Research on Identifying Abnormal Power Values in New Energy Generation Based on Extreme Learning Machines
New energy generation is significantly affected by weather conditions,resulting in significant fluctuations and uncertainties in power generation.These fluctuations and uncertainties may mask or simulate real outliers,making anomaly detection difficult.Therefore,a new energy generation power outlier recognition method based on extreme learning machines is proposed.Generate a flow rate interval dataset based on the coverage area of the power generation device,determine the number of abnormal flow rate intervals,and then divide the abnormal flow rate intervals of new energy power generation.Combining load fluctuations and extreme learning machines,identify the abnormal values of new energy power generation.The experimental results show that the MAE,MAPE,and RMSE values of the designed new energy generation power outlier limit learning machine recognition method are relatively low,proving that the recognition effect of the designed recognition method is good and can make a certain contribution to improving the comprehensive efficiency of new energy generation.
万方数据
维普
