Multivariate statistical process control-based hypothesis testing for damage detection in structural health monitoring systems

Abstract

The objective of this paper is to propose a new damage detection technique based on multiscale partial least squares (MSPLS) and optimized exponentially weighted moving average (OEWMA) generalized likelihood ratio test (GLRT) to enhance monitoring of structural systems. The developed technique attempts to combine the advantages of the exponentially weighted moving average (EWMA) and GLRT charts with those of multiscale input-output model partial least square (PLS) and multi-objective optimization. The damage detection problem is addressed so that the data are first modeled using the MSPLS method and then the damages are detected using the OEWMA-GLRT chart. The idea behind the developed OEWMA-GLRT is to compute an optimal statistic that integrates current and previous data information in a decreasing exponential fashion giving more weight to the more recent data and selects the EWMA parameters that minimizes the (MDR), the false alarm rate (FAR) and the average run length (ARL 1 ). This helps provide a more accurate estimation of the GLRT statistic and provide a stronger memory that enables better decision making with respect to damage detection. The performance of the developed technique is assessed and compared with PLS-based GLRT, PLS-based OEWMA, and PLS-based OEWMA-GLRT techniques using two illustrative examples, synthetic data and simulated International Association for Structural Control-American society of Civil engineers (IASC-ASCE) benchmark structure. The results demonstrate the effectiveness of the MSPLS-based OEWMA-GLRT technique over the PLS-based GLRT, PLS-based OEWMA, and PLS-based OEWMA-GLRT methods in terms of MDR, FAR, and ARL 1 values. KEYWORDS damage detection, exponentially weighted moving average, generalized likelihood ratio test, multiscale, partial least squares, structural health monitoring Struct Control Health Monit. 2019;26:e2287.wileyonlinelibrary.com/journal/stc

Effective fault detection in structural health monitoring systems

“…The validation of the proposed approach is also assessed using a nonlinear simulated benchmark structure. [36][37][38][39] The benchmark structures provided by More details on the IASC-ASCE benchmark are presented in Chaabane et al 42 This model was developed to generate simulated input-output response data. 42 In order to construct the KPLS model, the input and output data should be scaled first.…”

“…[36][37][38][39] The benchmark structures provided by More details on the IASC-ASCE benchmark are presented in Chaabane et al 42 This model was developed to generate simulated input-output response data. 42 In order to construct the KPLS model, the input and output data should be scaled first. Then, the data set is divided into training data which consists of 1000 samples and testing data which consists of 1000 samples and that will be used to validate the performance of developed FD technique.…”

“…Results using KPLS-based GLRT method.the IASC-ASCE Structural Health Monitoring Group are studied to verify the effectiveness of damage detection.The benchmark gives an analytical model based on the structure 12-degree-of-freedom (DOF) shearbuilding model. It consists of a 2:5 m 3 2:5 m plan with 3:6 m tall 41,42. The sections are designed for a scale model.…”

Robust Structural Damage Detection Using Analysis of the CMSE Residual’s Sensitivity to Damage

“…Structural damage identification is a fundamental element of structural health monitoring (SHM) that has become a vital tool in maintaining the safety and integrity of structures [1][2][3][4][5][6][7]. Research on vibration-based damage identification has been rapidly expanding over recent decades.…”

“…The adjective "cross" here indicates that MSE-like terms are product terms extending over the baseline finite element model (FEM) of the healthy structure and the measured damaged structure, also extending over various modes. The CMSE is defined as C * m,n = (φ i ) T K n φ * j i = 1, 2, · · · , N i , j = 1, 2, · · · , N j (5) where φ i and K n are the i-th analytical mode shape and the n-th stiffness submatrix of the baseline FEM, respectively, and N i is the number of analytical modes. In practice, it is easy to obtain the analytical modes of the baseline FEM, but difficult or expensive to extract the measured modes of the damaged structure [10]; therefore, one may select a much larger N i than N j , i.e., N i N j .…”