Short-Term Load Forecasting Based on Neural Network

Based on the fundamental principle that multi layer perceptron can approach linear or nonlinear functions at any accuracy, this paper presents a short term load forecasting method using multi layer feedforward neural network considering the influence of climate and formulates the construction and training method of the corresponding BP algorithm. Experiment results show that the short term load forecasting method based on neural network is of high precision. The relative error of load forecasting using neural network model is less than 3.67%.