Scholarlink

<p>The deflection of diamond wire saws—commonly referred to as wire bow—during stone machining significantly compromises the geometric accuracy of curved surfaces and corner regions, often resulting in the rejection of workpieces. To address this issue, we propose a systematic parameter optimization strategy that minimizes wire bow while maintaining a high feed rate. Initially, the relationship between machining parameters and wire deflection is theoretically established using a macroscopic mechanical model. A laser-based measurement system is then developed to quantify three novel evaluation metrics: average deflection (<i>D</i>), maximum deflection (<i>H</i>), and cross-sectional deformation area (<i>S</i>). The Taguchi method and analysis of variance (ANOVA) are employed to determine the primary effects of machining parameters on these metrics. Three machine learning models—generalized regression, backpropagation neural network (BPNN), and support vector regression (SVR)—are used to predict D, with the SVR model achieving the highest predictive accuracy (R<sup>2</sup> = 0.984). Based on these predictive models, an optimal parameter combination is derived and experimentally validated. The results show that, under a feed rate of 12 mm/min, the optimized parameters reduce corner-cutting error by 49.62% compared to conventional settings. This work presents a data-driven framework for enhancing the precision of stone machining using diamond wire saws.</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

Process parameter optimization for diamond wire saw machining of stone

  • 期刊论文

  • 发表日期：

    2025年5月9日

查看全文
  • Zhen Zhang,
  • Fangchen Yin,
  • Guoqin Huang,
  • Changcai Cui

摘要

The deflection of diamond wire saws—commonly referred to as wire bow—during stone machining significantly compromises the geometric accuracy of curved surfaces and corner regions, often resulting in the rejection of workpieces. To address this issue, we propose a systematic parameter optimization strategy that minimizes wire bow while maintaining a high feed rate. Initially, the relationship between machining parameters and wire deflection is theoretically established using a macroscopic mechanical model. A laser-based measurement system is then developed to quantify three novel evaluation metrics: average deflection (D), maximum deflection (H), and cross-sectional deformation area (S). The Taguchi method and analysis of variance (ANOVA) are employed to determine the primary effects of machining parameters on these metrics. Three machine learning models—generalized regression, backpropagation neural network (BPNN), and support vector regression (SVR)—are used to predict D, with the SVR model achieving the highest predictive accuracy (R2 = 0.984). Based on these predictive models, an optimal parameter combination is derived and experimentally validated. The results show that, under a feed rate of 12 mm/min, the optimized parameters reduce corner-cutting error by 49.62% compared to conventional settings. This work presents a data-driven framework for enhancing the precision of stone machining using diamond wire saws.