve the accuracy of the load forecasting with the relation to the daily load and temperature. The test results show that the proposed algorithm improves the accuracy of the load forecasting in 1996.