Enhanced particle filter for tool wear prediction

• A probabilistic tool wear model is derived to facilitate the system model construction in particle filter.
• An enhanced particle filter method is proposed by integrating the regressive analysis in the prediction stage of particle filter to improve the prediction accuracy.
• The presented new probabilistic tool wear prediction method takes advantage of physical knowledge and in-process measurement in one framework to account for uncertainty and nonlinearity in tool wear process.
• Two hybrid approaches (e.g. integrative particle filter and autoregressive model, and integrative particle filter and support vector regression) have been investigated and show better prediction performance than conventional particle filter.

Timely assessment and prediction of tool wear is essential to ensuring part quality, minimizing material waste, and contributing to sustainable manufacturing. This paper presents a probabilistic method based on particle filtering to account for uncertainties in the tool wear process. Tool wear state is predicted by recursively updating a physics-based tool wear rate model with online measurement, following a Bayesian inference scheme. For long term prediction where online measurement is not available, regression analysis methods such as autoregressive model and support vector regression are investigated by incorporating predicted measurement into particle filter. The effectiveness of the developed method is demonstrated using experiments performed on a CNC milling machine.