Grey correlation-oriented Random Forest and Particle Swarm Optimization Algorithm for Power Load Forecasting

Accurate power load prediction plays an important role in the design of power distribution equipment and distribution network. The traditional forecasting methods have the problems with low accuracy of power load forecasting and slow model training. In order to improve the accuracy of power load fore-casting, this paper proposes a new method combining Grey correlation-oriented random forest with par-tide swarm optimization algorithm for power load prediction. The method first uses Grey correlation projection to measure the similarity between the attributes of historical samples and the attributes of predicted samples, and it constructs the similar historical sample data set. Then the decision tree of ran-dom forest is optimized based on particle swarm optimization to improve the prediction accuracy. Fi-nally, Hadoop distributed cluster is used to realize the parallelization of power load prediction and im-prove the prediction efficiency. The experimental results show that the proposed model in this paper has better prediction performance than the traditional power load forecasting methods.