基于改进FOA优化的SVM在故障诊断中的应用

OPTIMIZED SVM BASED ON IMPROVED FOA AND ITS APPLICATION IN FAULT DAIGNOSIS

针对支持向量机(SVM)的分类性能受本身参数选择影响较大的问题,提出了基于改进果蝇优化算法(LFOA)的SVM参数优化方法。给出了基于改进果蝇算法的SVM参数优化步骤,并用标准数据集进行了仿真实验,验证了算法在收敛速度和收敛精度上均好于其他几种方法。以滚动轴承为实验对象,应用LFOA-SVM进行了常见故障的诊断,与FOA、GA和PSO等方法相比,LFOA算法改善了SVM的分类性能,提高了故障诊断准确率,可有效应用于故障诊断。

Aiming at the fact that the classification performance of support vector machine （ SVM ） highly depends on the parameters selection, a parameters optimize method of SVM based on improved fruit fly optimization algorithm （LFOA） was proposed. The steps of SVM parameters optimize based on LFOA was proposed, and the superiority of the algorithm in convergence speed and convergence accuracy when compared with some other methods is verified by simulation experiment of several standard datasets. Take the rolling bearing as experiment object, the common faults was diagnosed by LFOA-SVM The experiment results show that the LFOA improved the classification performance of SVM and has higher accuracy compared with FOA, GA and PSO, and can applied to fault diagnosis efficiently.