Scholarlink

<p>Over the last sixty years, exhaust system research has become increasingly important with advances in aerodynamic vehicle design. While many studies have explored nozzle performance through experimental and computational techniques, limited attention has been given to the role of geometric proportions in optimizing exhaust efficiency. This research introduces a novel approach by applying the Golden Ratio (1:1.618) to the design of rectangular convergent exhaust systems, comparing their performance against conventional square nozzles. The study combines experimental testing, CFD analysis, and a Multi-Layer Feedforward Neural Network (MLFFN) framework to model thrust efficiency and quantify over-expansion and under-expansion losses. The novelty of the work lies in integrating the mathematical and structural analysis of MLFFN with Golden Ratio-based nozzle geometry to establish a predictive optimization framework. The scope of this research is to demonstrate that exhaust nozzles designed using the Golden Ratio can enhance thrust efficiency, reduce expansion losses, and provide a systematic methodology for applying MLFFN in aerodynamic optimization. This integrated approach contributes to bridging the gap between geometric design principles and intelligent computational modeling in exhaust system research.</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

Optimization of exhaust systems using golden ratio through multi-layer feedforward network framework

  • 期刊论文

  • 发表日期：

    2025年9月29日

查看全文
  • Mohammed Asif Kattimani,
  • Habib Masum,
  • Sumeet Hangargi,
  • Hasibur Rahman,
  • Syed Mustafkhadri,
  • G. S. Girishkumar

摘要

Over the last sixty years, exhaust system research has become increasingly important with advances in aerodynamic vehicle design. While many studies have explored nozzle performance through experimental and computational techniques, limited attention has been given to the role of geometric proportions in optimizing exhaust efficiency. This research introduces a novel approach by applying the Golden Ratio (1:1.618) to the design of rectangular convergent exhaust systems, comparing their performance against conventional square nozzles. The study combines experimental testing, CFD analysis, and a Multi-Layer Feedforward Neural Network (MLFFN) framework to model thrust efficiency and quantify over-expansion and under-expansion losses. The novelty of the work lies in integrating the mathematical and structural analysis of MLFFN with Golden Ratio-based nozzle geometry to establish a predictive optimization framework. The scope of this research is to demonstrate that exhaust nozzles designed using the Golden Ratio can enhance thrust efficiency, reduce expansion losses, and provide a systematic methodology for applying MLFFN in aerodynamic optimization. This integrated approach contributes to bridging the gap between geometric design principles and intelligent computational modeling in exhaust system research.