Short Term Load Forecasting System

The short-term load forecast of electric power system is an important routine for power dispatch and utility departments. Its precision will influence the economic and secure operation of power systems and quality of power supply. The features of short-term load forecast can be generalized as followings: many data need to be forecasted, the physical factors which influence forecast are complicated and random, and high is demanded. Short-term load forecast software is necessary for automation of power system.There are many short-term load forecasting methods. Single model's precision can not be broken through for its merits and faults and limits of using information. Integrated model is an important mark of the progress of short-term load forecasting. The thesis builds an integrated model based on gray and fuzzy logic system by analyzing the load's characteristics of one area. Gray model needs fewer history data, far calculation and high precision to load forecast of areas with stable load. But the load is also influenced by many uncertain factors, for example, date type, weather and so on. Now there is no a determinate model to measure how these factors influence load. Fuzzy theory can deal with uncertain factors well and bring in expert's experience, so that it becomes a good method of short-term load forecast. The basic load is forecasted by gray model, then the random load which is influenced by weather, date type and holiday is forecasted by fuzzy rule pool which is constructed by each factor's subjecting function. Furthermore, some valuable experience is made full use by utilizing fuzzy language variables and numeric variables.The software of short-term load forecast is designed based on the method above. The software has been used now and shows the model has high precision. Furthermore, the software is easy to operate and maintain so that gets the acceptance of the user.