Virtual sensing for gearbox condition monitoring based on kernel factor analysis

Abstract

Vibration and oil debris analysis are widely used in gearbox condition monitoring as the typical indirect and direct sensing techniques. However, they have their own advantages and disadvantages. To better utilize the sensing information and overcome its shortcomings, this paper presents a virtual sensing technique based on artificial intelligence by fusing low-cost online vibration measurements to derive a gearbox condition indictor, and its performance is comparable to the costly offline oil debris measurements. Firstly, the representative features are extracted from the noisy vibration measurements to characterize the gearbox degradation conditions. However, the extracted features of high dimensionality present nonlinearity and uncertainty in the machinery degradation process. A new nonlinear feature selection and fusion method, named kernel factor analysis, is proposed to mitigate the aforementioned challenge. Then the virtual sensing model is constructed by incorporating the fused vibration features and offline oil debris measurements based on support vector regression. The developed virtual sensing technique is experimentally evaluated in spiral bevel gear wear tests, and the results show that the developed kernel factor analysis method outperforms the state-of-the-art feature selection techniques in terms of virtual sensing model accuracy.

Failure Region Determination of Linear Voltage Regulator Using Data-driven and Model-based Virtual Sensing

“…This situation occurs because performing system modeling is intricate in these areas. Other fields also use DDA to implement the concept of virtual sensing; examples include automotive (to monitor vehicle gear condition), structural health monitoring, and aerospace [28]- [30]. However, virtual sensing is not commonly used in testing and characterizing electronic systems and components, such as in determining the failure region of a power management circuit.…”

KPCA-WPHM-SCNs-based remaining useful life prediction method for motor rolling bearings

“…Linear mapping is represented by principal component analysis (PCA) (Zhou and Bai, 2021) and linear discriminant analysis (LDA) (Xing et al, 2018). The main methods for nonlinear mapping are local linear embedding (LLE) (Xiao et al, 2018), isometric feature mapping (ISOMap) (Jin et al, 2017), kernel principal component analysis (KPCA) (Xiao et al, 2017), and so on. LLE uses the local linearity property on the manifold, represented by the relative linearity between finite local samples and find sample sets in the low-dimensional space that satisfy the constructive weights between high-dimensional samples.…”