Structural Damage Detection of a Concrete Based on the Autoregressive All-pole Model Parameters and Artificial Intelligence Techniques

Over the past few decades, damage identification in structural components has been the crucial concern in quality assessment and load capacity rating of infrastructure, as well as in the planning of a maintenance schedule. In this regard, structural health monitoring based on efficient tools to identify the damages in early stages has been focused by researchers to prevent sudden failure in structural components, ensure the public safety and reducing the asset management costs. Therefore, the development and application of sensing technologies and data analysis using machine learning approaches to enable the automatic detection of cracks have become very important. The purpose of this research is to develop a robust method for automatic condition assessment of real-life concrete structures for the detection of relatively small cracks at early stages. A damage identification approach is proposed using the parametric modeling and machine learning approaches to analyze the sensors data. The data obtained from transducers mounted on concrete beams under static loading in laboratory. These data are used as the input parameters. The method relies only on the measured time responses. After filtering and normalization of the data, Autoregressive all-pole model parameters (Yule-Walker method) are considered as features and used as the inputs of a newly developed Self-Advising Support Vector Machine (SA-SVM) for the classification purpose in civil Engineering area. Finally, the results are compared with traditional methods to investigate the feasibility of our proposed method. It is demonstrated that the presented method can reliably detect the crack in the structure and thereby enable the real-time infrastructure health monitoring.