Degradation-Aware and Machine Learning-Driven Uncertainty Quantification in Crystal Plasticity Finite Element: Texture-Driven Plasticity in 316L Stainless Steel

2025-05-24

AI Summary

Quantifying plastic response uncertainty in 316L stainless steel welds is computationally prohibitive due to crystallographic texture variability. Method: We propose a degradation-aware, machine learning–driven uncertainty quantification framework integrating high-fidelity crystal plasticity finite element (CPFE) simulations, high-throughput electron backscatter diffraction (EBSD) characterization, and polynomial chaos expansion (PCE) surrogate modeling. A highly accurate RVE surrogate model is constructed using only 200 CPFE evaluations—reducing computational cost by several orders of magnitude. Contribution/Results: We identify Cube and Goss textures as the dominant microstructural drivers of degradation-sensitive plastic response for the first time. This enables rapid, high-fidelity quantification of stress–strain behavior uncertainty, overcoming the efficiency bottleneck of conventional Monte Carlo methods. The framework establishes a new paradigm for service reliability assessment of microstructure-sensitive materials.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Planning, Routing, and Scheduling: Planning under Uncertainty

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The mechanical properties and long-term structural reliability of crystalline materials are strongly influenced by microstructural features such as grain size, morphology, and crystallographic texture. These characteristics not only determine the initial mechanical behavior but also govern the progression of degradation mechanisms, such as strain localization, fatigue damage, and microcrack initiation under service conditions. Variability in these microstructural attributes, introduced during manufacturing or evolving through in-service degradation, leads to uncertainty in material performance. Therefore, understanding and quantifying microstructure-sensitive plastic deformation is critical for assessing degradation risk in high-value mechanical systems. This study presents a first-of-its-kind machine learning-driven framework that couples high-fidelity crystal plasticity finite element (CPFE) simulations with data-driven surrogate modeling to accelerate degradation-aware uncertainty quantification in welded structural alloys. Specifically, the impact of crystallographic texture variability in 316L stainless steel weldments, characterized via high-throughput electron backscatter diffraction (EBSD), is examined through CPFE simulations on calibrated representative volume elements (RVEs). A polynomial chaos expansion-based surrogate model is then trained to efficiently emulate the CPFE response using only 200 simulations, reducing computational cost by several orders of magnitude compared to conventional Monte Carlo analysis. The surrogate enables rapid quantification of uncertainty in stress-strain behavior and identifies texture components such as Cube and Goss as key drivers of degradation-relevant plastic response.

Problem

• Quantifying microstructure-sensitive plastic deformation in 316L stainless steel
• Reducing computational cost for degradation-aware uncertainty quantification
• Identifying key texture components driving degradation-relevant plastic response

Innovation

• Machine learning-driven CPFE surrogate modeling
• Polynomial chaos expansion for uncertainty quantification
• High-throughput EBSD for texture characterization