基于精英混沌蜂群算法优化小波神经网络的变压器故障诊断

Transformer Fault Diagnosis Using Wavelet Neural Network Based on Elite⁃chaos Artificial Bee Colony Algorithm

针对变压器故障类型的复杂难辨,结合油中气体分析法,提出了1种基于精英混沌人工蜂群算法的小波神经网络故障诊断模型对变压器故障进行诊断。模型对3层小波神经网络的输入和输出层的权值、小波元的伸缩和平移系数进行修正,改进算法混沌初始化蜜源,建立精英蜂群团体自适应引导采蜜蜂与跟随蜂搜索,同时引入混沌侦查蜂的行为模式,能够有效提高精度,加速收敛,避免局部最优。实验结果表明,该方法比传统BP小波神经网络和标准蜂群算法优化小波神经网络诊断速度更快,准确率更高,可有效运用于变压器故障诊断。

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 ini⁃tialization 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 standard bee colony algorithm.