利用残余力概念进行结构损伤识别

提出一种利用残余力进行结构损伤识别方法。用刚度损伤系数将有限元模型参数化来描述结构的损伤。用特征值和特征向量预测算法构造目标函数,利用两点交叉算子的二进制代码的遗传算法反复迭代优化刚度损伤系数,获得待测结构的损伤系数、损伤位置与损伤程度的信息。通过一个平面桁架对该方法进行验证,结果表明:利用残余力概念进行损伤识别所得结构的损伤系数与理论值很接近。该方法能够较准确地判断结构的损伤位置和损伤程度。

Method for Structure Damage Detection by Using the Concept of Residual Forces

This paper proposes a method of damage detection by using the concept of residual forces.To describe the damage in a structure,finite element models are parameterized by structural stiffness reduction parameters.An eigenvalue and eigenvector prediction algorithm along with normalized residual function is employed to formulate the objective function.Two-point crossover binary coded genetic algorithm is adopted in minimizing the objective and optimum set of stiffness reduction parameters are predicted.To illustrate the method,we use a plane truss as an example.And the result shows,the damage coefficients of the structure obtained from the method for structure damage detection by using the concept of residual force are very close to theoretical values.This method can estimate the damage location and degree of the structure with better precision.