Scholarlink

<p>Additive manufacturing (AM) has become increasingly popular due to its potential to improve production efficiency and meet sustainable manufacturing requirements. However, surface issues and the efficiency of post-processing AM components are still multiple challenges to fulfilling industrial application requirements. This work aims to develop machine learning models based on the Adaptive neuro-fuzzy inference system (ANFIS) and advanced Non-Dominated Sorting Genetic Algorithm III method (NSGA-III) to improve the surface quality of AM products and the efficient post-processing. To achieve this, statistical analysis and a new understanding based on the interaction of the Electron beam melting (EBM) raster path and the cutting tool are proposed to detail the influence of turning parameters, tool types, and their interaction on turning performance: Surface finish, tool wear, and cutting forces during turning of cylindrical γ-TiAl specimens. After that, an artificial intelligence system based on ANFIS and advanced NSGA-III method was developed for monitoring, predicting, and optimizing the post-processing parameters. The results reveal that the cutting conditions and tool types significantly affect the turning performance. This study demonstrated that the ANFIS models could provide a more precise and dependable estimation model for monitoring and predicting flank and rake wear, surface quality, and cutting forces with fewer errors, 0.863%, 0.53%, 1.06%, 0.552%, 1.121%, and 0.6428%, respectively, compared with measured experiments. Finally, the developed intelligent method achieved the optimal turning conditions to enhance surface quality, reduce cutting forces, and reduce tool wear of printed γ-TiAl. Henceforth, intelligent systems can be used to develop innovative systems that combine AM with post-processing technology to monitor post-processing performance and meet industry 4.0 requirements.</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

Developing an intelligent approach based on ANFIS and advanced NSGA-III for improving the turning performance of additively manufactured γ-TiAl alloy

  • 期刊论文

  • 发表日期：

    2025年3月13日

查看全文
  • Mustafa M. Nasr,
  • Saqib Anwar

摘要

Additive manufacturing (AM) has become increasingly popular due to its potential to improve production efficiency and meet sustainable manufacturing requirements. However, surface issues and the efficiency of post-processing AM components are still multiple challenges to fulfilling industrial application requirements. This work aims to develop machine learning models based on the Adaptive neuro-fuzzy inference system (ANFIS) and advanced Non-Dominated Sorting Genetic Algorithm III method (NSGA-III) to improve the surface quality of AM products and the efficient post-processing. To achieve this, statistical analysis and a new understanding based on the interaction of the Electron beam melting (EBM) raster path and the cutting tool are proposed to detail the influence of turning parameters, tool types, and their interaction on turning performance: Surface finish, tool wear, and cutting forces during turning of cylindrical γ-TiAl specimens. After that, an artificial intelligence system based on ANFIS and advanced NSGA-III method was developed for monitoring, predicting, and optimizing the post-processing parameters. The results reveal that the cutting conditions and tool types significantly affect the turning performance. This study demonstrated that the ANFIS models could provide a more precise and dependable estimation model for monitoring and predicting flank and rake wear, surface quality, and cutting forces with fewer errors, 0.863%, 0.53%, 1.06%, 0.552%, 1.121%, and 0.6428%, respectively, compared with measured experiments. Finally, the developed intelligent method achieved the optimal turning conditions to enhance surface quality, reduce cutting forces, and reduce tool wear of printed γ-TiAl. Henceforth, intelligent systems can be used to develop innovative systems that combine AM with post-processing technology to monitor post-processing performance and meet industry 4.0 requirements.