【作者】 赵林；

【导师】 何玉敖；

【作者基本信息】 天津大学 ， 结构工程， 2003， 硕士

【摘要】 结构振动半主动控制是目前性能价格比最高、最具工程应用前景的一种结构控制方法。它既可以克服主动控制需要大量外部能源的缺点,又基本沿用被动控制措施,同时具有良好而可靠的控制效果。虽然目前出现了多种控制方法和控制装置,但是由于土建结构振动控制问题的特殊性和复杂性,使得许多理论研究成果无法应用于工程实践当中。因此,在理论研究的基础上,结合结构工程实际,对结构振动半主动控制及其实用性进行深入研究是具有重要理论价值和工程实用价值的。本文针对上述问题,借鉴相关学科的最新研究成果,对目前半主动控制中存在的主要问题分别提出新的解决方案。其中主要的研究性工作包括:利用模态变换技术,基于单步状态预测控制律,提出了多自由度结构体系主动变刚度阻尼(AVSD)多振型开关控制律;在神经网络对MR阻尼器恢复力特性精确辨识的基础之上,基于LQG控制(即线性二次型高斯最优控制算法)提出了MR阻尼器应用于半主动控制的跟踪控制算法;阐述了人工神经网络在结构振动控制中的应用进展,并提出了基于BP(Back Propagation)神经网络的跟踪辨识方法;对结构振动半主动控制的时滞问题进行了探讨,并介绍了两种切实可行的时滞补偿方法,即:移相法和状态预测补偿法;针对主动变刚度阻尼AVSD半主动控制控制装置的优化设置问题,基于多模态优化设置准则,提出了一种具体的控制装置优化设计方法;最后,探讨了结构振动半主动控制工程应用中所面临的几个问题,并针对上述问题指出了半主动控制工程应用中值得进一步研究的课题及其发展的方向。更多还原

【Abstract】 At present, It it has been proved by theory and experiment that structural control as a new technologythe semi-active control is a validthe most effective and economical method to reduce the seismic disaster. As one method in of the structural control, the semi-active control combines the active control and with the passive control, and it can approach to the control effectiveness of the active control with only little energy, this advantagewhich makes it the developmental tendency of the structural control. But with the strong nonlinear speciality, the semi-active control is not perfect in theory. Thus it is very important in theory and practice that the semi-active control technology of structural vibration and its application practicability are studied deeply.Following points are the achievement of my research:Based on single-step state predicting control law, AVS/D multi-modal switching control law for many degrees of freedomMDOF structural systems is presented.Put forward a tracking control algorithm based on LQG control theory for MR dampers to be applied in the structure vibration semi-active control field of structural vibration.Analyze the application development of Artificial Neural Network in the control field of structural vibration at firston structure vibration control field, and then a precise method of identifying to the structure system identifications based on BP Neural Network is suggested. Ploblems Problems about time-delay of the active or semi-active control systems are studied firstly, thenand then put forward two methods for time-delay compensation are introduced.Optimal placement of AVS/D control devices in the semi-active control system is discussed. Based on multi-modal optimal placement criterion, a method of optimal placement is given.Analyze the current practicabilityapplication problems of the semi-active control of structural vibration, based based on thesethe above ploblemsproblems, some future research directions are pointed out.更多还原