Transformer winding vibration signal identification method based on ITD permutation entropy and CGWO-SVM

Abstract

The invention discloses a transformer winding vibration signal identification method based on ITD permutation entropy and CGWO-SVM. The transformer winding vibration signal identification method comprises the steps that vibration signals of a transformer winding in normal and fault states in the switching process are collected; and ITD is adopted to decompose a vibration signal, and permutation entropy is utilized to perform complex calculation on the decomposed signal to extract a feature vector, and the feature vector is used as an input quantity of CWGO-SVM for training, and the feature vector is compared with GWO-SVM and SVM. The transformer winding vibration signal identification method can effectively extract the fault features of the transformer winding, is high in diagnosis resultprecision, achieves the intelligent diagnosis of the fault of the transformer winding, can find the fault at the early stage of a latent fault, is higher in operability, and is superior to a conventional SVM in technical effect.

Description

technical field [0001] The invention belongs to the technical field of transformer winding fault diagnosis, and in particular relates to a transformer winding vibration signal recognition method based on ITD permutation entropy and CGWO-SVM. Background technique [0002] In recent years, my country's power grid is gradually moving towards ultra-high voltage, long-distance large-capacity power transmission, and smart grids. The interrelationships within the system are increasingly diversified, compounded, and huge, and the requirements for the continuous and stable operation of individual links in the system are increasingly demanding. strict. Ensuring that every single link in the power system does not fail is a prerequisite for the overall structure to not fail, which is also an indispensable part of building a strong smart grid. According to the national power grid operation safety assessment report of the State Grid Corporation of my country for a total of ten years from ...

AI Technical Summary

Problems solved by technology

[0004] Because the diagnostic accuracy of existing transformer winding fault detection methods is not high enough, and the diagnostic effect needs to be improved, it is urgent to realize the intelligent diagnosis of transformer winding faults with high precision, and to detect faults in the early stage of latent faults, and the method needs to be more advanced. strong maneuverability

Images