Seismic Monitoring of Reinforced Concrete Building Using Artificial Neural Network

Abstract: Effective damage diagnosis is required for existing buildings in seismically active region for quick rehabilitation just after earthquake. In this context research on seismic monitoring of Reinforced Concrete (RC) building is gaining importance in recent time. Internal defects to buildings are sometimes unseen, which may cause major failure due to aftershocks. To avoid this type of failure a robust method for diagnose is required. This paper provides an artificial neural network (ANN) based technique for detecting seismic damage in RC building. A feed-forward multilayer neural network with one hidden layer that translates input samples to output parameters makes up the network. The input to the ANN model is considered as key parameters of real-time earthquake ground motion, while the output is considered as key design parameters for seismic safety of structures susceptible to real-time earthquake ground motion. The suggested ANN model is trained using several forms of real-time earthquake ground motion data to provide a suitable range of design parameters based on the seismic safety of particular RC Structure. The model is also tested against new earthquake ground motion data to ensure its applicability and accessibility for seismic health monitoring purposes. A typical RC structure is considered for the development of the model, and numerical simulation is performed using the finite element method to generate the data required for the development of the ANN model. The ANN model is created using MATLAB software, and the results received from this study are encouraging since the ANN modelbased seismic structural health monitoring may be rapid and efficient. The suggested technique may be utilized to efficiently monitor post-earthquake health of RC buildings, which is critical for any such structures.