Can hidden Markov models (HMMs) effectively monitor tool wear in machining processes? This paper presents a new modeling framework for tool wear monitoring in turning using HMMs. Feature vectors are extracted from vibration signals measured during turning, and vector quantization is used to convert these features into a symbol sequence for the HMM. The research explores the effectiveness of this approach for different lengths of training data and observation sequences. A series of experiments were conducted to evaluate the performance of the HMM-based tool wear monitoring system. The experimental results demonstrated successful tool state detection rates as high as 97%, indicating the effectiveness of the approach. The method offers a new way to detect tool states within machining processes. In conclusion, this research demonstrates the potential of HMMs for tool wear monitoring in machining processes. With high tool state detection rates, this approach offers a valuable tool for optimizing machining operations and improving manufacturing efficiency. The method appears to be very effective with the results of the report.
Published in the Journal of Manufacturing Science and Engineering, this paper aligns with the journal's focus on advancing manufacturing processes and technologies. By presenting a new modeling framework for tool wear monitoring, the paper contributes to the journal's mission of improving manufacturing efficiency and quality through innovative engineering solutions.