Scholarlink

<p>This work presents TenMARO, a new Tent Mapping Artificial Rabbit Optimization algorithm, designed for effective fault parameter optimization in Gas Turbine Aero-Engines (GT-AE). As opposed to the conventional metaheuristics, TenMARO combines deterministic tent mapping with adaptive foraging strategies to avoid convergence stagnation and local optimum difficulties. TenMARO minimizes Mean Square Error (MSE) between DT-predicted and healthy performance parameters under physical constraints. A validated MATLAB Simulink and ANSYS Fluent DT model combined with NASA’s C-MAPSS dataset serves as the test environment. Comparative experiments with IPSO, CGWO, MARO, and SSO confirm TenMARO’s superiority in convergence speed (39.2% faster), fitness value (97.84), and computational efficiency. The outcomes illustrate TenMARO's viability as a scalable and resilient design optimization platform for real-time propulsion system diagnosis. Statistical evaluation (Wilcoxon test, <i>p</i> &lt; 0.05) and error bar analysis establish robustness. The work provides a transparent, reproducible, and computationally efficient framework that aligns optimization algorithm design with data science reproducibility principles and DT-based aero-engine fault management.</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

TenMARO: a deterministic–chaotic metaheuristic framework for data-driven fault parameter optimization in aero-engine digital twin

  • 期刊论文

  • 发表日期：

    2025年12月6日

查看全文
  • Naga Venkata Rama Subbarao Tadepalli,
  • Ramji Koona

摘要

This work presents TenMARO, a new Tent Mapping Artificial Rabbit Optimization algorithm, designed for effective fault parameter optimization in Gas Turbine Aero-Engines (GT-AE). As opposed to the conventional metaheuristics, TenMARO combines deterministic tent mapping with adaptive foraging strategies to avoid convergence stagnation and local optimum difficulties. TenMARO minimizes Mean Square Error (MSE) between DT-predicted and healthy performance parameters under physical constraints. A validated MATLAB Simulink and ANSYS Fluent DT model combined with NASA’s C-MAPSS dataset serves as the test environment. Comparative experiments with IPSO, CGWO, MARO, and SSO confirm TenMARO’s superiority in convergence speed (39.2% faster), fitness value (97.84), and computational efficiency. The outcomes illustrate TenMARO's viability as a scalable and resilient design optimization platform for real-time propulsion system diagnosis. Statistical evaluation (Wilcoxon test, p < 0.05) and error bar analysis establish robustness. The work provides a transparent, reproducible, and computationally efficient framework that aligns optimization algorithm design with data science reproducibility principles and DT-based aero-engine fault management.