Scholarlink

<p>This review presents a comprehensive analysis of the design, characterization, and performance evaluation of polymer-based gears fabricated using additive manufacturing (AM) technologies. Emphasizing the integration of experimental testing and machine learning (ML), the study highlights how AM techniques, such as Fused Filament Fabrication (FFF), Stereolithography (SLA), and Selective Laser Sintering (SLS), impact the mechanical integrity of printed gears, particularly in terms of anisotropy, dimensional stability, and surface roughness. Experimental findings on material selection, build orientation, and printing parameters are critically examined alongside ML algorithms including Artificial Neural Networks (ANN), Genetic Algorithms (GA), and ensemble models, which are employed for predicting wear, optimizing process settings, and accelerating design cycles. The synergistic use of empirical data and predictive modeling is shown to significantly reduce prototyping costs, enhance design reliability, and enable intelligent maintenance strategies. This review concludes by outlining future research opportunities involving hybrid simulation-ML frameworks, the deployment of embedded sensing systems, and the development of high-performance and composite polymers for next-generation gear 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

Experimental and machine learning approaches for the design and optimization of additively manufactured polymer gears: a review

  • 期刊论文

  • 发表日期：

    2025年12月15日

查看全文
  • Ali Hussein Alwan,
  • Mohsin Noori Hamzah,
  • Alaa Abdulhady Jaber

摘要

This review presents a comprehensive analysis of the design, characterization, and performance evaluation of polymer-based gears fabricated using additive manufacturing (AM) technologies. Emphasizing the integration of experimental testing and machine learning (ML), the study highlights how AM techniques, such as Fused Filament Fabrication (FFF), Stereolithography (SLA), and Selective Laser Sintering (SLS), impact the mechanical integrity of printed gears, particularly in terms of anisotropy, dimensional stability, and surface roughness. Experimental findings on material selection, build orientation, and printing parameters are critically examined alongside ML algorithms including Artificial Neural Networks (ANN), Genetic Algorithms (GA), and ensemble models, which are employed for predicting wear, optimizing process settings, and accelerating design cycles. The synergistic use of empirical data and predictive modeling is shown to significantly reduce prototyping costs, enhance design reliability, and enable intelligent maintenance strategies. This review concludes by outlining future research opportunities involving hybrid simulation-ML frameworks, the deployment of embedded sensing systems, and the development of high-performance and composite polymers for next-generation gear applications.