Abstract:International Journal of Structural Stability and Dynamics, Ahead of Print. An accurate finite element model (FEM) plays a critical role in the structural damage identification. However, due to the existence of the uncertainties, such as material properties and modeling errors, it always exists some gaps between the analytical FEM and experimental structure. While an artificial neural network (ANN)-based model updating methods have been widely adopted to narrow the gap and obtain a baseline FEM, it still faces inaccurate results and fails to meet the physical law. In this regard, the study proposes a novel physics-based loss function inspired by modal sensitivity analysis and incorporates it into the residual neural network, thereby forming a novel physics-guided neural network (PGNN) method. The mapping relationship between the input of structural responses and output of model updating variables is constrained to retain its physical meaning by guiding the training process instead of pure data association, which aims to improve the accuracy of the ANN-based method and achieve accurate and high-efficiency model updating. An experimental example of a continuous rigid frame bridge is adopted to verify the feasibility of the proposed method. Additionally, other common model updating methods, including moth-flame optimization and regularization method, are used to make a comparison. The noise-robustness of the proposed method is investigated as well. Compared to the existing method, the results illustrate that the proposed PGNN method can achieve better model updating and good noise-robustness under high uncertainties, which means the introduction of the physics-based loss function significantly enhances the parameters updating ability of the neural network. The proposed method exhibits high efficiency and promising potential for large-scale bridge structure model updating.

Model Updating of a Steel Truss Based on Artificial Neural Networks

Bio-based Connections and Hybrid Planar Truss: A Parallel Genetic Algorithm Approach for Model Updating

Model Updating of Suspended-Dome Using Artificial Neural Networks

Condition Assessment of Joints in Steel Truss Bridges Using a Probabilistic Neural Network and Finite Element Model Updating

Study on Updating Finite Element Model of Steel Truss Structure Based on Knowledge-Enhanced Deep Reinforcement Learning

A DIRECT MODEL UPDATING METHOD USING ADDED KNOWN STIFFNESS

A Model Updating Method for Truss Structure Using Stepwise Uniform Design Schemes Considered Primary Factors

High-Efficiency Finite Element Model Updating of Bridge Structure Using a Novel Physics-Guided Neural Network

Study on Finite Element Model Updating of Nonlinear Structures Using Neural Network

Model Parametric Updating Based on Forward Analysis

Research for Pedestrian Steel Bridge Design of Neural Network in Structural Model Updating

Structural Model Updating of a Steel Truss Structure Utilizing Ambient Vibration Measurement

Application of Artificial Neural Networks to the Response Prediction of Geometrically Nonlinear Truss Structures

A Polynomial Algorithm for Model Updating of Engineering Truss#

Vibration-based Bayesian model updating of an actual steel truss bridge subjected to incremental damage

Finite Element Model Fractional Steps Updating Strategy for Spatial Lattice Structures Based on Generalized Regression Neural Network

Neural Network Model and Zerrilli-Armstrong Model for Prestrained Copper

Beam Structure Finite Element Model Updating via BP Neural Network Optimized

Model Updating and Verification of Multi-scale Model of the Welded Steel Truss

Bayesian Model Updating of a Simply-Supported Truss Bridge Based on Dynamic Responses

Artificial-Neural-Network-Based Mechanical Simulation Prediction Method for Wheel-Spoke Cable Truss Construction