Statistical Damage Sensitive Feature for Structural Damage Detection Using AR Model Coefficients

Abstract

Structural Health Monitoring (SHM) and damage detection techniques have captured much interest and attention of researchers and structural engineers owing to their promising ability to provide spatial and quantitative information regarding structural damage and the performance of a structure during its life-cycle. With the development of smart sensors and communication technologies, Wireless Sensor Networks (WSN) has empowered the advancement in SHM. Recently, time series models have been widely used for structural damage detection due to the sensitivity of the model coefficients and residual errors to the damages in the structure. This paper presents a simple index that is computed using the Auto-Regressive (AR) model coefficients as an effective damage sensitive feature (DSF) for the detection of structural damage. Based on this feature, a damage identification method is developed. The Fisher information criterion of the computed DSF is used to statistically decide on the location of damage. This method has been implemented in a simulation environment and the verification of its accuracy in structural damage detection has been carried out experimentally. Experimental data is obtained using wireless sensors from a series of tests performed on a steel beam. The novel damage feature combined with the Fisher criterion for statistical evaluation has shown potential in effective structural damage detection.