Structural Health Monitoring in Multi-Story Frames Based on Signal Processing and Rbf Neural Networks

Abstract eng:
In the past few decades, using signal processing tools in structural health monitoring has risen considerably due to recent advances in the field of sensors and other electronic technologies. These advances provide a wide range of response signals such as velocity, acceleration and displacement caused by low to high intensity earthquakes and environmental loads on building structures and bridges. Structural health monitoring i.e. the detection of presence, location and type of damage in structure in order to quantify the amount of damage and predicting the remaining lifetime of structure for service. In this research, wavelet packet transform has been employed in combination with Hilbert transform due to its favorable performance in detection of the structural damages and also its capability for denoising of response signals. In the proposed method, radial basis function (RBF) neural network has been used with the aim of reducing the number of required sensors in order to identify the location and determine the severity of damage caused to the structure. To achieve the proposed goal, the extracted data from each response signal should be increased to provide some information with regard to the higher modes. Finally, the obtained data is used to train the RBF neural network. The performance of the proposed method has been verified by means of numerical examples. To demonstrate the capabilities of the proposed algorithm, numerical simulations are performed on a four-story two-bay shear frame with different damage scenarios using OpenSees. The results show that this method can detect the occurrence, location and severity of damage with good accuracy even in the presence of measurement noise.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket