基于最大熵法的钢框架组合结构地震易损性分析

发布时间：2018-04-05 18:14

本文选题：钢框架结构　切入点：有限元计算　出处：《江西理工大学》2015年硕士论文

【摘要】：地震、海啸、泥石流等自然灾害给人们的生活带来了不小的影响,而强烈的地震是较为严重的、破坏力极强的自然灾害之一,以往地震所造成的人员伤亡和财产损失不计其数,且大多数的损失是由于建筑物的破坏所导致的。建筑设计者一直都在努力寻找提高结构的抗震能力的方法,同样有众多研究人员也在不断的探寻在地震发生前如何更为有效的分析结构在地震作用下的响应,为设计人员提供一定设计资料,因此,结构地震易损性研究越来越受到人们的重视。在地震发生前通过结构地震易损性研究,可以获得震区内结构在地震作用下可能的损伤状态,为结构设计人员有针对的进行抗震设计提供一定的帮助;在地震发生后,对灾区的震害评估同样是一项重要的工作任务,这对灾后的组织救援有着重大的作用。如何在地震未发生前事先对灾区可能发生的破坏形式及数量做出灾害评估,这是地震灾害的又一种预防方式。结构易损性分析不仅可以在地震发生前通过结构设计思路的优化减少地震发生区域由于地震带来的损失,又能在地震发生后给救援组织者提供一项明确的救援思路,合理分配救灾资源。所以,结构地震易损性分析对结构的抗震设计、防灾减灾都有着积极的意义。钢材以其优越的抗拉抗压性能成为建筑结构中必不可少的材料,而钢框架结构则将钢材的性能充分发挥出来,以其优越的抗震性能以及较小的自重成为了建筑业的宠儿,是空间跨度大、抗震性能要求高的建筑物首选结构。然而,在某些地震中钢框架结构同样表现出易损的现象,这使得人们对钢框架结构的抗震性能需要有更加进一步的认识,通过对钢框架结构在地震作用下的结构响应进行易损性分析,更加全面细致的认识钢框架结构在地震作用下的表现。通过对待分析结构进行基于增量动力法的有限元模拟计算,通过设定一系列的地震动强度,模拟结构在不同强度水平地震作用下结构所产生的响应,求解结构响应所服从的概率分布,根据事先定义的结构性能水准,确定分析结构在地震作用下达到或超越某一状态的概率,然后对不同强度等级的地震作用下达到同一状态下的概率点进行拟合,从而得出结构的易损性曲线。对钢框架结构进行地震易损性分析可以预测出该结构在不同强度水平的地震动作用下,结构达到或超越某一性能水准的条件超越概率。由于在进行结构易损性分析时,在确定概率密度分布函数方面,研究者们大多都是对结构响应的分布模式进行假定,按照假定的已知分布模式来求解概率密度函数,这种做法无疑增加了求解结果中人为主观因素对最终结果的影响,使得求解结果的真实性和准确性不能得到很好的保障。本文针对结构地震易损性分析中数据统计的不确定性问题,引入信息论中的最大熵法应用到概率密度函数的求解当中,通过对具有代表性的小样本数据进行分析,并采用遗传算法优化求解概率密度函数中的参数,确定统计数据的概率密度函数,得到更加准确且符合实际分布的概率密度函数。最大熵法是解决在测量数据处理中掺杂主观因素,且不能够准确反映客观事实等问题的有效方法,是解决不确定问题的有效工具,在概率统计中得到广泛应用。该方法主要是通过提取测量数据中数据矩的相关信息来求解被测量变量的概率密度函数,而不是将被测量数据的分布形式假设成为某个经典模型进行概率求解,大大降低了计算过程中人为主观因素的对结果的影响。最大熵法可以求解出蕴含信息量最多的概率密度分布函数,分析矩约束与小样本数据之间的关系,从而得出更加符合实际情况的概率分布。因此,最大熵法是解决不确定问题的最为有效的方法。但是,在使用最大熵法求解概率密度函数时,由于存在多个未知数变量且函数关系式复杂,是一个多元多峰的极值问题,一般的数学方法很难快速的求得符合要求的数值,所以本文在求解参数时使用了遗传算法来求解多个未知参数,通过算法程序中对所求数据的一步步迭代计算,最终获得最符合目标函数要求的参数变量的近似数值。本文通过对钢框架结构进行地震易损性分析,将结构最大顶点位移和最大层间位移作为结构的整体量化指标,通过有限元软件分析利用结构的动力时程分析方法获得结构响应数据,对结构在不同强度水平的地震作用下的破坏形式进行统计分析,确定结构响应所服从的概率密度分布函数及结构的易损性曲线,为基于位移的抗震设计提供参考意见,同样为灾后组织救援提供一定的资料。
[Abstract]:The earthquake, tsunami, landslides and other natural disasters brought no small impact to people's life, and strong earthquake is more serious, one of the most destructive natural disasters caused by the seismic casualties and property losses and most of the losses are too many to count, because of the destruction of buildings caused by the building designers. Have been trying to find a way to improve the seismic capacity of the structure, there are also many researchers are constantly exploring before the earthquake, how to more effective response analysis of structures under earthquake, in order to provide some design information, therefore, the seismic vulnerability research has attracted more and more attention. By study on the seismic vulnerability before the earthquake, the damage state can be obtained in this area may structure under earthquake, for the structural designers have the anti Provide some help in seismic design; after the earthquake, the damage assessment of the disaster area is also an important task, which has a great effect on the post disaster relief organization. How to make the disaster assessment in the failure form and the number of earthquakes occurred before the disaster may occur, this is another prevention of earthquake disaster vulnerability analysis. Not only can the structure by optimizing the structure of the design before the earthquake, the earthquake region due to reduce the losses caused by the earthquake, and after the earthquake to rescue the organizers to provide a clear idea of the rescue, rational allocation of relief resources. Therefore, the seismic vulnerability analysis of earthquake the design of structure, has positive significance to disaster prevention and mitigation. Steel become indispensable in building structure material with superior tensile and compressive stress, and the steel frame structure is steel material The performance of full play, with its superior seismic performance and smaller weight has become the construction industry darling, is the space span, the seismic performance requirements of high buildings preferred structure. However, in some seismic steel frames show the same vulnerability phenomenon, which makes people on the seismic performance of steel frame structure need to have further understanding of vulnerability through the analysis on the structural response of steel frame structure under earthquake, knowledge of the performance of the steel frame structure