Abstract:

In order to give full play to the advantages of the combined kernel function in the relevance vector machine forecasting model and effectively improve the accuracy of load forecasting, a short-term load forecasting model based on the Bagging-combined kernel function correlation vector machine is proposed. Firstly, the forecasting model of combined kernel function relevance vector machine is constructed by weighted combination of Gaussian kernel function and Morlet wavelet kernel function, and then the particle swarm optimization algorithm is used to optimize the optimal weights of the two kernel functions. In order to improve the generalization ability of the model, the Bagging algorithm is used to sample the original data multiple times to construct a training sample set. Through the simulation of actual example, compared with a variety of relevance vector machine forecasting models, it is verified that the proposed model has good prediction accuracy.

Key words: Short-term load forecasting, relevance vector machine, combinatorial kernel function, Bagging algorithm, Morlet wavelet kernel function