大型结构系统地震反应分析平台研究

发布时间：2018-04-05 15:21

本文选题：地震反应分析　切入点：子结构方法　出处：《清华大学》2013年硕士论文

【摘要】：目前对结构地震反应的研究方法中，振动台试验受到振动台大小的限制，对大型复杂结构试验能力有限；混合试验方法，尤其是近些年通过子结构技术实现了分布式的试验平台，由于可以有效的利用位于不同地点的试验室设备资源，在国内外得到了较好的发展，但是由于这些方法通常采用了静力凝聚的方法并且子结构边界处自由度通常较少，难以有效的应用于大型复杂结构的数值模拟中。进行精确的地震反应仿真分析是研究工程抗震的重要手段。大型复杂结构的地震反应分析计算量巨大，使得在个人计算机有限的内存和计算能力下，利用常规的分析方法很难进行。并且复杂结构系统可能包括多个具备不同力学特征的子结构，对不同的子结构常常需要采用不同的有限元软件进行模拟。如果利用子结构技术将复杂结构系统划分为多个子结构，允许针对不同子结构使用不同的有限元软件进行分析计算将非常有吸引力。因此整合现有的有限元分析软件，利用各自的优势进行结构地震反应模拟是一种有效的解决问题的方法。本文提出的结构地震反应分析平台将整体结构划分为多个子结构进行计算，计算中考虑子结构间的相互作用。为了满足不同类型的子结构对不同分析软件的需求，在此系统中每个子结构都被视作高度独立的系统，并为每个结构分析软件提供标准化的接口，用来传输各子结构边界处的力和位移。各子结构边界处采用“协调器”满足力的平衡和位移协调，该方法的核心基于柯西-牛顿迭代过程，其实现参考了潘鹏等人针对分布式拟动力实验系统提出的“协调器”算法。研究中建立了应用于商业有限元软件ABAQUS和开源分析软件OpenSees的子结构系统，“协调器”与两个子结构计算系统构成本文中的分析平台。研究中建立了一个拥有四个塔楼的复杂结构模型对该平台做了相应的验证，，通过整体模型数值试验与地震分析平台的计算结果的对比分别讨论了分析平台对于线性问题和非线性问题的有效性。最后利用该平台进行了土-结相互作用（SSI）分析，土壤使用OpenSees进行建模，上部结构使用Abaqus进行建模，并通过本文的结构分析平台考虑土壤与结构相互间的作用，结果表明该分析平台对于具有非线性特征的多个子结构系统的分析具有较强的能力。
[Abstract]:At present, the shaking table test is limited by the size of the shaking table, and its ability to test large and complex structures is limited.Especially in recent years, the distributed test platform has been realized by substructure technology. Because of the effective use of laboratory equipment resources located in different locations, it has been well developed at home and abroad.However, these methods are usually based on static coacervation and the degree of freedom at the boundary of substructures is usually less, so it is difficult to be effectively applied to the numerical simulation of large and complex structures.Accurate seismic response simulation analysis is an important method to study earthquake resistance of engineering.Seismic response analysis of large and complex structures has a huge amount of computation, which makes it difficult to use conventional analysis methods under the limited memory and computing power of personal computer.And the complex structure system may include many substructures with different mechanical characteristics. Different substructures often need to be simulated with different finite element software.If the substructure technique is used to divide the complex structure system into multiple substructures, it will be very attractive to use different finite element software for different substructures.Therefore, it is an effective method to solve the problem by integrating the existing finite element analysis software and making use of their respective advantages to simulate the seismic response of structures.The structural seismic response analysis platform proposed in this paper divides the whole structure into several substructures for calculation, in which the interaction between substructures is taken into account.Used to transmit forces and displacements at the boundaries of each substructure.The balance and displacement coordination of "coordinator" is adopted at the boundary of each substructure. The core of the method is based on Cauchy Newton iteration process. The implementation of the method refers to the "coordinator" algorithm proposed by Pan Peng and others for distributed pseudo-dynamic experimental system.In this paper, the substructure system of commercial finite element software ABAQUS and open source analysis software OpenSees is established. The "coordinator" and two substructure computing systems constitute the analysis platform in this paper.In the study, a complex structure model with four towers was built to verify the platform.The effectiveness of the analytical platform for linear and nonlinear problems is discussed by comparing the numerical results of the whole model and the seismic analysis platform.Finally, this platform is used to analyze soil-junction interaction, OpenSees is used to model soil, Abaqus is used to model superstructure, and the interaction between soil and structure is considered through the structural analysis platform in this paper.The results show that the platform has a strong capability for the analysis of multiple substructure systems with nonlinear characteristics.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU311.3

【参考文献】