Online diagnosis method for mechanical fault of high-voltage circuit breaker

Abstract

The invention provides an online diagnosis method for a mechanical fault of a high-voltage circuit breaker, which comprises the steps of S1, initializing an online diagnosis model; S2, for a new sample, acquiring a feature vector by using a pretreatment method, putting the feature vector into the online diagnosis model to calculate a discrimination function value and a fault diagnosis value; S3, inputting the discrimination function value and the fault diagnosis value of the new sample into an update triggering module to carry out calculation, judging whether an update triggering condition is satisfied or not, switching to a step S4 if the update triggering condition is satisfied, and resetting an update triggering counter; outputting a model diagnosis result directly if the update triggering condition is not satisfied so as to act as a final diagnosis result; and S4, updating the online diagnosis model, and switching to the step S2 so as to carry out diagnosis again on the current sample. According to the online diagnosis method, a historical training result can be fully utilized, the storage space requirement and the calculation time of the model can be greatly reduced, the calculation efficiency is improved, and online update of the fault diagnosis model is realized, thereby being better applied to mechanical fault diagnosis for equipment.

Description

technical field [0001] The invention relates to the technical field of power systems, in particular to an online diagnosis method for mechanical faults of high-voltage circuit breakers. Background technique [0002] High-voltage circuit breaker is the key main equipment in the power system, which undertakes the control and protection tasks of the power system. The International Large Power Grid Conference has conducted two worldwide surveys on the reliability of high-voltage circuit breakers, and my country has also conducted a large number of statistical analyzes on high-voltage switch accidents. Corresponding surveys and statistical reports show that most of the failures of high-voltage circuit breakers are the problems of the operating mechanism. The diagnosis of mechanical life faults of high-voltage circuit breakers is of great significance not only for the safety and reliability of circuit breaker operation, but also for the life cycle management of power grid assets....

AI Technical Summary

Problems solved by technology

[0004] The inventors have found that the above three evaluation methods all have deficiencies. The disadvantages are: in the first and second methods, the fault judgment is a relatively mechanical judgment method. Once the parameters are fixed, the It cannot be changed and cannot adapt to complex and changeable fault situations in the field, and misdiagnosis or missed diagnosis may occur; the third diagnosis method based on machine learning methods requires a fault library and trains a diagnostic model for diagnosis

However, in the current big data environment, with the development of the Internet of Things and online monitoring technology, a large amount of new data is generated in the entire system every day, which often contains new failure modes and personalized failure information of equipment, which cannot be fixed in the database The trained model accurately recognizes

If data expansion is performed on the fault database to update the model, due to the lack of effective data screening rules, the computational complexity of the model often increases exponentially with the addition of a large amount of data, resulting in difficulties in model storage and solution, affecting the model diagnostic accuracy

Images