Understanding the degree of reinforcing bar corrosion in reinforced concrete (RC) structures is crucial for evaluating the behavior. This study develops a simulation system for estimating the corrosion distribution along the rebar of a RC beam member based on surface…
Understanding the degree of reinforcing bar corrosion in reinforced concrete (RC) structures is crucial for evaluating the behavior. This study develops a simulation system for estimating the corrosion distribution along the rebar of a RC beam member based on surface crack widths. The system integrates the rigid body spring model (RBSM) with machine learning methods. The RC beam is modeled and RBSM simulations with different expansion distributions are run with 500 analysis steps. The expansion data and the corresponding surface crack width data generated from the simulations are used to build the training dataset for machine learning. A large number of training data samples are obtained by extracting the simulation results step-by-step. The inputs are surface crack widths from several locations and the desired output is the internal corrosion-induced expansion. After training with the dataset, the neural network is able to correlate inputs and outputs, allowing it to estimate an expansion distribution from given cracking data. The estimated expansion distribution is then used to simulate the surface cracks using RBSM, and the error between the given cracking data and simulated cracks is returned as an input to the trained network in order to optimize the expansion estimation and enhance performance of the system. The feasibility of the proposed RBSM-neural network system is validated using both synthetic and experimental test data. The estimation results align well with the target data, demonstrating the effectiveness of the system in estimating internal expansion along the rebar and reproducing the cracking distribution using surface crack data. Internal distributions of cracking and stress states are extracted from the simulations, providing additional information for further analysis of structural performance.