Structural System Identification via Validation and Adaptation

2025-06-25

AI Summary

This study addresses the longstanding challenge of tightly coupling parameter identification and model validation in structural system dynamics modeling. We propose a generative co-design framework wherein a physics-informed neural generator maps measurement noise to physical parameters under constraints imposed by governing equations of motion, while a discriminator network jointly guides parameter learning and real-time model validation. By synergistically optimizing adversarial loss and mean squared error, the framework unifies parameter estimation and model validity assessment. Evaluated on multiple strongly nonlinear structural systems, the method achieves significantly lower parameter estimation errors than conventional approaches and enables rigorous structural model validation on independent test sets. The core contribution lies in the first integration of generative modeling, hard physical constraints, and formal model validation—thereby bridging the persistent gap between data-driven identification and theory-based model verification.

Technology Category

• Machine Learning: Structured Learning
• Search and Optimization: Mixed Discrete/Continuous Search
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Estimating the governing equation parameter values is essential for integrating experimental data with scientific theory to understand, validate, and predict the dynamics of complex systems. In this work, we propose a new method for structural system identification (SI), uncertainty quantification, and validation directly from data. Inspired by generative modeling frameworks, a neural network maps random noise to physically meaningful parameters. These parameters are then used in the known equation of motion to obtain fake accelerations, which are compared to real training data via a mean square error loss. To simultaneously validate the learned parameters, we use independent validation datasets. The generated accelerations from these datasets are evaluated by a discriminator network, which determines whether the output is real or fake, and guides the parameter-generator network. Analytical and real experiments show the parameter estimation accuracy and model validation for different nonlinear structural systems.

Problem

• Estimating governing equation parameters from data
• Validating learned parameters using independent datasets
• Identifying and predicting dynamics of complex systems

Innovation

• Neural network maps noise to physical parameters
• Fake accelerations compared via mean square error
• Discriminator network validates generated accelerations