Improvement of Cutting Force Estimation by Lead Error Compensation for Machine Tool

There is a considerable demand for cutting force estimation without additional sensors to improve the cutting process by machine tools. Some methods of cutting force estimation based on disturbance observer (DOB) have been developed. However, some problems still exist in their practical application to cutting force estimation. This study aims to clarify and resolve the problems. First, two cutting force estimation methods based on the two-inertia model of the feed system of the machine tool are compared. Because the two methods are based on DOB and perform time differentiation of the encoder signal, they are vulnerable to the quantization error. Therefore, the two methods are analyzed, in terms of their robustness to the quantization error, based on the discrete Lyapunov equation. The method that is tolerant to the quantization error is chosen for further analysis. The problem of the chosen method is the lead error, which is not considered in the two-inertia model. This lead error degrades the accuracy of the cutting force estimation. Therefore, identification and compensation methods for tackling the lead error are proposed. Experiments with an actual industrial machine tool are performed, and the presented method is validated.