Short-term load forecasting based on similar day approach and intelligent algorithm using analytic hierarchy process

Short-term load forecasting (STLF) is a basic work in power network planning. The accuracy of its prediction has an important impact on the actual power generation and distribution. Power sector develop efficient and economical power generation plans also based on it. A load forecasting method is proposed in this paper, using grey relational analysis to correlate factors and fuzzy clustering to select load similar day. This preprocessing method meets the requirement of different types of electricity load forecasting. Then analytic hierarchy process (AHP) is used to model the BP neural network, particle swarm optimization support vector machine and time series intelligent algorithm, forming a combined prediction model. In experiments, the comparison between the error of the combined prediction result and the single prediction models verifies the improvement effect of the fusion model based on similar day and algorithm using analytic hierarchy process.