Fault Diagnosis for Transmission System of Large Equipment Using Nonlinear Output Frequency Response Function

The fault diagnosis of the transmission system of large numerical control equipment is studied based on the nonlinear output frequency response function (NOFRF). A frequency domain variable step size least mean square adaptive algorithm is used to estimate NOFRF and get nonlinear spectral data. After obtaining nonlinear frequency spectrum data, the spectrum features are extracted by kernel principal component analysis(KPCA) method. To consider global characteristics and local characteristics of nonlinear spectrum, a kind of mixed function is used as kernel function of KPCA model. According to different frequency domain scale characteristics, a multi-fault classifier based on support vector machine (SVM) fusion is constructed, and a classification reliability-based fusion method is adopted.