斜拉桥多尺度模型修正及模型确认方法研究
发布时间：2019-06-16 17:49
【摘要】：桥梁结构的健康监测、损伤预后以及安全预后是当前桥梁工程界的研究热点,建立一个准确的、能同时兼顾结构整体特性和局部细节特性的有限元模型是实现桥梁健康监测目标的有效手段之一。同时,为使建立的有限元模型能较好地反应桥梁结构的实际状况,应采用有效的模型修正技术来提高模拟精度,并考虑参数的不确定性对修正后的模型进行确认。本文以江苏沿海高速(G15)灌河大桥(结合梁斜拉桥)为工程背景,建立灌河大桥多尺度模型,并对该多尺度模型进行模型修正和确认,主要工作如下：1.对比分析材料多尺度模拟和结构多尺度模拟的基本理论和方法,为实际结构多尺度模拟提供参考。2.采用结构多尺度模拟方法实现不同类型单元跨尺度的连接,从而建立灌河大桥初始多尺度有限元模型。3.通过分析灌河大桥多尺度模型的特点及误差来源,基于多项式响应面方法和支持向量回归机方法建立两阶段模型修正的基本框架,开展灌河大桥多尺度模型修正。4.结合灌河大桥健康监测系统实测数据,从计算结果和试验结果相关性分析、不确定性正反向传递分析、模型有效性评估三方面进行了灌河大桥多尺度模型确认。得到的主要结论如下：1.本文借鉴材料多尺度模拟的思想,基于Arlequin结构多尺度模拟方法建立了面向结构健康监测的灌河大桥多尺度有限元模型。2.提出了两阶段结构多尺度模型修正方法,其中第一阶段针对大尺度模型采用三阶多项式响应面方法进行修正;第二阶段针对实测数据分别采用三阶多项式响应面方法与支持向量回归机方法进行修正,两种方法分别建立了正反向代理模型。两阶段修正后的灌河大桥多尺度模型的计算频率与实测频率之间的最大误差由21%降低为3%。3.建立了多尺度模型确认的基本框架,研究了参数不确定性的正反向传递规律。若不考虑均值误差,正向传递时各阶频率的重合度指标均大于75%,反向传递时各结构参数的重合度指标均大于65%。模型有效性评估结果表明修正后的灌河大桥多尺度模型在整个参数空间均具有较高精度,因此确认后的灌河大桥多尺度模型可用于结构损伤预后与安全预后分析。
[Abstract]:The health monitoring, injury prognosis and safety prognosis of bridge structure are the research hotspots in the field of bridge engineering at present. The establishment of an accurate finite element model which can take into account the overall and local details of the structure is one of the effective means to achieve the goal of bridge health monitoring. At the same time, in order to make the finite element model reflect the actual situation of the bridge structure, the effective model correction technique should be used to improve the simulation accuracy, and the uncertainty of the parameters should be taken into account to confirm the modified model. In this paper, based on the engineering background of Jiangsu Coastal Expressway (G15) Guanhe Bridge (combined beam cable-stayed bridge), the multi-scale model of Guanhe Bridge is established, and the multi-scale model is modified and confirmed. The main work is as follows: 1. The basic theories and methods of material multi-scale simulation and structure multi-scale simulation are compared and analyzed, which provides a reference for the actual structure multi-scale simulation. 2. The structural multi-scale simulation method is used to realize the cross-scale connection of different types of elements, and the initial multi-scale finite element model of Guanhe Bridge is established. Based on the analysis of the characteristics and error sources of the multi-scale model of Guanhe Bridge, the basic framework of two-stage model correction is established based on the multinomial response surface method and the support vector regression method, and the multi-scale model correction of Guanhe Bridge is carried out. 4. Combined with the measured data of the health monitoring system of Guanhe Bridge, the multi-scale model of Guanhe Bridge is confirmed from three aspects: the correlation analysis between the calculated results and the test results, the positive and negative transfer analysis of uncertainty, and the evaluation of the validity of the model. The main conclusions are as follows: 1. In this paper, based on the idea of material multi-scale simulation and Arlequin structure multi-scale simulation method, a multi-scale finite element model of Guanhe Bridge for structural health monitoring is established. 2. A two-stage structure multi-scale model correction method is proposed, in which the third-order polynomial response surface method is used to modify the large-scale model in the first stage, and the third-order multinomial response surface method and the support vector regression method are used to modify the measured data in the second stage, and the forward and backward proxy models are established by the two methods respectively. The maximum error between the calculated frequency and the measured frequency of the two-stage modified multi-scale model of Guanhe Bridge is reduced from 21% to 3%. The basic framework of multi-scale model confirmation is established, and the forward and backward transfer law of parameter uncertainty is studied. Without considering the mean error, the coincidence index of each order frequency is more than 75% in forward transmission and 65% in reverse transfer. The evaluation results of the validity of the model show that the modified multi-scale model of Guanhe Bridge has high accuracy in the whole parameter space, so the confirmed multi-scale model of Guanhe Bridge can be used to analyze the prognosis and safety prognosis of structural injury.
