Study On Structural Damage Identification Based On The Control Chart Method

Structural damage identification is the core function and the most difficult in structural health monitoring. Many practical difficulties such as modeling errors, environment effects and incomplete measured data reduce the efficiency of damage identification methods. The development of effective and applicable methods for structural damage identification has become urgent in structural health monitoring of civil engineering.This paper focuses on application of statistical process control techniques(control chart)to vibration-based damage identification, and shake table test on a five floor steel frame model was used to investigate this method. Time series analysis is chosen as a tool to extract characteristic parameters for structural damage identification. A unique aspect of this study is the coupling of various projection techniques such as principal component analysis/factor analysis with the statistical process control. We also compare the effectiveness of X chart and EWMA chart in detecting damage.1. First, an auto-regressive (AR) model is fit to the measured acceleration-time histories from an undamaged structure. Residuals and coefficients of the AR model are selected as the damage-sensitive features for the subsequent control chart analysis. Then, the AR residuals and coefficients of the models fitting to subsequent new data are monitored relative to the control limits. A statistically significant number of error terms outside the control limits indicate a system transit from a healthy state to a damage state. In this chapter we also compare the two damage-sensitive features to see which is more effective.2. When data from multiple measurement points are available, Principal Component Analysis (PCA) / Factor Analysis (FA) are used to transforms the time series from multiple measurement points into a single time series prior to the feature extraction process. Then the control chart is used to monitor the single time series, and comparison of the effectiveness of two dimension reduction techniques is made.3. Unlike the X control chart, exponentially weighted moving average (EWMA)control chart uses the entire sequence of points and considers both...