Identification of voltage sag source based on BAS-BP classifier model

Abstract

In order to improve the recognition accuracy of different voltage sag disturbance sources and effectively control the voltage sag, a method of voltage sag source identification based on beetle antennae search (BAS)-back propagation (BP) classifier model constructed by longicorn BAS and BP neural network is proposed. In this paper, the improved S-transform is used to extract 16 characteristic indicators to form a voltage sag source identification indicator system. In order to eliminate the influence of redundant information on the classification results, 9 indicators are selected as the input of the classifier using the combination weighting method. By optimizing the initial weights and thresholds of BP neural network by BAS, the BAS-BP classifier model is constructed to identify different types of voltage sag sources in distribution network. The simulation results show that the classifier model has certain anti-noise ability and applicability, and has a better classification than the conventional classifier model dose.

Keywords

• voltage sag
• improved s-transform
• combination weighting
• beetle antennae search-back propagation (bas-bp) classifier
• classification and recognition
• back propagation (bp) neural network