千米级斜拉桥施工控制系统关键问题研究

发布时间：2019-05-09 13:37

【摘要】：在收集国内外相关研究资料并分析其研究成果的基础上,针对传统的施工控制方法无法满足特大跨度斜拉桥控制精度需求的问题,对千米级斜拉桥施工控制体系、参数识别与结构行为预测方法存在的问题进行了系统研究,主要完成了以下几个方面的工作：1.在收集并分析国内外相关研究资料的基础上,对基于几何控制法的桥梁全过程控制方法进行研究,并将该方法应用于千米级斜拉桥的施工控制中。2.对千米级斜拉桥施工全过程引起施工误差的参数进了较全面的敏感性分析,根据分析结果,确定出主要参数及其影响范围,对以后同级别的斜拉桥的理论计算和施工控制具有一定的指导意义。3.针对千米级斜拉桥的施工控制,提出了采用灰色神经网络结合有限元的最优化反演算法,对施工控制过程中涉及到的部分密切相关的结构参数进行了参数估计与参数识别工作。通过对多参数的综合识别,根据参数识别结果对结构计算模型进行修正,来指导后续的施工过程控制。该方法在样本库相对较小的情况下能够得到比传统识别方法精度更高的结果,可避免识别陷入局部最优解的错误。4.采用D-S证据理论结合灰色神经网络的混合方法,对与千米级斜拉桥施工控制相关的各种误差进行预测工作,对预测到的超过施工控制误差允许范围的各项误差提出了误差处理方法,对其中部分误差的产生机理、处理方式及其影响进行了详细的论述,结合实际控制过程中的数据算例,验证了本文理论的正确性和可行性。5.建立了千米级斜拉桥施工控制数据库系统,在全过程施工控制体系的基础上,提出了几何控制法的控制系统数据处理解决方案,以满足几何控制过程中的参数识别和结构状态预测对数据完备性的要求。在识别和预测工作中能够考虑结构各构件在制造安装等过程中的误差来源及其误差累积过程,使控制过程考虑的影响因素更加全面。6.针对斜拉桥大节段梁段的施工控制,提出了基于最优化方法的误差控制方式,针对梁段在制造、安装过程中产生的各种误差,建立结构最优化问题的数学模型,利用多目标、多约束条件的优化算法,对结构后续不同的施工阶段进行逐步控制和修正,达到降低最终成桥状态误差水平的目的。
[Abstract]:On the basis of collecting the relevant research data at home and abroad and analyzing the research results, aiming at the problem that the traditional construction control method can not meet the requirement of control precision of the long span cable-stayed bridge, the construction control system of the kilometer-scale cable-stayed bridge is studied. The problems existing in parameter identification and structural behavior prediction methods are systematically studied, and the following aspects are completed: 1. On the basis of collecting and analyzing relevant research data at home and abroad, this paper studies the whole process control method of bridge based on geometric control method, and applies this method to the construction control of kilometer cable-stayed bridge. This paper makes a comprehensive sensitivity analysis of the parameters causing construction errors in the construction process of a kilometer cable-stayed bridge. According to the results of the analysis, the main parameters and their influence ranges are determined. It has certain guiding significance for the theoretical calculation and construction control of cable-stayed bridges of the same grade in the future. In view of the construction control of the kilometer cable-stayed bridge, the grey neural network combined with the finite element optimization inversion algorithm is proposed. The parameter estimation and parameter identification of some closely related structural parameters involved in the construction control process are carried out. Through the comprehensive identification of multi-parameters, the structural calculation model is modified according to the results of parameter identification to guide the subsequent construction process control. When the sample base is relatively small, the proposed method can obtain higher accuracy than the traditional recognition method, and avoid the error of falling into the local optimal solution. 4. Based on the mixed method of DES evidence theory and grey neural network, all kinds of errors related to construction control of kilometer cable-stayed bridge are predicted. This paper puts forward the error treatment method for each error which exceeds the allowable range of construction control error, and discusses in detail the mechanism of part of the error, the processing mode and its influence, combined with the data examples in the actual control process, and gives a detailed discussion on the mechanism, processing mode and influence of some of the errors. The correctness and feasibility of the theory in this paper are verified. 5. The construction control database system of kilometer cable-stayed bridge is established. On the basis of the construction control system of the whole process, the data processing solution of geometric control method is put forward. In order to meet the requirements of parameter identification and structural state prediction for data completeness in the process of geometric control. In the process of identification and prediction, the error source and error accumulation process of each structural component in the process of manufacturing and installation can be taken into account, which makes the influence factors considered in the control process more comprehensive. 6. Aiming at the construction control of beam section in large segment of cable-stayed bridge, the error control method based on optimization method is put forward. Aiming at all kinds of errors produced in the manufacture and installation of beam segment, the mathematical model of structural optimization problem is established, and the multi-objective is used. The optimization algorithm with multiple constraints is used to control and correct the subsequent construction stages of the structure step by step, so as to reduce the state error level of the final bridge.
【学位授予单位】：西南交通大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U448.27;U445.4

【参考文献】