Measurement selection for vibration-based structural damage identification

Many researchers have intensively investigated the methods of using vibration properties of a structure to identify its damage in the past two decades. It has been realized that the modal data included in damage identification analysis have a great influence on the accuracy of identification results. In this paper, a technique is presented to select a subset of measurement points and corresponding mode to be used in model updating for damage identification. A new concept of damage measurability is introduced in terms of two factors, namely the sensitivity of a residual vector to the structural damage, and the sensitivity of the damage to the measurement noise. Based on the two factors, the structural damage measurability is estimated quantitatively. It yields more accurate and reliable damage identification results if the mode and measurement points corresponding to the largest damage measurability value are chosen in field test and numerical analysis. The advantage of the technique is that it is based on the undamaged state of structures and thus independent of the damage configuration. Therefore, it is applicable in practice to determine the measurement selection prior to field testing and damage identification analysis. A frame structure is used to demonstrate the validity and versatility of the method. It demonstrates a significant improvement on the accuracy and reliability of damage identification results when the measurement points and mode are determined according to the present technique. (C) 2000 Academic Press.