Study of Super Short-Term Bus Load Forecasting Model Based on Similar Ranges

Abstract

With the increasing demand of electricity dispatching and the request of energy conservation and emission reduction, the dispatching plan and operating of electric power has become more and more important. On one hand, electric power companies try to reduce power reserve as much as possible to increase the efficiency; on the other hand, some power reserve is necessary to deal with emergency and to ensure the safety and stability of grid. Therefore, this paper proposes a super short-term bus load forecasting model which is based on similar ranges to track the variation of weather and load. By using the method of fruit flies optimizing grey neural network in the real time, it can reduce the size of the network computing and solve the problem of divergence. Since the system based on the model has operated for one year, it proves that this model can meet the requirement of the precision for the electricity dispatching and adapt to the changes in different regions.