基于模态数据的长细结构小损伤精确识别研究

Research on Precise Damage Identification for Long and Slender Structures Using Modal Data

本文针对现有的损伤识别方法不能满足部分结构损伤识别精度要求的现状,对结构的小损伤精确识别方法开展研究.以长细结构为研究对象,对具有不同损伤位置和损伤程度的圆柱形的轻阻尼梁结构进行了数值分析和实验研究,应用数值计算方法和实验确定的特征向量和特征频率对长细结构裂缝参数进行识别计算.本文在研究过程中编制了一个创新性的预测程序,通过其一次性生成目标函数图来选择合适的初始参数,从而对识别结果进行分析.研究结果表明,应用本文提出的识别方法,裂缝位置的识别误差可以控制在0.05%~0.28%范围内,裂缝深度识别误差低于7%.

In view of the fact that the existing damage identification methods can not meet the requirements of some structural damage identification accuracy,this paper studies the small damage identification methods of structures.Taking the slender structure as the research object,the numerical analysis and experimental study on the cylindrical lightly damped beam structures with different damage locations and degrees are carried out.The crack parameters of slender structure are identified and calculated by the numerical method based on the characteristic modes and frequencies determined by the experiment.In this paper,an innovative forecasting program is compiled to select the appropriate initial parameters through one-time generation of the objective function graph,so as to analyze the recognition results.The results show that the identification error of crack location can be controlled within the range of 0.05%~0.28%and the recognition error of crack depth is less than 7%by using the method proposed in this paper.