Scholarlink

<p>Machining of hardened SKD11 tool steel is a challenging task due to its high hardness and strength, which accelerates tool wear, increases cutting temperature, and deteriorates surface quality. Conventional cooling strategies are often insufficient to effectively control heat generation under high-speed milling conditions. To address these limitations, this study proposes an integrated hybrid machining strategy that combines ultrasonic-assisted milling with through-tool internal cooling to enhance heat dissipation and reduce friction at the cutting interface. The high-frequency, low-amplitude ultrasonic vibration induces intermittent tool-workpiece contact, consequently improving chip evacuation and minimizing adhesion, while internal cooling delivers coolant directly into the cutting zone for efficient temperature control. Comparative experiments conducted under Z-axis, XY-plane, and combined XYZ ultrasonic modes demonstrated that Z-axis ultrasonic assistance combined with internal cooling achieved the best performance, reducing surface roughness by 20–31% and cutting temperature by 15–28% compared to conventional milling. Additionally, multiple machine learning, models, including Decision Tree (DT), Random Forest (RF), and Adaptive Boosting (AdaBoost), were used to predict cutting tool wear. Among these, AdaBoost achieved the highest accuracy and precision (0.964) and F1-score (0.944), and its performance was further enhanced through a voting ensemble strategy. Unlike previous studies which considered vibration-assisted machining and data-driven monitoring method, this study integrates hybrid machining and intelligent prediction within a unified framework for hardened SKD11 tool steel, thereby contributing to smarter and more sustainable manufacturing.</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

Hybrid ultrasonic-assisted milling with internal cooling and ensemble machine learning for cutting-tool wear prediction in SKD11 hardened steel

  • 期刊论文

  • 发表日期：

    2026年3月30日

查看全文
  • Shen-Yung Lin,
  • Zahid Abbas Shah,
  • Hong-Xuan Peng

摘要

Machining of hardened SKD11 tool steel is a challenging task due to its high hardness and strength, which accelerates tool wear, increases cutting temperature, and deteriorates surface quality. Conventional cooling strategies are often insufficient to effectively control heat generation under high-speed milling conditions. To address these limitations, this study proposes an integrated hybrid machining strategy that combines ultrasonic-assisted milling with through-tool internal cooling to enhance heat dissipation and reduce friction at the cutting interface. The high-frequency, low-amplitude ultrasonic vibration induces intermittent tool-workpiece contact, consequently improving chip evacuation and minimizing adhesion, while internal cooling delivers coolant directly into the cutting zone for efficient temperature control. Comparative experiments conducted under Z-axis, XY-plane, and combined XYZ ultrasonic modes demonstrated that Z-axis ultrasonic assistance combined with internal cooling achieved the best performance, reducing surface roughness by 20–31% and cutting temperature by 15–28% compared to conventional milling. Additionally, multiple machine learning, models, including Decision Tree (DT), Random Forest (RF), and Adaptive Boosting (AdaBoost), were used to predict cutting tool wear. Among these, AdaBoost achieved the highest accuracy and precision (0.964) and F1-score (0.944), and its performance was further enhanced through a voting ensemble strategy. Unlike previous studies which considered vibration-assisted machining and data-driven monitoring method, this study integrates hybrid machining and intelligent prediction within a unified framework for hardened SKD11 tool steel, thereby contributing to smarter and more sustainable manufacturing.