Mechanical Fault Diagnosis of Transformer Core Based on VMD and GWO-HKELM

Aiming at the problem that it is difficult to accurately identify the mechanical fault of the transformer core, a method of transformer core mechanical fault diagnosis based on variational mode decomposition (VMD)-multi-scale permutation entropy (MPE) and a gray wolf optimization algorithm to optimize the parameters of hybrid kernel extreme learning machine (GWO-HKELM) is proposed. Firstly, VMD is used to decompose the transformer vibration signal and select multiple IMF components that are highly correlated with the original vibration signal. Secondly, we calculate the MPE values of the IMF components and use them to construct a feature dataset. Furthermore, GWO is used to optimize the key kernel parameters of HKELM and establish a GWO-HKELM diagnostic model. Then, the feature dataset is imported into the GWO-HKELM diagnostic model for recognition. Finally, the effectiveness and superiority of the suggested method are verified through experimental data and comparison with other methods. The experimental results show that the suggested method can effectively identify different types and degrees of mechanical faults caused by transformer core looseness, with a diagnostic accuracy of over 98.0%. It can provide theoretical guidance for the formulation of on-site maintenance strategies for transformers.