Fault Diagnosis Scheme Based on Grid Fault Degree Sample Data and AI Classification Method
Abstract
Considering the differences of fault characteristics in the process of fault diagnosis, loss and false positives of protection and circuit breaker information, power grid fault diagnosis method based on integrated fault degree is proposed. Firstly , this method uses wavelet transform to analyze the amplitude characteristics and energy characteristics in the electrical quantity information on the fault recording data, and extracts the key indicators; Then, according to the binary information from remote communication system, directed bipartite graph is used to update fault symptoms, determine the fault boundary, and to calculate suspected degree; finally, integrated fault degree is determined by the improved D-S evidence theory, and final fault diagnosis result is determined by improved support vector machine method. Simulation results show that the proposed scheme improves the accuracy of diagnosis, and has good application prospects.
Keywords
fault diagnosis, wavelet transform, directed bipartite graph, D-S evidence theory, support vector machine
DOI
10.12783/dtetr/mcaee2020/35052
10.12783/dtetr/mcaee2020/35052
Refbacks
- There are currently no refbacks.