基于混沌步长果蝇算法优化支持向量机的齿轮故障诊断

Gear Fault Diagnosis Based on SVM Optimized by Chaotic Step Fruit Fly Optimization Algorithm

为提高齿轮的故障诊断精度,同时针对果蝇算法(FOA)用于优化支持向量机(SVM)参数时存在陷入局部最优的不足,对FOA算法中搜索步长的确定方式进行了改进,以此提出了基于混沌步长果蝇算法(CSFOA)的齿轮故障诊断方法。齿轮故障诊断实例结果表明,CSFOA优化得到的SVM参数比FOA更优,有效地提升了齿轮故障诊断的精度,同其它一些方法的比较也证明了CSFOA的优势。

In order to improve fault diagnosis accuracy of gears,also aiming at the problem that the fruit fly optimization algorithm(FOA)falls into local optimum when optimizing the parameters of support vector machine(SVM),the way to confirm search step was improved in this paper and a gear fault diagnosis method based on SVM optimized by chaotic step fruit fly optimization algorithm(CSFOA)is proposed.The results of gear diagnosis showed that the parameters optimized by CSFOA is better than FOA,and the gear diagnosis accuracy got an effective promotion.Meanwhile,comparison results with some other optimization algorithms also showed that the method has certain advantages.