An iterative order determination method for time-series modeling in structural health monitoring

Abstract

Statistical time-series modeling has recently emerged as a promising and applicable methodology to structural health monitoring. In this methodology, an important step is to choose robust and optimal orders of time-series models for extracting damage-sensitive features. In this study, an iterative order determination method is proposed to determine optimal orders based on residual analysis. The proposed technique consists of identifying the best time-series model, determining the maximum orders, and selecting the optimal orders that enable the model to extract uncorrelated residuals. The application of low-pass signal filters to the process of order determination is also evaluated. In a comparative study, the influence of optimal orders on damage is assessed to perceive whether features extracted from optimal models, with and without using low-pass filters, are sensitive to damage. Experimental data of a three-story laboratory frame and a large-scale bridge are applied to demonstrate the performance and capability of the proposed method. Results show that the proposed iterative method is an efficient tool for the determination of optimal orders along with the extraction of uncorrelated residuals.

Damage Identification of a Steel Frame Based on Integration of Time Series and Neural Network under Varying Temperatures

“…Time series analysis, which uses sequence of data points with uniform time intervals to create systems, can be applied to feature extraction, damage diagnosis, and damage classification without modal parameter identification in SHM [3]. Various time series models, such as AR [7,8,26], ARX (autoregressive model with exogenous input) [27], and ARMA (autoregressive-moving average) [28], have been extensively explored, whose features based on their coefficients have been extracted as damage indices by the SHM community. Das et al [1] conducted a comparative study of the effectiveness of different vibration-based damage identification methods and proved that time series analysis outperformed other methods in damage identification with the presence of operational or environmental nuisances.…”

Fast unsupervised learning methods for structural health monitoring with large vibration data from dense sensor networks

“…To robustly select the order, with an emphasis on extracting uncorrelated residuals, Entezami and Shariatmadar 38 proposed an iterative procedure through the Ljung–Box Q (LBQ) test, assuring model accuracy and adequacy. Rezaiee-Pajand et al 39 also presented a two-stage iterative algorithm, to select the model order on the basis of a residual analysis through statistical hypothesis tests and signal filtering.…”

Structural Health Monitoring for Condition Assessment Using Efficient Supervised Learning Techniques

“…Statistical decision making refers to the application of statistical methods for the classification of the extracted features. This step is related to the implementation of machine learning algorithms, to classify the structural state conditions and identify possible damage states [5,[14][15][16]. The simple idea of machine learning relies on identifying a relationship between some features derived from the measured data in the undamaged or damaged conditions, as a training data set.…”