Power electronic converter fault diagnosis method and diagnosis system

Abstract

The invention provides a power electronic converter fault diagnosis method and diagnosis system. The method comprises the following steps: establishing a sample library matched with the running stateof the power electronic converter according to an obtained signal of the power electronic converter containing a plurality of sub-signals; obtaining samples serving as training data from the sample library, training a fault classifier based on a cerebellum model neural network algorithm through the samples, and implanting the trained fault classifier passing the fault recognition rate into a DSP for diagnosis; and acquiring a signal of a to-be-diagnosed power electronic converter, performing data preprocessing on the signal, inputting the signal into the DSP, and judging whether the to-be-diagnosed power electronic converter has a fault or not according to a processing result of the DSP. According to the method, the interference of other factors such as noise on single state information iseffectively reduced, and the defects of unreliability and uncertainty of single fault sample information used in a traditional method are overcome.

Description

technical field [0001] The present application belongs to the field of power electronics, and in particular relates to a fault diagnosis method and a diagnosis system of a power electronic converter. Background technique [0002] As the main device of power electronic AC-DC conversion, the power electronic converter plays an important role in the process of ensuring high-efficiency conversion of electric energy transmission. Once the power electronic converter fails during operation, it will seriously endanger the stability and reliability of the entire system. Therefore, the application of fault diagnosis technology to power electronic converters is of great significance for the stable operation of power systems to realize rapid diagnosis of faulty equipment when faults occur. [0003] The traditional power electronic converter fault diagnosis method is to collect a single state information of the converter, and extract the fault characteristics of the single information f...

AI Technical Summary

Problems solved by technology

Although the traditional method can also be used to diagnose faults, due to the interference of other factors such as noise in the collected single fault information, the uncertainty of the collected fault information is increased, and the fault diagnosis rate is often not reliable.

Images