A wavelet neural network diagnosis model based on the chaotic artificial bee colony algorithm with elite swarm search strategy is proposed to accommodate to complicated fault types of transformer

and the analysis method of gases in oil is combined with the model. The proposed model is based on the three-layer wavelet neural network structure and the weighting coefficients of input and output

as well as the stretching and translation coefficients of wavelet elements. The continuous optimization algorithm is based on the elite swarm search strategy

which simulates the foraging behavior of worker bees and follower bees. Meanwhile

the search mechanism of scout bees and the initialization of honey source are enhanced by constructing chaos strategy. Experimental results show that the proposed method is more applicable to the transformer fault diagnosis for its faster and more accurate search results than those obtained by conventional BP wavelet neural network diagnosis and wavelet neural network diagnosis optimized by standardbeecolonyalgorithm.