Scholarlink

<p>This study investigates the precision cutting of hybrid material stacks comprising Ti-6Al-4V, CFRP, and Al7075 using Abrasive Water Jet Machining (AWJM). These materials are frequently utilized in aerospace and automotive sectors because of their superior mechanical properties. The research focuses on optimizing key AWJM parameters, including jet pressure (<i>P</i>), stand-off distance (<i>SOD</i>), and nozzle speed (<i>V</i>) to achieve high-quality kerf cuts through multi-objective optimization (MOO). The objectives include minimizing kerf taper angle (<i>θ</i><sub><i>K</i></sub>) and surface roughness (<i>SR</i>) while maximizing material removal rate (<i>MRR</i>) across all layers. A Bayesian Artificial Neural Network (ANN) model with a (3-17-9) architecture was developed to predict cutting performance, effectively capturing complex non-linear interactions between process parameters and responses. Non-dominated Sorting Genetic Algorithm-II (NSGA-II) was utilized to generate Pareto-solutions, which were further ranked using Entropy-TOPSIS technique to identify the optimal parameter combination and evaluate parameter significance. The results demonstrated the ANN model’s high prediction accuracy (<i>R</i> ~ 1) and the Pareto front’s ability to balance trade-offs among <i>θ</i><sub><i>K</i></sub>, <i>SR</i>, and <i>MRR</i> for hybrid stacks. Optimal machining conditions were identified as <i>P</i> = 350.6 MPa, <i>V</i> = 297.5&#xa0;mm/min, and <i>SOD</i> = 7&#xa0;mm, with water jet pressure contributing the most significant influence (53.31%) on cutting performance. This integrated framework combining ANN, NSGA-II, and Entropy-TOPSIS offers a robust and systematic approach for multi-dimensional optimization in hybrid material machining, enhancing both efficiency and precision in advanced manufacturing applications.</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

A hybrid machine learning approach to optimize kerf quality in AWJM of Ti-6Al-4V/CFRP/Al7075 stacks

  • 期刊论文

  • 发表日期：

    2025年9月9日

查看全文
  • Mohamed Abouhawa,
  • Abdulaziz Elshinawi,
  • Abeer Eisa,
  • Mohamed Fattouh

摘要

This study investigates the precision cutting of hybrid material stacks comprising Ti-6Al-4V, CFRP, and Al7075 using Abrasive Water Jet Machining (AWJM). These materials are frequently utilized in aerospace and automotive sectors because of their superior mechanical properties. The research focuses on optimizing key AWJM parameters, including jet pressure (P), stand-off distance (SOD), and nozzle speed (V) to achieve high-quality kerf cuts through multi-objective optimization (MOO). The objectives include minimizing kerf taper angle (θK) and surface roughness (SR) while maximizing material removal rate (MRR) across all layers. A Bayesian Artificial Neural Network (ANN) model with a (3-17-9) architecture was developed to predict cutting performance, effectively capturing complex non-linear interactions between process parameters and responses. Non-dominated Sorting Genetic Algorithm-II (NSGA-II) was utilized to generate Pareto-solutions, which were further ranked using Entropy-TOPSIS technique to identify the optimal parameter combination and evaluate parameter significance. The results demonstrated the ANN model’s high prediction accuracy (R ~ 1) and the Pareto front’s ability to balance trade-offs among θK, SR, and MRR for hybrid stacks. Optimal machining conditions were identified as P = 350.6 MPa, V = 297.5 mm/min, and SOD = 7 mm, with water jet pressure contributing the most significant influence (53.31%) on cutting performance. This integrated framework combining ANN, NSGA-II, and Entropy-TOPSIS offers a robust and systematic approach for multi-dimensional optimization in hybrid material machining, enhancing both efficiency and precision in advanced manufacturing applications.