is more comprehensive and detailed in earthquake. By treating structural analysis of the finite element simulation method based on incremental dynamic, by setting a series of ground motion intensity response simulation structure produced in different seismic intensity levels under the action of probability distribution to solve structural response, according to the structural performance level defined, determine Analysis of the structure to achieve or exceed probability of one state in the earthquake, and the earthquake on the different intensity levels reached the same point probability under the condition of fitting, thus obtains the structural vulnerability curves. The seismic vulnerability analysis can predict the seismic action of different intensity levels using of steel frame structure, structure to achieve or exceed a certain level of performance. Due to the conditional exceedance probability in structural vulnerability analysis, in determining the probability density distribution function, the researchers are mostly on the distribution mode of structure response was assumed, in accordance with the known distribution mode assumption to solve the probability density function, this approach will undoubtedly the increase in solving the problems of subjective factors on the final results, which makes the calculation results of the authenticity and accuracy can not be guaranteed. According to the structure. Vulnerability analysis of earthquake data in the statistical uncertainty problem in information theory, introduce the maximum entropy method is applied to solve the probability density function, through the analysis of small sample data of the representative, and adopting the optimization parameters for the probability density function of the genetic algorithm to determine the probability density function of statistics, probability the density function is more accurate and consistent with the actual distribution. The maximum entropy method is solved in the measurement data processing method and the subjective factor, can not accurately reflect the objective facts and other issues, is the effective tool to solve the uncertain problem, widely used in probability and statistics. This method is mainly by the probability density function of relevant information data extraction data measured in the moment to solve the measured variables, will instead be distributed form of measurement data are assumed to be a classical model in probability The rate of solution, greatly reduces the calculation process and subjective factors influence on the results. The maximum entropy method can solve the information contained the highest amount of probability density function, analysis of the relationship between the moment constraints and small sample data, so as to get more in line with the actual situation of the probability distribution rate. Therefore, the maximum entropy method is to solve the most an effective method for uncertain problems. However, in the use of the maximum entropy method for solving the probability density function, because there are many unknown variables and the function relation is complex, is a multi peak extremum problem, general mathematical methods are difficult to quickly obtain to meet the requirements of the value, so this paper in solving parameters the genetic algorithm is used to solve the many unknown parameters, the calculation of the data through a step by step iterative algorithm in the procedure, finally obtain numerical approximations of the parameter variables of the objective function requirements. In this paper. Through the analysis of seismic vulnerability of steel frame structure, the maximum displacement and the maximum drift index as a whole structure, through the finite element analysis software using structural dynamic time history analysis method to obtain the structural response data, statistical analysis on the failure mode of structure in earthquake with intensity levels. The response of the structure to determine the probability density function and the structure of the vulnerability curve, provide reference for the displacement based seismic design, also provides a data organization for post disaster relief.

【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU311.3;TU398

【相似文献】