基于倒数变量与广义柔度矩阵的桁架损伤识别 后印本

Краткое изложение: In order to improve the accuracy of damage identification, this paper introduces the concept of reciprocal variable and proposes a truss damage identification method combining reciprocal variable and generalized flexibility matrix. This method only needs low order modal parameters. In this paper, a control equation based on incomplete modes is introduced, and an iterative strategy based on non-negative least square method is used to solve the damage identification problem. In each cycle, the non-negative least square method is used to determine the damage location and degree of the structure. By excluding the non-damaged elements, a new coefficient matrix is reconstructed, and the new coefficient matrix and the original right-hand term are used to solve the next cycle until the convergence accuracy requirements are met. Numerical simulation results show that compared with the original recognition method based on generalized flexibility matrix, the recognition effect of the method combining reciprocal variable with generalized flexibility matrix is better. In the case of noise and incomplete measured mode shapes, the method combining the incomplete mode shape control equation with the iterative strategy can maintain certain identification accuracy.