大型结构模态综合及子结构损伤识别方法研究

发布时间：2018-04-02 03:19

本文选题：动态子结构　切入点：自由界面模态综合　出处：《福州大学》2014年硕士论文

【摘要】：健康监测系统在大型土木工程结构中得到了日益广泛的应用和发展,但仍有一些问题未得到解决。比如在搭建大型结构健康监测系统时,需要建立具有一定精度的基准模型以进行损伤识别和结构健康状态评估,而在修正基准模型过程中,由于参数过多,容易导致计算量太大,优化病态以及迭代不收敛等问题。鉴于此,本文通过研究动态子结构方法,并将其应用于土木工程结构中,进而研究动态子结构损伤识别方法。主要完成以下几方面的工作：1、自由界面模态综合法和双协调自由界面模态综合法在土木工程结构上的应用研究,并分别针对2种模态综合方法提出相应的模态截取准则和子结构划分原则,以保证良好的综合精度。研究结果表明子结构的划分不仅要考虑其几何形状和材料特性的影响,若综合考虑结构各振型反弯点的影响,将获得更精确的综合效果。2、数值仿真中选取了一个以实际工程为原型的大跨度连体结构,结果验证了动态子结构方法在大型结构上应用的可行性,以及提出的模态截取准则和划分原则的正确性,同时也进一步分析了子结构划分个数对大型结构计算精度及计算效率的影响。3、在传统的基于灵敏度的模型修正方法基础上引入了动态子结构方法,首先基于自由界面模态综合方法和双协调自由界面模态综合方法推导出了结构的特征值和特征向量关于单元参数的灵敏度矩阵表达式,由此建立了基于动态子结构的模型修正方法。接着与传统方法相比较,发现动态子结构模型修正方法在每个迭代步求取灵敏度矩阵时,只需要对有限个自由度对应的项进行求解,从而将在很大程度上提高计算效率。4、将推导出的基于自由界面模态综合的模型修正方法应用于框架结构的损伤识别中。结果表明,与传统的基于整体结构灵敏度的模型修正方法相比较,采用本文提出的基于自由界面模态综合的模型修正方法时收敛将更为稳定,且不容易产生误判,损伤识别能力也相对较优。5、通过框架结构损伤识别的算例,证实基于双协调自由界面模态综合的模型修正方法的损伤识别能力。由于该方法在每个迭代步仅需要在自由度较小的某个子结构中进行求解,从而可以大大减少迭代时间,因而预期这种方法将更适合于具有庞大自由度的大型复杂结构,具有更好的应用前景。
[Abstract]:Health monitoring system has been widely used and developed in large civil engineering structures, but there are still some problems that have not been solved. It is necessary to establish a reference model with certain accuracy for damage identification and structural health assessment. However, in the process of revising the benchmark model, too many parameters will easily lead to too much calculation. In this paper, the dynamic substructure method is studied and applied to civil engineering structures. Then the damage identification method of dynamic substructure is studied. The main work is as follows: 1. The application research of free interface modal synthesis method and double coordinated free interface modal synthesis method in civil engineering structure. The corresponding modal interception criteria and substructure partition principles are proposed for the two modal synthesis methods to ensure good synthesis accuracy. The results show that the substructure partition should not only take into account the effects of geometric shape and material characteristics. If we consider the influence of the reverse bending point of the structure, we can get a more accurate synthetic effect. In the numerical simulation, a large span conjoined structure with actual engineering prototype is selected. The results verify the feasibility of applying dynamic substructure method to large structures, and the correctness of the proposed modal interception criteria and partition principles. At the same time, the influence of the number of substructure partition on the calculation accuracy and efficiency of large structures is also analyzed. The dynamic substructure method is introduced based on the traditional sensitivity based model modification method. Firstly, based on the free interface modal synthesis method and the dual coordinated free interface modal synthesis method, the eigenvalues of the structure and the sensitivity matrix expressions of the eigenvector for the element parameters are derived. Then, compared with the traditional method, it is found that the dynamic substructure model modification method is used to obtain the sensitivity matrix in each iteration step. Only the terms corresponding to the finite number of degrees of freedom need to be solved, which will greatly improve the computational efficiency. Finally, the derived model modification method based on free interface modal synthesis is applied to the damage identification of frame structures. Compared with the traditional model correction method based on the sensitivity of the whole structure, the proposed model correction method based on free interface modal synthesis will converge more stably and will not be prone to misjudgment. The ability of damage identification is also relatively superior. 5. Through the example of frame structure damage identification, The damage detection ability of the model correction method based on double coordinated free interface modal synthesis is confirmed. Since the method only needs to be solved in a substructure with less degree of freedom per iteration step, the iterative time can be greatly reduced. Therefore, it is expected that this method will be more suitable for large and complex structures with large degrees of freedom, and will have a better application prospect.
【学位授予单位】：福州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU317

【相似文献】