Ball screws need to be applied with different preloads in most applications, however, very few studies had paid attention to preload detection, which is one of the most common failure modes of ball screws. To detect the preload of a ball screw, we propose a novel approach using vibration signals and ensemble empirical mode decomposition (EEMD) combined with linear discriminant analysis (LDA) and a support vector machine (SVM). The vibration signals under different preloads are obtained for the same ball screw through degradation experiments, and decomposed and reconstructed by EEMD. Statistical features are extracted from the reconstructed signals in the time and frequency domains. Linear discriminant analysis is used to reduce the dimensionality of the original feature space. Finally, the preload can be identified by the SVM, in which the identification accuracy can be higher than 90%. Further analysis showed that the high-frequency signal has a stronger correlation with preload degradation than the low-frequency signal, and the proposed method can be more effective than other existing methods, especially when the degradation of preload is more than 300N.

As a critical precision transmission element in numerical control (NC) machines, the linear rolling guide (LRG) suffers from surface wear degradation, which significantly impairs machining accuracy and operational reliability. Despite its importance, effective identification methods for LRG degradation remain limited. In this study, a hybrid approach combining multi-scale fuzzy entropy (MFE) with a gray wolf-optimized random forest (GWO-RF) algorithm was proposed to identify the surface wear state of the LRG. Preload degradation and vibration signals were collected at three surface wear stages throughout the LGR’s service life. The vibration signals were decomposed and reconstructed using complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), followed by multi-scale fuzzy entropy analysis of the reconstructed signals. After dimensionality reduction via kernel principal component analysis (KPCA), the processed features were fed into the GWO-RF model for classification. Experimental results demonstrated a recognition accuracy of 97.9%.