A New Algorithm for Power System Shortterm Load Forecasting

Short-term load forcasting is an important routine work in power system. This paper proposed a novel Volterra model for short-term load prediction. In the first place whether the load time series has chaotic character is determined by chaos theory. There is one more point that the load time series are reconstructed by phase space reconstruction theory. The new load prediction algorithm based on Volterra model which makes no truncation and contains more information of system. The load data of a city in China are used to verify the effectiveness of the new algorithm. Results indicate that the error between predict load and true load is less than 5%. The prediction short-term load is accurate enough for electric power dispatching.