A Short-term Load Forecasting Based On Fuzzy Identification In Power System

Due to the industrial structure adjustment, the change of resident consumption ability and pattern of consumption, and market-oriented and so on, Short-term power load forecasting for urban plans faces considerable difficulties. Aiming at the limitation of short-term load forecasting technique for power system under the particular circumstance, we presents the methods of classifying load by using fuzzy pattern identify theory, then constructs all kinds of ANN according to class. The method is applied to a short-term load forecasting, and guaranteed exactness and stability of the results. In this paper, a neural network prediction model was established, the example in Heilongjiang was given to validate the accuracy of the algorithm.