A Switching Hidden Semi-Markov Model for Degradation Process and Its Application to Time-Varying Tool Wear Monitoring

Abstract

Hidden semi-Markov model (HSMM) has been widely used in equipment condition monitoring. However, the HSMM is usually modeled in fixed working mode. It is incompetent to monitor the condition when the working mode is varying in the equipment's lifetime. In this article, taking time-varying working mode into account, we propose a novel switching HSMM (SHSMM) to represent the equipment's degradation process. The reciprocal of duration is modeled and utilized to quantize the influence of working mode on the degradation process. Compared to traditional HSMM and time-varying HMM, the proposed SHSMM has a more generalized form and a more powerful ability to describe the degradation process with time-varying working mode. The proposed SHSMM is then applied to tool wear monitoring with time-varying cutting mode. Experimental results show that, via the proposed SHSMM, the monitoring confidence increases and the estimation of remaining useful life has a great improvement.