A Highly Fault-Tolerant Distribution Network Fault Diagnosis Method Based on KMP Algorithm and Rough Set
Yiwen Gao, Cheng Long, Hua Zhang, Siyuan Jiang, Hongjun Gao · 2022
Currently, the use of multi-source fault telemetry data in distribution networks is becoming a mainstream method to diagnose faults in distribution networks, but the results are often unsatisfactory due to misreporting and omission of information. Based on this, a highly fault-tolerant distribution network fault diagnosis method combining KMP algorithm and rough set theory is proposed in this paper, which firstly uses KMP algorithm to calibrate the multi-source fault telematics data, then uses rough set theory to filter the features of the calibrated data to obtain the best attribute parsimony combination, and finally feeds the telematics data corresponding to the best attribute parsimony combination into a BP neural network for Finally, the telematics data corresponding to the best combination of attributes are fed into a BP neural network for training. The results of the simulation tests conducted on the Python 3.7 platform validate the effectiveness of the proposed method.