Scholarlink

<p>Multi-material composites offer significant advantages over traditional single-material parts, particularly in aerospace, automobile, biomedical, and architectural applications. Their layered structure enables the combination of distinct material properties, such as pairing lightweight and high-strength materials, to achieve superior strength-to-weight ratios, enhanced flexibility, and improved damage tolerance. In this study, acrylonitrile butadiene styrene (ABS) and polyethylene terephthalate glycol (PETG) were alternately layered using a dual-nozzle fused deposition modeling (FDM) printer to fabricate specimens. ABS contributes strength and impact resistance, making it ideal for structural components and enclosures, while PETG offers flexibility and chemical stability, suitable for protective housings and functional prototypes. To optimize mechanical performance, a hybrid artificial neural network (ANN) and gorilla troops optimizer (GTO) approach was implemented. The ANN (5-15-1 architecture) modeled the nonlinear relationship between five process parameters: layer height (LH), deposition rate (DR), infill density (ID), raster angle (RA), and extruder temperature (ET), and the output tensile strength (TS). The trained model served as the fitness function for GTO, which identified optimal parameters resulting in a predicted TS of 53.80&#xa0;MPa. Validation experiments on three specimens yielded TS of 52.95, 53.20, and 52.70&#xa0;MPa, with an average prediction error of only 1.58%, confirming the framework’s practical effectiveness.</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 ANN-GTO-Based Optimization of Tensile Strength in Layer-by-Layer ABS–PETG Multi-material FDM Composites

  • 期刊论文

  • 发表日期：

    2025年11月24日

查看全文
  • Gayatri Dhankhar,
  • Manoj Kumar Satyarthi

摘要

Multi-material composites offer significant advantages over traditional single-material parts, particularly in aerospace, automobile, biomedical, and architectural applications. Their layered structure enables the combination of distinct material properties, such as pairing lightweight and high-strength materials, to achieve superior strength-to-weight ratios, enhanced flexibility, and improved damage tolerance. In this study, acrylonitrile butadiene styrene (ABS) and polyethylene terephthalate glycol (PETG) were alternately layered using a dual-nozzle fused deposition modeling (FDM) printer to fabricate specimens. ABS contributes strength and impact resistance, making it ideal for structural components and enclosures, while PETG offers flexibility and chemical stability, suitable for protective housings and functional prototypes. To optimize mechanical performance, a hybrid artificial neural network (ANN) and gorilla troops optimizer (GTO) approach was implemented. The ANN (5-15-1 architecture) modeled the nonlinear relationship between five process parameters: layer height (LH), deposition rate (DR), infill density (ID), raster angle (RA), and extruder temperature (ET), and the output tensile strength (TS). The trained model served as the fitness function for GTO, which identified optimal parameters resulting in a predicted TS of 53.80 MPa. Validation experiments on three specimens yielded TS of 52.95, 53.20, and 52.70 MPa, with an average prediction error of only 1.58%, confirming the framework’s practical effectiveness.