Generative adversarial framework to calibrate excursion set models for the 3D morphology of all-solid-state battery cathodes

2025-03-21

AI Summary

Acquiring high-fidelity 3D cathode microstructures for all-solid-state batteries (ASSBs) remains challenging, and purely data-driven generative adversarial networks (GANs) suffer from poor interpretability and limited controllability. Method: This work proposes a hybrid generative framework integrating a low-parameter stochastic geometric model—specifically, the generalized random field run-length set—as a structural prior with a GAN. The model is calibrated solely from 2D microscopy images and generates physically realistic, high-fidelity 3D cathode digital twins. Contribution/Results: This is the first approach unifying run-length set modeling and GAN-based generation, jointly ensuring morphological fidelity and explicit interpretability of microstructural parameters. It overcomes the fundamental limitation of conventional GANs in systematically controlling microstructural configurations. The generated 3D digital twins enable virtual materials testing and macroscopic performance simulation, establishing a new paradigm for structure–property relationship analysis and targeted optimization of porous functional materials.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

This paper presents a computational method for generating virtual 3D morphologies of functional materials using low-parametric stochastic geometry models, i.e., digital twins, calibrated with 2D microscopy images. These digital twins allow systematic parameter variations to simulate various morphologies, that can be deployed for virtual materials testing by means of spatially resolved numerical simulations of macroscopic properties. Generative adversarial networks (GANs) have gained popularity for calibrating models to generate realistic 3D morphologies. However, GANs often comprise of numerous uninterpretable parameters make systematic variation of morphologies for virtual materials testing challenging. In contrast, low-parametric stochastic geometry models (e.g., based on Gaussian random fields) enable targeted variation but may struggle to mimic complex morphologies. Combining GANs with advanced stochastic geometry models (e.g., excursion sets of more general random fields) addresses these limitations, allowing model calibration solely from 2D image data. This approach is demonstrated by generating a digital twin of all-solid-state battery (ASSB) cathodes. Since the digital twins are parametric, they support systematic exploration of structural scenarios and their macroscopic properties. The proposed method facilitates simulation studies for optimizing 3D morphologies, benefiting not only ASSB cathodes but also other materials with similar structures.

Problem

• Generate realistic 3D morphologies from 2D images
• Combine GANs with stochastic models for interpretability
• Optimize 3D structures for battery cathodes virtually

Innovation

• Generative adversarial networks calibrate stochastic models
• Combines GANs with low-parametric geometry models
• Generates 3D digital twins from 2D images