【作者】 姜南；

【导师】 李忠献；

【作者基本信息】 天津大学 ， 结构工程， 2008， 博士

【摘要】 磁流变(MR)阻尼器就作为一种新型的半主动控制装置,具有构造简单、速度响应快,阻尼力连续可调等优点。基于MR阻尼器的半主动控制系统可以有效控制结构的地震反应。本文对应用MR阻尼器的相邻建筑地震反应半主动控制进行了理论与试验研究,主要研究工作和成果如下:(1)在不同振幅和不同频率正弦波激励下,对施加不同励磁电流的MR阻尼器进行了动力性能测试试验。提出了MR阻尼器的修正的sigmoid模型,该模型表达式简单、含义明确、参数较少,其参数可以根据阻尼器试验曲线做灵活的修改,能够准确描述MR阻尼器的动力特性。(2)基于Kalman滤波器的特点,将加速度反馈方法扩展应用于IOC、COC和LQR等算法。数值分析表明,基于不同算法的加速度反馈方法不需要被控结构响应的全状态反馈就能达到原控制算法对被控结构的控制效果。说明采用该方法可以减少控制系统中传感器的数量,降低控制系统的软硬件成本,适于实际工程应用。(3)基于MR阻尼器控制力幅值对被控结构控制效果的影响,指出应按照优化后的控制力幅值设计制造MR阻尼器。计算表明,当控制力幅值达到一定值时,被控结构的控制效果提升缓慢,说明可以用较小的MR阻尼器控制力幅值达到理想的控制效果。由于控制力幅值将作为MR阻尼器设计制造的依据,故按照优化后控制力幅值制造的阻尼器体积更小、重量更轻、MR液需求量更少,方便MR阻尼器在工程实际应用中的设置、安装和调试。(4)分别对半主动控制、开关控制和两种被动控制方法下,MR阻尼器对相邻结构地震反应的控制效果进行了仿真分析计算。计算结果表明,基于MR阻尼器的半主动控制系统对相邻结构的控制效果在四种控制方法中最好,说明半主动控制方法考虑了MR阻尼器在出力范围内控制力连续可调的特点,充分发挥了MR阻尼器的性能,对相邻结构是一种非常有效的控制方法。(5)在不同地震激励下采用两种被动控制和三种加速度反馈控制方法,应用MR阻尼器对相邻结构模型进行了地震模拟振动动台试验。试验结果表明,基于MR阻尼器的半主动控制系统可以有效的降低模型结构的振动响应,半主动控制系统充分利用了MR阻尼器的性能,对相邻结构达到了较好的控制效果。而且三种加速度反馈方法仅需要结构的加速度响应作为反馈信息,大大减少了试验中传感器数量,使半主动控制更容易应用于工程实际。更多还原

【Abstract】 Magnetorheological (MR) fluid damper is one of the new semi-active control devices. It has great features of mechanical simplicity, rapid response, and adjustable damping force capacity. The semi-active control system can effectively control the seismic responses of structures based on MR dampers. In this dissertation, semi-active control of seismic responses of adjacent buildings using MR dampers is investigated through theoretical and experimental study, the main work and achievement are as follows:(1) The dynamic performance test of MR damper with different applied current levels is conducted under sinusoidal excitations with different frequencies and amplitudes. A modified Sigmoid model, which has simple expression, clear meaning, and less parameters, is proposed. The parameters in this model can be modified according to the performance test results.(2) Based on the characteristics of Kalman filter, the acceleration feedback control method is introduced to Instantaneous Optimal Control algorithm, Classical Optimal Control algorithm and Linear Quadratic Regulator Optimal Control algorithm. The numerical analysis results indicate that the acceleration feedback based control algorithms can obtain similar control effects as full state feedback control algorithms, which results in fewer sensors needed in the control system and more suitable for engineering application.(3) After investigating the relationship between the control effectiveness and the amplitude of the control force, it is suggested that the optimized control force amplitude should be designed for the MR damper used in a particular control system. The simulation results indicate that the control effectiveness improves slowly with the increasing of the control force when it reaches a certain value. And because the control force amplitude is the basis for designing and manufacturing of MR damper, so the volume and weight of manufactured MR damper based on the optimized control force amplitude can be reduced and less MR fluid is required in the device. Therefore, it would be more convenient for installing and debugging the MR damper in practical engineering application.(4) The application of MR damper in the adjacent structures is studied. Adjacent structures employed with semi-active control algorithm, switch control algorithm, and two passive control algorithms are simulated. The numerical results show that the semi-active control method can obtain the best control effectiveness within four control algorithms. The results further prove that the semi-active control algorithm takes into account the continuous adjustability characteristic of the MR damper, and thus it can make full use of the control devices and obtains great performance on the control of adjacent structure.(5) A series of shaking table tests of the adjacent model structures installed with MR damper were carried out under different seismic excitations. Two passive cases and three acceleration feedback based control cases are considered. The experimental results show that the semi-active control systems can reduce the structural responses significantly. And the acceleration feedback based control algorithms can greatly reduce the number of sensors required in the control system, which makes it more convenience for applying the control algorithms in engineering application.更多还原