【作者】 傅彬；

【导师】 唐小兵；

【作者基本信息】 武汉理工大学 ， 结构工程， 2008， 硕士

【摘要】 现代斜拉桥以其良好的结构性能和跨越能力以及优美的建筑造型在现代桥梁结构中占据重要地位,斜拉桥的仿真计算技术以及施工控制理论也日趋成熟,本文在目前已有的研究成果基础上,以铁锣坪大桥为研究对象,就斜拉桥施工控制仿真计算、主梁分段悬臂浇注过程控制等问题进行了初步探讨。本文首先回顾了国内外斜拉桥发展的历史,分析了国内外斜拉桥施工控制的现状以及发展前景,然后介绍了斜拉桥施工控制系统的主要组成部分,主要内容以及影响斜拉桥施工控制的各项因素,并提出了一些解决办法。其次,深入讨论了斜拉桥施工控制过程中的结构模拟计算分析方法以及结构分析中需考虑的有关问题,根据斜拉桥结构设计原理以及悬臂施工方法的特点,对大跨度斜拉桥的施工过程进行仿真分析。采用前进分析方法模拟每一阶段施工状态,得到各个工况下结构的索力、内力、位移值,并且选择具有代表性的斜拉索和施工梁段进行研究,然后对仿真计算的结果进行了详细的分析,总结出了结构在施工过程中的一些变化规律,为施工控制的实施奠定了理论轨迹。最后,具体阐述了铁锣坪大桥的施工方案和施工工艺、施工监测的主要内容和操作流程,详细说明了实际施工过程中须注意的问题以及施工控制的精度。由于在实际悬臂浇注施工中梁段前端的标高变化和仿真计算的标高变化有一定的差异,进而影响到立模标高的确定,所以本文引入了BP人工神经网络对未施工梁段的前端标高变化进行预测,将影响梁段标高变化的因素如已施工梁段距离所在索塔中心的距离,对应斜拉索第三次张拉力的大小,张拉时的大气温度,理论计算的标高变化作为输入样本,将实测标高变化作为输出样本对BP网络进行训练,利用BP网络具备输入参数与输出参数之间的非线性映射关系的特点对后续梁段在斜拉索第三次张拉完毕后的标高变化进行预测,为施工控制提供参考。通过对预测值与实测值的比较表明,二者吻合较好,证明了将BP神经网络用于斜拉桥施工控制是合理的,可行的。并且总结出了一些影响网络预测精度的原因以供参考。更多还原

【Abstract】 Modern cable-stayed bridge with its good performance and the structure and the ability to leap beautiful architectural modeling in the modern bridge structure play an important role, the cable-stayed bridge construction simulation technology and control theory are becoming increasingly mature, in the present paper has been based on the research results, take Tie Luo Ping cable-stayed bridge as the object of study, cable-stayed bridge construction control simulation, the main sub-cantilever beam gating process control issues was also discussed.This paper first reviews the history of the development of domestic and international cable-stayed bridge, the construction of a cable-stayed bridge control of the domestic and international situation and development prospects, and then introduced a cable-stayed bridge construction control of the basic theory, implementation procedures, as well as the contributing factor of bridge construction control were recommended.Second, the analysis and calculation method of structure model were discussed in-details, some problems were taken into account, according to the design theory of the cable-stayed bridge and to the feature of the cantilever construction, emulation analysis for construction sequences of cable-stayed bridge is presented. A forward-analysis method is introduced, in which the actual construction process is considered, by all conditions of the cable force, the internal force, displacement values, and choose a representative of the cable and beam construction research shows that in the process of construction some changes in the law for the implementation of Construction Control laid the theoretical trajectory.Finally, Tie Luo Ping Bridge Construction Programme and construction technology, construction of the main elements of the monitoring and operation processes was elaborated, detailed description of the actual construction process should be noted that the issue of control and accuracy. As in the actual casting of cantilever beam in the front of the elevation changes of the simulation calculation and the elevation changes are some differences, thereby affecting the determination of elevation model legislation, this paper introduces BP artificial neural networks to the construction beam of the front elevation changes predict, it will affect the beam elevation changes in factors such as distance has been the construction of the tower beam from the center of, the corresponding tension cables third the size of tension when the atmospheric temperature, the elevation changes theoretical calculation as input samples , will be measured as output samples elevation changes of BP network training, use of BP network with the input parameters and output parameters in nonlinear relationship between the characteristics of the follow-up beam in the third tensioned cables after the elevation changes forecast for the construction control reference. Through the predictive value compared with the measured values show that the two good agreement, BP will prove that the neural network for cable-stayed bridge construction control is reasonable and feasible. In addition to summing up and some of the reasons for the prediction accuracy of the network for reference.更多还原