Circuit breaker mechanical fault identification method and device based on DS evidence fusion

Abstract

The invention discloses a circuit breaker mechanical fault identification method and device based on DS evidence fusion. The method comprises three parts of training sample sampling, LSTM network training and DS evidence fusion of multiple diagnosis results. Firstly, a differentiated training sample set is obtained through a Latin hypercube sampling mode, secondly, an LSTM algorithm with the same network structure is constructed and trained to obtain differentiated sub-fault diagnosers, and finally, a DS evidence theory is adopted to fuse all LSTM diagnosis results to obtain a final circuit breaker operation mechanical fault identification result. According to the invention, the accuracy of circuit breaker operation mechanism fault diagnosis based on the LSTM algorithm can be further improved, and the circuit breaker operation mechanism can be reasonably and reliably diagnosed.

Description

technical field [0001] The invention belongs to the technical field of circuit breaker mechanical fault identification, and in particular relates to a circuit breaker mechanical fault identification method based on DS evidence fusion, and also relates to a circuit breaker mechanical fault identification device based on DS evidence fusion. Background technique [0002] The high-voltage circuit breaker is an important control and protection device for the power grid, and it is also a switching device that can cope with the most diverse tasks and requirements in the power system. Once a fault occurs, it will cause serious grid operation accidents. Therefore, how to timely and accurately discover the defects and faults of the high-voltage circuit breaker and ensure the normal operation of the circuit breaker has become a major problem in the power system. [0003] Mechanical failure is the main failure type of high-voltage circuit breakers. According to relevant statistics, mech...

AI Technical Summary

Problems solved by technology

On the one hand, for any single fault identification algorithm, there are poor generalization ability, extreme dependence on the characteristics of training samples, and poor accuracy of circuit breaker fault diagnosis performance, which limits the further application of related diagnostic methods

On the other hand, without considering the conflict of relevant diagnosis results, the identification results of multiple independent classification algorithms cannot be well integrated, which further limits the improvement of the accuracy of mechanical fault identification.

Images