Scholarlink

<p>With the ongoing increase in demand for lightweight products within the manufacturing industry, thin-walled components are increasingly being utilized across various engineering fields, accompanied by increasingly stringent requirements for machining accuracy. However, due to the inherently low structural rigidity of thin-walled components, they are highly susceptible to various complex factors during the cutting process, which can significantly affect both machining accuracy and surface quality. This paper proposes the instantaneous milling force prediction model for the side milling of thin-walled components investigating three key influencing factors: tool runout, workpiece deformation, and tool vibration. Firstly, a geometric parameter model is established based on the trochoidal motion principle of the tool, taking into account the coupled interaction between tool runout and workpiece deformation. Subsequently, a tool vibration model is constructed to describe the relationship between actual and ideal cutting thickness. By integrating cutting force coefficients calibrated using the average milling force method, an instantaneous milling force prediction model tailored for side milling of thin-walled components is established. Under 27 groups experimental conditions, the proposed model achieved average improvements in X direction of 0.551% and 0.325% in the overall error indicators <i>RMSE</i> and <i>MAPE</i> compared to the two-factor model, and 3.002% and 2.491% in Y direction, respectively. In the context of peak force prediction, which is of particular significance in thin-walled components machining, the proposed model in the X-direction demonstrated improvements of 7.52% and 3.022% in <i>RMSE</i><sub><i>peak</i></sub> and <i>MAPE</i><sub><i>peak</i></sub>, respectively, when compared to the two-factor model. In the Y-direction, the improvements were approximately 6.985% and 8.715%, respectively. The experimental results indicate that the proposed model achieves comparable overall error metrics to the two-factor model, while demonstrating superior stability in peak force forecasting. Consequently, it delivers significantly enhanced prediction accuracy in the machining of thin-walled components.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Instantaneous milling force prediction for side milling of thin-walled components considering tool runout, workpiece deformation and tool vibration

  • 期刊论文

  • 发表日期：

    2026年1月8日

查看全文
  • Chunhua Feng,
  • Jiaqi Chen,
  • Weidong Li,
  • Jinhuan Su

摘要

With the ongoing increase in demand for lightweight products within the manufacturing industry, thin-walled components are increasingly being utilized across various engineering fields, accompanied by increasingly stringent requirements for machining accuracy. However, due to the inherently low structural rigidity of thin-walled components, they are highly susceptible to various complex factors during the cutting process, which can significantly affect both machining accuracy and surface quality. This paper proposes the instantaneous milling force prediction model for the side milling of thin-walled components investigating three key influencing factors: tool runout, workpiece deformation, and tool vibration. Firstly, a geometric parameter model is established based on the trochoidal motion principle of the tool, taking into account the coupled interaction between tool runout and workpiece deformation. Subsequently, a tool vibration model is constructed to describe the relationship between actual and ideal cutting thickness. By integrating cutting force coefficients calibrated using the average milling force method, an instantaneous milling force prediction model tailored for side milling of thin-walled components is established. Under 27 groups experimental conditions, the proposed model achieved average improvements in X direction of 0.551% and 0.325% in the overall error indicators RMSE and MAPE compared to the two-factor model, and 3.002% and 2.491% in Y direction, respectively. In the context of peak force prediction, which is of particular significance in thin-walled components machining, the proposed model in the X-direction demonstrated improvements of 7.52% and 3.022% in RMSEpeak and MAPEpeak, respectively, when compared to the two-factor model. In the Y-direction, the improvements were approximately 6.985% and 8.715%, respectively. The experimental results indicate that the proposed model achieves comparable overall error metrics to the two-factor model, while demonstrating superior stability in peak force forecasting. Consequently, it delivers significantly enhanced prediction accuracy in the machining of thin-walled components.