The problem of short-term power grid load forecasting is studied,a power grid load forecasting method based on im-proved grey wolf optimizer(IGWO)algorithm and extreme learning machine is proposed.The IGWO is proposed,the self var-ying weight coefficient changes and elite disturbance update strategies are introduced to improve the global optimization ability of IGWO.The density based spatial clustering of application with noise(DBSCAN)algorithm is used for clustering analysis of power load data to minimize the impact of data differences on prediction accuracy.The IGWO optimized ELM model(IGWO-ELM)is used to predict the multi cluster power load data-set.The simulation results show that compared with other prediction methods,the proposed classification IGWO-ELM has higher prediction accuracy.
关键词
电力负荷预测/灰狼算法/DBSCAN/极限学习机/预测精度
Key words
power load forecasting/gray wolf optimizer/DBSCAN/extreme learning machine/prediction accuracy
维普
万方数据