基于模态的损伤识别过程中的传感器优化布置

Optimal sensor placement for mode-shape based damage identification on bridges

为了提高损伤识别的准确性,开发了一种基于结构振动模态和Gauss-Newton算法的结构损伤识别方法。根据测量误差在损伤识别过程中的扩散,建立了损伤识别偏差的上限估计公式。用优化传感器布置的方法来减小损伤识别偏差上限,设计了相应的组合优化算法。通过数值算例验证了该方法的有效性。该方法被应用于识别一座实际桥梁的损伤情况,也得到了比较满意的结果。

A damage identification method to evaluate structural damage in bridges was developed based on mode-shapes and the Gauss-Newton method to improve the accuracy of damage identification. This paper describes the propagation of measurement errors in the damage identification process and obtains the upper bound estimate equation of the identification bias. Then, the analysis is used to reduce the upper bound of the identification bias through optimizing the sensor placement using a combinatorial algorithm. The algorithm is verified by a numerical example, and then implemented for the damage identification of a real bridge.