In order to improve the accuracy of fault diagnosis of transformer

a kind of improved hybrid shuffled frog leaping algorithm is proposed to optimize the model parameters of support vector machine for transformer fault diagnosis in this paper. Firstly

the combination method of uniform randomization and anti-approximate opposition are adopted for the improvement of the population initialization of shuffled frog leaping algorithm in order to improve the quality of initial solution. Secondly

the differential mutation operator is introduced

and the cosine adaptive skeleton differential evolution algorithm is embedded into the local search algorithm of shuffled frog leaping algorithm and

at the same time

Levy mutation operator is used instead of random update operation

which improves the accuracy of the local search algorithm and accelerates the convergence speed. Finally

the improved hybrid shuffled frog leaping algorithm is used to optimize the model parameters of support vector machine and set up the fault diagnosis model.The simulation results of both classical test functions and example verification results show that the performance of the improved hybrid shuffled frog leaping algorithm proposed in this paper is greatly improved. Compared with the optimization result of SFLA

AS-SFLA and ODSFLA algorithm

the optimized support vector machine fault diagnosis model has higher diagnosis accuracy

faster iteration speed

and can effectively diagnose fault of transformer.