Prediction of cutting force via machine learning: state of the art, challenges and potentials

• Published in: Journal of intelligent manufacturing Vol. 36; no. 2; pp. 703 - 764
• Main Authors: Liu, Meng, Xie, Hui, Pan, Wencheng, Ding, Songlin, Li, Guangxian
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.02.2025; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Artificial intelligence; Business and Management; Control; Cutting force; Cutting parameters; Cutting wear; Cuttings; Finite element analysis; Finite element method; Machine learning; Machines; Machining; Manufacturing; Mathematical models; Mechatronics; Metal cutting; Numerical models; Predictions; Processes; Production; Real time; Robotics; Surface stability; Tool wear; Machine learning (ML); Accuracy; Typical models; Force prediction; Efficiency; Long-term prediction
• ISSN: 0956-5515; 1572-8145; 1572-8145
• DOI: 10.1007/s10845-023-02260-8

Cutting force is a critical factor that reflects the machining states and affects tool wear, cutting stability, and the quality of the machined surface. Accurate prediction of cutting force has been the subject of extensive research in machining technology for decades. Generally, the predicting methods are based on the physical principles of metal cutting processes and they can be divided into two main categories: calculation of cutting forces by using analytical models and numerical simulation of cutting forces with finite element analysis. With the advance of artificial intelligence and machine learning (ML), various algorithms have been developed to predict cutting force with high accuracy and high efficiency. This paper provides a comprehensive review of force prediction methods, with a focus on ML-based algorithms. The mechanisms and characteristics of various force prediction methods, such as analytical models and finite element analysis, as well as different ML-based algorithms, are introduced in detail. The challenges of current algorithms and their potential in long-term and real-time prediction are discussed. The review highlights the potential of ML-based algorithms in improving the accuracy and efficiency of cutting force prediction and emphasizes the need for further research to address the current challenges and advance the field of force prediction in metal-cutting processes.