Adaptive Leakage Protection for Low-Voltage Distribution Systems Based on SSA-BP Neural Network

The fluctuation of normal leakage current has a great influence on the fixed-threshold protector. To address this issue, paper proposes an adaptive protection method based sparrow search algorithm (SSA)-backpropagation (BP) neural network. Based analysis generation mechanism, uses SSA optimized BP network to construct prediction model current. By dividing range into several intervals and setting corresponding action threshold, threshold interval is automatically selected in advance, predicted value model, so as realize faults. Experiments have proved that protector can identify fault more sensitively increase ratio put operation by predicting development adjusting advance.