A study and implementation of an intelligent load forecast support system

Abstract

Computer decision support systems have been more widely applied in the planning of urban power systems, especially the distribution systems. A load forecast is a basic stage in the process of the distribution system planning. Hence, a load forecast support system (LFSS) is a basic sub-system in the package of a decision support system for urban power systems planning. The knowledge and experience of planners play a vital part in the practical forecasting process. Thus, it is urgent for researchers to find an approach to accumulate planners' knowledge and experience during load forecasting and reuse them in the further forecasting. The combination of artificial intelligence (AI) technology and LFSS provides an effective approach to solve this problem. This paper has proposes a model and an implementation frame of an intelligent load-forecast support system (ILFSS), which provides a total solution to store and reusing load forecasting knowledge. This approach has applied the AI technology to the whole process of load forecasting, including model definition, model selecting, result adjusting and decision making. Some AI technologies like rule-based reasoning and case-based reasoning are involved in the implementation of ILFSS, which cooperate with conventional load forecasting models. Up to now, some of the proposed architecture has been implemented in a real decision support system for urban power system planning called CNP, which is widely used in China.