Summer Load Forecasting And Design Of Database System

A high-rate development of economy leads to the increase of power peak load, especially under some special weather conditions, power supply could not meet load demand. Two models of peak load forecasting based on analysis of the relation between weather information and peak load are proposed, and a three layers development model is used to design database system, which can forecast summer peak load accurately.Considering the characteristic of summer weather information and three load series decomposed by Multi-Resolution Analysis (MRA), different artificial neural networks (ANN) are used to establish a load forecast model. The model is applied to a practical example and recomposed by adding human body amenity indicator as a parameter. Several key factors are chosen into ANN by the rough set theory to form a new model while the less useful ones are reduced. The comparison of different models shows the last one has a more accurate forecast value.A load forecasting system using database technology is designed, which is composed of application layer, business logic layer and database service layer and most parts of the system are realized by component object model technology. Middleware technology is also applied in the SQLServer2000 database, which can be used not only in a single computer but also in web application.