Fault diagnosis method of convertor based on time convolution network

Abstract

The invention relates to a fault diagnosis method of a power electronic convertor based on a time convolution network technology. The fault diagnosis method comprises the following steps that S1, an electrical signal of a measurement point is collected and treated by noise reduction, and sample data with fault information are obtained; S2, the dimension of the sample data is reduced by adopting normalization, and obtained fault characteristics and fault types are corresponded one by one to establish a data sample base; S3, a fault classifier based on the time convolution network is constructed, and training and testing are conducted according to the data sample base to obtain optimal network structure parameters; S4, a fault classifier based on the time convolution network is reconstructedaccording to the optimal network structure parameters, and the fault classifier with optimal parameters is obtained; and S5, the fault classifier with the optimal parameters is written into simulink,and real-time fault diagnosis and location are conducted on the power electronic convertor in actual operation. According to the fault diagnosis method, the health condition of the convertor can be judged more accurately and reliably.

Description

Technical field [0001] The invention relates to the technical field of power electronics, in particular to a fault diagnosis method of a converter based on a time convolutional network. Background technique [0002] With the advent of the Industry 4.0 era, power electronics technology has been more widely used in various fields of production and life. Accordingly, power electronics fault diagnosis technology is also indispensable. [0003] First of all, power electronic converters are mostly used as control equipment or core power sources. If the failure type is not scientifically diagnosed, it will only raise the carbuncle legacy and affect the self-isolation and self-recovery of the device failure. Moreover, as the scope of faults expands and functional failures increase, there are great safety hazards. [0004] Second, as the complexity of power electronic equipment increases, maintenance costs are increasing day by day. The converter, the main body of energy conversion in power...

AI Technical Summary

Problems solved by technology

The support vector machine is relatively simple in calculation, but because it is susceptible to the noise of the sampling signal, it will cause misjudgment of the output result

The fault dictionary has strong anti-interference ability, but it needs a large fault sample to achieve good results

Images