Case-Based Reasoning Combined with Information Entropy and Principal Component Analysis for Short-Term Load Forecasting

Short-term load forecasting (STLF) plays a vital part in the operation of electric power system, and it relates the security, stability, and economic dispatch of the system. In this paper, rough sets information entropy (IE) and principal component analysis (PCA) are applied to the attributes reduction of load cases, and respectively, the significance and relativity of load data are disposed. Thus, not only is the training time in the process of retrieval decreased, but also is effective control implemented aiming at petit factors to essential ones. In the process of revise, some impactful amendments are presented to improve prediction precision. Finally, this scheme is performed on the data of Bao Ding Electric Power Company (BDEPC) during 2000-2004, and the testing result shows that the proposed model is feasible and promising for load forecasting. Index Terms - short-term load forecasting, case-based reasoning, information entropy, principal component analysis, neural network