IPSO-BP Power Quality Disturbance Identification Method Based on Feature Screening

Among the current methods for power quality disturbance analysis, the time domain analysis method requires multiple calculations so that the characteristics of the power quality disturbance signal cannot be quickly extracted. How to effectively utilize the time domain characteristics of the original signal within a specific range needs to be solved. Key issues, and when extracting time-domain features, multiple features need to be collected. Multiple features lead to high feature dimensions of the dataset, which increases the structural complexity of the classifier and makes it difficult to optimize model parameters. Therefore, this paper adopts a feature score, and The genetic algorithm is a two-stage feature selection algorithm. After screening a large number of features, the features are screened before the evolutionary calculation starts, and the screened features are fully utilized in the evolutionary calculation process to improve the classification accuracy, and an IPSO-BP algorithm is used, reducing the time spent on classification.