A novel method for predicting cutting force considering tool wear by power consumption in milling process

Accurate prediction of cutting forces is crucial for machining process monitoring, parameter optimization, and tool wear management. Traditional dynamometer-based methods are limited by cost and applicability constraints, especially under complex cutting conditions. This study proposes an innovative indirect method for predicting cutting forces by integrating tool wear considerations into a power consumption model. A mechanistic model incorporating shearing and edge forces affected by flank wear is developed, with power consumption further characterized by cutting parameters and tool wear conditions. A simulated annealing (SA) optimization algorithm is employed to identify key model parameters, enabling the accurate prediction of instantaneous cutting forces. Experimental validation, conducted under varying spindle speeds, feed rates, and cutting depths, demonstrates the effectiveness of the proposed method. The mean tool wear and cutting force prediction errors are maintained below 11 μm and 20 N, respectively. Comparative analysis of search strategies highlights the superior performance of exponential cooling strategy in SA optimization, ensuring convergence efficiency and prediction accuracy. Furthermore, the model's robustness is validated across novel cutting conditions and independent datasets. This methodology offers a cost-effective alternative to conventional monitoring systems by leveraging machine power signals, enhancing process adaptability while ensuring reliability. The proposed method significantly contributes to machining precision, proactive maintenance, and real-time process optimization. Future work will expand its applicability to multi-axis milling and complex surfaces, addressing challenges in instantaneous force prediction under diverse cutting scenarios.