PC-SRGAN: Physically Consistent Super-Resolution Generative Adversarial Network for General Transient Simulations

2025-05-10

AI Summary

Scientific super-resolution (SR) images often lack physical consistency, limiting their utility in interpretable simulations. To address this, we propose PC-SRGAN—a novel generative adversarial network that for the first time embeds a numerically stable temporal integrator into the GAN framework, explicitly coupling physics-based constraints during image generation to ensure causality and fidelity of transient dynamics. Our method jointly optimizes a physics-constrained loss, PSNR, and SSIM, and introduces a causal quality assessment metric. Experiments demonstrate that PC-SRGAN surpasses SRGAN using only 13% of the training data, achieving significant improvements in PSNR (+2.1 dB) and SSIM (+0.042). It enables high-fidelity surrogate modeling, particularly excelling in low-data regimes for scientific simulation enhancement.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Causal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• 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

Machine Learning, particularly Generative Adversarial Networks (GANs), has revolutionised Super Resolution (SR). However, generated images often lack physical meaningfulness, which is essential for scientific applications. Our approach, PC-SRGAN, enhances image resolution while ensuring physical consistency for interpretable simulations. PC-SRGAN significantly improves both the Peak Signal-to-Noise Ratio and the Structural Similarity Index Measure compared to conventional methods, even with limited training data (e.g., only 13% of training data required for SRGAN). Beyond SR, PC-SRGAN augments physically meaningful machine learning, incorporating numerically justified time integrators and advanced quality metrics. These advancements promise reliable and causal machine-learning models in scientific domains. A significant advantage of PC-SRGAN over conventional SR techniques is its physical consistency, which makes it a viable surrogate model for time-dependent problems. PC-SRGAN advances scientific machine learning, offering improved accuracy and efficiency for image processing, enhanced process understanding, and broader applications to scientific research. The source codes and data will be made publicly available at https://github.com/hasan-rakibul/PC-SRGAN upon acceptance of this paper.

Problem

• Enhances image resolution with physical consistency
• Improves accuracy metrics with limited training data
• Provides reliable surrogate for time-dependent simulations

Innovation

• Enhances resolution with physical consistency
• Uses numerically justified time integrators
• Improves accuracy with limited training data