研究了粒子群优化改的进的模糊线性PSVM在旋转机械故障诊断的应用.常规的PSVM对噪声或野值敏感,模糊PSVM可以很好的解决这种问题;对于非平衡样本,PSVM分类面会偏重于数据点较多的一类,从而降低正确分类性能,通过为不同样本分别设计不同的惩罚因子,提高分类器性能;模糊线性PSVM分类器的惩罚因子采用经典粒子群优化算法进行优化,避免传统方法对初始点和样本的依赖.通过旋转机械故障分类应用实例进行了设计方法的验证,首先对振动信号进行滤波,然后以不同频率频谱的谱峰能量作为模糊线型PSVM分类器的输入特征参数,用于区分旋转机械的5种典型故障,试验结果表明了方法的有效性.%Based on particle swarm optimization (PSO) and fuzzy linear proximal support vector machine (FLPSVM), a rotating machinery fault diagnosis method was proposed. Fuzzy PSVM can solve the question that standard PSVM is sensitive to outliers and noises in training sets. When PSVM is applied to the problem with two classes on unbalanced datasets, it tends to fit better the class with more data points. This leads to the poor classification performance. Different penalty factors were designed for different samples in order to improve the classification performance. The penalty factors of FLPSVM were optimized by the canonical particle swarm optimization to avoid the dependence on initial parameters and training samples. The rotating machinery fault-classification data were used to demonstrate the designed method. The vibration signals were filtered by the filters at first and then the energy at spetral peaks of different frequencies was taken as the input feature parameters of FLPSVM classifier to identify the five typical rotating machinery faults. The experiment results demonstrate that the modeling method is correct and precise.
展开▼
