Scholarlink

<p>Assessing TRIP steel joint reliability is challenging under small-sample conditions. This study develops a Generalized Maximum Entropy (GME) regression within an Accelerated Failure Time (AFT) framework to predict Mean Time to Failure (MTTF) without assuming rigid distributional shapes (e.g., Weibull). Using a 3x5 design of GMAW lap joints, the model yielded a robust <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(R^2_{PRESS}\)</EquationSource> <EquationSource Format="MATHML"><math> <msubsup> <mi>R</mi> <mrow> <mi mathvariant="italic">PRESS</mi> </mrow> <mn>2</mn> </msubsup> </math></EquationSource> </InlineEquation> of 0.86. Results quantify the high sensitivity of fatigue life to process parameters: a 1 J increase in heat input yields a 0.789% gain in MTTF, while a 1 MPa increase in stress reduces it by 1.968%. Specifically, at a reference stress of 148.93 MPa and 273 J heat input, the model predicts an MTTF of 216,486 cycles. The proposed information-theoretic GME approach yields an interpretable, closed-form MTTF surface, offering a defensible decision support tool for RCM in advanced manufacturing subject to limited experimental evidence availability.</p> Graphical abstract <p></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

Estimation of the mean time to failure (MTTF) of welded joints in TRIP steels using the maximum entropy principle

  • 期刊论文

  • 发表日期：

    2026年3月12日

查看全文
  • Octavio P. Gaona,
  • David Salvador González-González,
  • Rolando Javier Praga Alejo,
  • Marco A. Fuentes-Huerta,
  • Elan Iñaky Laredo Alcalá,
  • Argelia Fabiola Miranda Pérez

摘要

Assessing TRIP steel joint reliability is challenging under small-sample conditions. This study develops a Generalized Maximum Entropy (GME) regression within an Accelerated Failure Time (AFT) framework to predict Mean Time to Failure (MTTF) without assuming rigid distributional shapes (e.g., Weibull). Using a 3x5 design of GMAW lap joints, the model yielded a robust \(R^2_{PRESS}\) R PRESS 2 of 0.86. Results quantify the high sensitivity of fatigue life to process parameters: a 1 J increase in heat input yields a 0.789% gain in MTTF, while a 1 MPa increase in stress reduces it by 1.968%. Specifically, at a reference stress of 148.93 MPa and 273 J heat input, the model predicts an MTTF of 216,486 cycles. The proposed information-theoretic GME approach yields an interpretable, closed-form MTTF surface, offering a defensible decision support tool for RCM in advanced manufacturing subject to limited experimental evidence availability.

Graphical abstract