【学位授予单位】：东南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U448.27
本文编号：2500706
[Abstract]:The health monitoring, injury prognosis and safety prognosis of bridge structure are the research hotspots in the field of bridge engineering at present. The establishment of an accurate finite element model which can take into account the overall and local details of the structure is one of the effective means to achieve the goal of bridge health monitoring. At the same time, in order to make the finite element model reflect the actual situation of the bridge structure, the effective model correction technique should be used to improve the simulation accuracy, and the uncertainty of the parameters should be taken into account to confirm the modified model. In this paper, based on the engineering background of Jiangsu Coastal Expressway (G15) Guanhe Bridge (combined beam cable-stayed bridge), the multi-scale model of Guanhe Bridge is established, and the multi-scale model is modified and confirmed. The main work is as follows: 1. The basic theories and methods of material multi-scale simulation and structure multi-scale simulation are compared and analyzed, which provides a reference for the actual structure multi-scale simulation. 2. The structural multi-scale simulation method is used to realize the cross-scale connection of different types of elements, and the initial multi-scale finite element model of Guanhe Bridge is established. Based on the analysis of the characteristics and error sources of the multi-scale model of Guanhe Bridge, the basic framework of two-stage model correction is established based on the multinomial response surface method and the support vector regression method, and the multi-scale model correction of Guanhe Bridge is carried out. 4. Combined with the measured data of the health monitoring system of Guanhe Bridge, the multi-scale model of Guanhe Bridge is confirmed from three aspects: the correlation analysis between the calculated results and the test results, the positive and negative transfer analysis of uncertainty, and the evaluation of the validity of the model. The main conclusions are as follows: 1. In this paper, based on the idea of material multi-scale simulation and Arlequin structure multi-scale simulation method, a multi-scale finite element model of Guanhe Bridge for structural health monitoring is established. 2. A two-stage structure multi-scale model correction method is proposed, in which the third-order polynomial response surface method is used to modify the large-scale model in the first stage, and the third-order multinomial response surface method and the support vector regression method are used to modify the measured data in the second stage, and the forward and backward proxy models are established by the two methods respectively. The maximum error between the calculated frequency and the measured frequency of the two-stage modified multi-scale model of Guanhe Bridge is reduced from 21% to 3%. The basic framework of multi-scale model confirmation is established, and the forward and backward transfer law of parameter uncertainty is studied. Without considering the mean error, the coincidence index of each order frequency is more than 75% in forward transmission and 65% in reverse transfer. The evaluation results of the validity of the model show that the modified multi-scale model of Guanhe Bridge has high accuracy in the whole parameter space, so the confirmed multi-scale model of Guanhe Bridge can be used to analyze the prognosis and safety prognosis of structural injury.
【学位授予单位】：东南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U448.27
【相似文献】
相关期刊论文 前8条
1 张令弥;计算仿真与模型确认及在结构环境与强度中的应用[J];强度与环境;2002年02期
2 黄丽丽;韩景龙;员海玮;;考虑气动不确定性的气动弹性系统模型确认[J];航空学报;2009年11期
3 吕剑,何颖波,陈成军,邓宏见;泰勒杆试验的动态本构模型确认应用研究[J];环境技术;2005年02期
4 陈志国;邓忠民;毕司峰;;基于Monte Carlo法的结构动力学模型确认[J];振动与冲击;2013年16期
5 朱跃;张令弥;郭勤涛;;基于SVR方法的有限元模型确认中不确定性建模研究[J];振动与冲击;2010年09期
6 臧朝平;刘银超;;轴对称结构的模态振型描述和模型确认[J];南京航空航天大学学报;2012年05期
7 张保强;陈国平;郭勤涛;;模型确认热传导挑战问题求解的贝叶斯方法[J];航空学报;2011年07期
8 ;[J];;年期
相关硕士学位论文 前5条
1 刘琦齐;斜拉桥多尺度模型修正及模型确认方法研究[D];东南大学;2015年
2 黄丽丽;具有不确定性的飞机气动弹性系统模型确认研究[D];南京航空航天大学;2009年
3 刘银超;矩函数在结构动力学模型确认中的应用[D];南京航空航天大学;2012年
4 贾晴晴;基于矩函数的结构动力学模型确认方法研究[D];南京航空航天大学;2012年
5 李同玉;电子设备环境振动仿真分析及其模型确认研究[D];南京航空航天大学;2014年
，本文编号：2500706
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2500706.html
