框架结构损伤的检测诊断与评估方法研究

Research on damage identification diagnosis and assessment for frame structures

通过构造以损伤参数为未知量的超定非线性方程组,提出了3种改进的框架结构损伤的检测、诊断、评估方法——改进的一阶迭代法、改进的混合迭代法、改进的动静结合法。通过对框架结构损伤识别的数值模拟分析可以得出:在单柱或梁发生20%及以下损伤的情况下,动静结合法、一阶迭代法、混合迭代法均能较好地识别出损伤的位置与程度。在求解40%及以上的损伤或者多处梁柱损伤时,一阶迭代法和混合迭代法分别只经过25次迭代和22次迭代就达到设定的10-8迭代误差而收敛停止计算,均能准确地识别出损伤单元的位置与损伤程度;与一阶迭代法相比,混合迭代算法由于采用了精确度较高的二阶非线性解析解作为迭代修正的初值,因而迭代修正精度更高,收敛性更好;动静结合法由于结合了结构的动态信息和静态信息,识别精度与混合迭代算法接近且减少了运算时间,并且通过提高计算中所使用的模态阶数可大大提高其识别精度。

The first order iterative algorithm,mixed iteration algorithm and structural damage identification algorithm using static and dynamic data are proposed.The first and second order sensitivity matrixes of modal parameters with respect to the damage member are derived,and the modal truncation error which occurs during the derivation of modal mode sensitivity is improved.The first and second order sensitivity equations are established respectively based on the principle of Taylor series expansion and the solution method of these sensitivity equations is studied.It is observed that the first order iterative algorithm need to iterate only 25 times and mixed iterative algorithm need to iterate only 22 times to converge to the given limited iteration error 10-8.both algorithms could identify the damage location and damage degree accurately.It is shown that the mixed iteration algorithm has a better convergence and a faster iteration speed because the higher precision second order nonlinear analytical solution is adopted.Due to the fact that the damage identification algorithm using static and dynamic data combines the static and dynamic information of the structure,it has the nearly same high precision to the mixed iterative algorithm but needs less operation consumption and the precision can be improved greatly by increasing the modal order when multiple damages and large damage exists.