Method of short-term load forecasting based on BAYESIAN theorem

Bayesian learning is a probability method that makes optimal decision based on known probability distribution and recently observed data. In the paper, by using the Bays estimate method, the weight of every forecasting model is obtained. Support Vector machines and Spectrum analysis are selected to construct the Bays combined model, which are applied to forecast. The forecasting method gives bigger weight to the models, which better conform to the variation of power load, and improves the precision. The sample calculation shows the combined model is better than those of the singular one.