An online chatter detection and recognition method for camshaft non-circular contour high-speed grinding based on improved LMD and GAPSO-ABC-SVM

The camshaft is a crucial part of the engine. However, its non-circular contour surface is prone to chatter in high-speed grinding, seriously affecting the processing quality and efficiency. Therefore, an online detection and recognition method for camshaft non-circular contour high-speed grinding chatter based on improved LMD and GAPSO-ABC-SVM is proposed. Firstly, the local mean decomposition (LMD) algorithm is improved by the mirror extension method, moving average algorithm, and adaptive soft screening stopping criterion. Its ability to deal with unsteady vibration signals is verified by simulation signals and experiments. Then, considering the influence of the curvature change of the non-circular contour grinding surface on the chatter features, the signal features are automatically extracted according to the contour curve characteristics. Finally, a recognition algorithm based on GAPSO-ABC-SVM is proposed to improve the accuracy and robustness of high-speed grinding chatter recognition. A new hybrid swarm intelligent optimization algorithm is proposed through the intelligent fusion of Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Artificial Bee Colony (ABC) algorithms. The support vector machine (SVM) optimization is implemented by the hybrid swarm intelligence algorithm. In the high-speed grinding chatter verification experiment of camshaft non-circular contour, the detection and recognition method based on improved LMD and GAPSO-ABC-SVM can achieve an accuracy of 97.917 % for chatter recognition. And it has good fault tolerance.