Analysis And Application Of On-line Super Short-time Load Forecasting

Along with the development of electric power industry and the scale-up of power network, the structure and operation mode become more and more complex. It is an important goal to improve the security and economy of power system. The sudden load increase or decrease may effect on security of electric power system badly. So it is important to forecast load and track the load alteration.With the advancing of electric power market, people have paid a lot of attention on short-time forecasting, focusing on how to improve the precision of load forecasting, but using the load forecasting to improve the security of system is quite limited. When the electric power system operation mode is changed, the department of dispatch needs to know the future alteration of load from a few minutes to less than an hour. The short-time load forecasting usually is forecasting hourly, which cannot meet the needs of the requirement of dispatch department. In this paper, two models of online super short-time load forecasting are proposed. One is based on the linear extrapolation, the other on the BP. Using this model in flow computation and security analysis can help to analysis the secure operation state of power system at the next interval when the load alteration violate the fixed value, which would help operators to control the status of power network at the next interval, and take some corresponding operation to improve the security and reliability of power network.The forecasting model presented in this paper has a reasonable structure, and it can meet the real time requirement of power system. Using it into flow computation and security analysis can improve the security and reliability of electric power system.