Vibration-based damage detection using statistical process control

Currently, vibration-based damage detection is an area of significant resea rch activity. This paper attempts to extend the research in this field thro ugh the application of statistical analysis procedures to the vibration-bas ed damage detection problem. The damage detection process is cast in the co ntext of a statistical pattern recognition paradigm. In particular, this pa per focuses on applying statistical process control methods referred to as 'control charts' to vibration-based damage detection. First, an autoregress ive (AR) model is fit to the measured acceleration-time histories from an u ndamaged structure. Residual errors, which quantify the difference between the prediction from the AR model and the actual measured time history at ea ch time interval, are used as the damage-sensitive features. Next, the X-ba r and S control charts are employed to monitor the mean and variance of the selected features. Control limits for the control charts are constructed b ased on the features obtained from the initial intact structure. The residu al errors computed from the previous AR model and subsequent new data are t hen monitored relative to the control limits. A statistically significant n umber of error terms outside the control limits indicate a system transit f rom a healthy state to a damage state. For demonstration, this statistical process control is applied to vibration test data acquired from a concrete bridge column as the column is progressively damaged. (C) 2001 Academic Pre ss.