Short-term Load Forecasting Based on Hierarchical Clustering and ISA-LSSVM Model

Abstract

For short-term load forecasting of different types of users, support vector machine and deep learning model are widely used at present. A hybrid model is proposed to solve the problems of the least squares support vector machine (LSSVM) model, such as the difficulty in determining the hyperparameters, the high data quality requirements of the model, and the slow optimization speed and easy to fall into the local optimization of the integrated conventional optimization algorithm. In this model, firstly, the original feature data is clustered by hierarchical clustering (HC) and then the corresponding LSSVM model is established for the same prediction day. Then, the super parameters in LSSVM are heuristic searched by the improved simulated annealing algorithm (ISA). Finally, by comparing the performance of the load forecasting model with that of various load forecasting models, the results show that the proposed model can effectively improve the accuracy of load forecasting and shorten the forecasting time.