Research on Abnormal Data Repair Model in Smart Grid
To further improve the quality of power load abnormal data repair,an improved beetle antennae search(IBAS)algo-rithm is proposed to optimize the parameters of the radial basis function(RBF)kernel neural network for abnormal data repair.Based on BAS,dynamic inertia weights and Levi's flight trajectory optimization mechanism are introduced to improve the IBAS algorithm.IBAS is applied to optimize RBF network parameters and an abnormal data repair model for IBAS-RBF is construc-ted.The abnormal data of single and continuous power load points are repaired,and the repair quality is evaluated through quality evaluation indicators.The experimental results show that compared to the BAS algorithm,PSO algorithm and FPA al-gorithm,the IBAS algorithm significantly improves its optimization speed and accuracy.After using the IBAS-RBF model for power load data restoration,the accuracy and consistency of the load series are improved by 7.8％and 7.6％,respectively,and the trend and effectiveness are improved by 6.6％and 2.1％,respectively.This indicates that this model can achieve the elimi-nation of extreme power load anomalies and the repair of continuous point anomalies,significantly improves the smoothness of the repair trajectory.
power loadabnormal detectionIBASRBF networkdata repair
万方数据
维普
