论文部分内容阅读
针对故障诊断中传统的线性多变量统计分析方法不能解决线性不可分问题,提出了一种基于核判别分析的非线性特征约简方法。在核判别分析中,核函数决定了其非线性映射的能力;为此,提出了一种基于K均值聚类的核函数参数优化方案,并将该方法应用于滚动轴承的故障诊断中。结果表明,与主成分分析及线性判别分析相比,核判别分析能够更有效地区分轴承的4种状态,适用于故障诊断中的非线性特征约简。
In view of the fact that the traditional linear multivariate statistical analysis method in fault diagnosis can not solve the problem of linear inseparability, a nonlinear feature reduction method based on kernel discriminant analysis is proposed. In the kernel discriminant analysis, the kernel function determines its nonlinear mapping ability. To solve this problem, a kernel function parameter optimization scheme based on K-means clustering is proposed. The proposed method is applied to the fault diagnosis of the rolling bearing. The results show that compared with the principal component analysis and linear discriminant analysis, nuclear discriminant analysis can distinguish the four states of bearings more effectively and is suitable for the nonlinear feature reduction in fault diagnosis.