摘要
以基于灵敏度分析的有限元模型修正方法为基础,提出了一种基于1范数正则化过程的结构损伤识别方法。通过与以Tikhonov正则化为代表的二次型正则化过程相比较,本文的理论分析表明1范数正则化方法在迭代计算过程中能根据上一迭代步损伤识别结果自适应地调整正则化项中的损伤参数权系数,从而显著改善了Tikhonov正则化识别结果过度光滑的缺陷,更利于识别结构的局部损伤。为解决引入1范数造成的数值计算困难,文中还对基于1范数正则化的模型修正算法进行了改进。以二维框架模型为例的损伤识别数值模拟表明:1范数正则化方法与模型修正方法相结合可以有效抑制实测模态参数中噪声的影响,体现出较好的鲁棒性;在模态噪声水平达到10%的情况下,仍能有效抑制噪声干扰,凸显结构局部损伤位置,准确识别损伤程度。
In this paper a damage identification method is developed via the finite element model updating technique based on the sensitivity-analysis with 1-norm-based regularization scheme.Compared with classical quadratic regularization process,1-norm-based regularization method can adjust the regularization parameters matrix in regularization items adaptively according to the damage identification results in each iterative step.Therefore,the proposed method can overcome the smearing effect of Tikhonov regularization,and identify the local damage more accurately.The model updating algorithm with 1-norm-based regularization is also improved to avoid the numerical difficulties.Numerical simulations of two-dimensional frame show that the model updating method with 1-norm-based regularization can effectively suppress the effects of measurement noise and identify the location and extent of the local structure damages correctly even in the case that the noise level is 10%.
出处
《应用力学学报》
CAS
CSCD
北大核心
2013年第5期756-761,807,共6页
Chinese Journal of Applied Mechanics
基金
国家自然科学基金(51268045)
教育部高等学校博士点基金资助项目(20103601110006
20123601120011)
交通运输部科技项目资助(20113187801370)
