Research on Power Quality Anomaly Monitoring Method for Distributed New Energy Access Power Grid
Power quality of power grid is vulnerable to shocks from distributed new energy access power grid.The shocks can destroy the integrity of voltage data and make it impossible to determine the monitoring threshold,resulting in long monitoring time and low accuracy.For this reason,the power quality abnormality monitoring method for distributed new energy access power grid is studied.A least squares support vector machine model is constructed,and the model hyperparameters are determined by an integrated learning particle swarm algorithm,and the optimized model is used to fill in the missing data of the voltage signal.The wavelet transform is used to extract the distribution characteristics of power quality data and obtain the difference between the energy distribution of each layer and the standard signal energy distribution.The back propagation(BP)neural network initial weights and monitoring thresholds are optimized by introducing the salp swarm optimization algorithm,and the difference values are input into the trained BP neural network to realize power quality abnormality monitoring.The experimental results show that the proposed method has an accuracy and completeness rate of more than 95%,and the training time and testing time are about 15 ms.The method can accurately and efficiently monitor the abnormal data and ensure the stable operation of the power grid.
Distributed new energyAccess power gridPower quality anomaly monitoringSalp swarm algorithmLeast squares support vector machineParticle swarm algorithm
NETL
NSTL
万方数据
