【作者】 李雪峰；

【导师】 王建国；

【作者基本信息】 合肥工业大学 ， 结构工程， 2008， 博士

【摘要】 桥梁作为重要的社会基础设施,是抗震防灾、危机管理系统的一个重要组成部分,提高桥梁的抗震性能是减轻地震损失、加强区域安全的基本措施之一,是我国公路交通建设中所面临的重大课题。建筑结构振动控制经过几十年的发展,已被理论和实践证明是抵御地震破坏的有效手段。随着大跨度桥梁的普遍兴建和高效能建桥材料的广泛应用,桥梁结构的振动问题日趋突出。而大跨度斜拉桥由于其本身所具有的一些特殊性,国内外对其进行振动控制的研究水平还远没有达到对高层房屋建筑结构一样的研究程度。所以在深入了解桥梁结构的地震反应特性的基础上,研究合理而可行的控制措施保护桥梁结构免遭地震破坏,将是一个具有极大工程应用价值且时间紧迫的研究课题。随着科学技术的发展,以及人们对振动环境、对产品与结构振动特性越来越高的要求,振动被动控制的局限性就暴露出来了,难以满足人们的要求。主动控制技术由于具有效果好、适应性强等潜在的优越性,很自然地成为一条重要的新途径。本文以大跨度斜拉桥结构模型为研究对象,研究在地震激励下采用主动控制方法进行斜拉桥结构的振动控制,对主动控制方法中的一些关键性问题进行研究,抑制斜拉桥结构的地震响应,增强其动力稳定性,从而达到提高大跨度斜拉桥结构的抗震能力。首先在原有性能评价指标的基础上,进一步完善斜拉桥结构主动控制Benchmark性能评价体系,更加全面的评价不同的斜拉桥主动控制策略对结构的减振效果;研究平衡降阶的理论及其实现,通过对二次型性能泛函指标进行平衡转换,找出平衡系统中对系统能量贡献较大的平衡系统状态变量,进行模型降阶,得到一个能够反映原结构模型动力响应的降阶模型,提出基于内平衡系统二次型性能泛函进行模型降阶,以适应于日益增长的大型土木工程结构实际振动控制的需要。在主动控制系统设计中,研究致动器/传感器的最优位置以及其最优数目,根据致动器对结构系统的状态影响矩阵确定满足控制要求的最优致动器数目;根据系统二次型性能指标和本文提出的传感器优化配置准则,进行致动器/传感器的优化配置,优化计算方法和流程,得到斜拉桥结构振动主动控制设计的致动器/传感器优化配置方案。迭代学习控制是一种比较理想的控制策略,其本身具有某种智能,能够在控制过程中能不断地完善自身,以使控制效果越来越好,逐渐成为令人关注的课题。针对斜拉桥结构的地震响应,结合控制领域内的最新研究成果,将迭代学习控制与其它控制方法相结合,对斜拉桥结构进行有效的控制。首先基于线性二次型最优控制与迭代学习控制相结合的思想,研究线性二次型迭代学习混合控制方法,提高迭代学习控制的收敛速度,提高了控制系统的控制效果;其次基于迭代学习控制与滑模控制策略各自的优点,将两者相结合提出一种新的控制策略,滑模迭代学习控制策略。并且采用二次型迭代学习混合控制方法和滑模迭代学习控制方法对Emerson Memorial大桥进行地震响应主动控制计算,得出Benchmark控制指标,计算结果表明二次型迭代学习混合控制方法和滑模迭代学习控制策略均能够对Emerson Memorial大桥地震响应进行有效地控制,并且控制效果得到了一定的改进。更多还原

【Abstract】 As the important infrastructure, bridge is an important component of the seismic disaster prevention system and crisis management system. Improving the bridge seismic performance is one of the basic measures to reduce earthquake damage and strengthen regional security. Building structure vibration controlling has been proven to be an effective means to withstand earthquake damage by the theory and practice after decades of development. With long-span bridges constructed and extensive usage of high-performance material, bridge structure vibration problem had become more prominent. But because of the special nature of the long-span cable-stayed bridge, the level of bridge vibration control is far from the high-building vibration control. Therefore, base of understanding the bridge seismic response characteristics, studying the reasonable and practicable control measures to protect the bridge structure against earthquake damage will be a great project.With the development of science and the ever-increasing demands for vibration environment and structure vibration characteristics, passive vibration control was difficult to satisfy and exposed its limitation. Active control became a new important ways because of its effective and adaptable. This paper will study bridge structure active control and its key problems against earthquake base of long-span cable-stayed bridge. So improve the bridge anti-seismic capability and enhance its dynamical stability.Firstly, further improve benchmark performance system of cable-stayed bridge active control base of original indicators. More comprehensively evaluates different active control strategy for cable-stayed bridge. Then study the theory of balance reduction and its achieving. Reduce the model by internal balance transform for quadratic characteristic indexes function and find out the state variables which are great contribution to system energy. So obtain a reduced model which can reflect the dynamic response of original model in order to meet the active control of large-scale civil engineering.In the design of active control system, research the optimal number and position of actuators/sensors. Determine the optimal number of actuators according to the state effect matrix by controlling. According to the quadratic performance index and the proposed optimal sensors distribution guidelines, design optimal actuators/sensors configuration and optimize the calculation process. So obtain the optimal distribution of actuators/sensors for active vibration control of cable-stayed bridge.Iterative learning control is a more satisfactory control strategy, with its own intelligence , which be able to constantly improve itself in controlling process. So it is gradually becoming an issue of concern. Based of the latest research results in the control field, combine the iterative learning control and other control strategy to reduce the earthquake response of cable-stayed bridge. Firstly, combine the linear quadratic optimal control and iterative learning control to obtain a new mixed control strategy which named quadratic iterative learning control. It can improve the convergence rate of iterative learning control and improve the effect of control. Secondly, with respective advantages of iterative learning control and sliding mode control strategy, combine them and obtain a new control strategy which named slide mode iterative learning control. Use these two new mixed control strategies to control the Emerson Memorial Bridge against earthquake and calculate the benchmark performance indicators. The result show that new control strategies were able to effectively control the Emerson Memorial Bridge against earthquake and control effect were improved.更多还原