Autoregressive Modelling and Synthetic Generation of High-Fidelity, Statistically Equivalent 3D Microstructures for As-Manufactured Misalignments in Fiber-Reinforced Composites

2026-06-18

AI Summary

This study addresses the challenge of efficiently generating high-fidelity three-dimensional microstructures of fibrous composites from X-ray micro-CT data. The authors propose an integrated framework that extracts fiber misalignment features via elliptical intersection slice segmentation and constructs a deep statistical model by combining copula-based dependence modeling, a latent-variable autoregressive process, and explicit representation of extreme misalignment patterns. A physics-informed, layer-by-layer synthesis strategy is then employed to generate realistic 3D microstructures. This approach uniquely integrates autoregressive modeling, copula dependence structures, and extreme misalignment characterization, successfully producing simulation-ready configurations comprising approximately 2,400 fibers. The generated microstructures exhibit strong geometric plausibility and statistical equivalence to real data, with key statistical metrics typically deviating by less than 10%.

Abstract

This study presents an integrated framework for processing, modelling, and generating statistically representative three-dimensional fiber microstructures from experimental X-ray-$μ$CT observations. First, an analytical slice-segment ellipse-intersection method is introduced to extract per-slice and per-fiber in-plane and out-of-plane misalignment profiles along the fiber depth. These descriptors are then used to construct a stochastic model that captures slice-wise misalignment distributions and their depth-wise evolution through, copula-based in-plane dependence, latent autoregressive continuity, and rare extreme-misalignment motifs. The model hyperparameters are calibrated using Bayesian optimization, achieving close agreement with the original statistical descriptors, with deviations generally below 10\%. The optimized statistical model is coupled with a physical generation strategy that begins with variable-radius fiber seeding layer and proceeds through an iterative slice-by-slice 3D growth scheme, where the statistical layer guides fiber evolution and Delaunay-based neighbourhood construction with ellipse-based contact resolution ensures non-overlapping, radius-augmented synthetic microstructures. The framework successfully generates about 2400 synthetic fibers while preserving strong statistical fidelity to the original X-ray-$μ$CT data. The proposed pipeline provides a promising and scalable route for generating statistically equivalent, geometrically admissible, and simulation-ready fiber composite microstructures for virtual testing and analysis.

Problem

• 3D microstructures
• fiber-reinforced composites
• misalignments
• statistical equivalence
• synthetic generation

Innovation

• autoregressive modeling
• statistically equivalent microstructures
• fiber misalignment
• synthetic generation
• X-ray μCT