【作者】 周建中；

【导师】 赵鸿铁；

【作者基本信息】 西安建筑科技大学 ， 结构工程， 2003， 博士

【摘要】 结构振动混合控制是一种新型的抗震技术，它不仅适用范围广，而且具有很好的控震性能。因此，国内外已有越来越多的专家学者投入到此方面进行研究，并已取得了一些成果。但从总体而言，目前国内外在抗震结构混合控制减震技术方面还处于试验及理论研究阶段。运用模糊-神经元网络控制技术，进行抗震结构混合控制方面的研究将开拓结构控制的一个全新领域，有关这一领域内的问题有待得到解决完善。 本文利用模糊-神经元智能控制技术，旨在研究PTMD(或加入模糊控制器的PTMD)与消能减震(或基于神经元的消能减震)相结合的广义混合控制的设计理论和方法。主要完成了以下工作： 1)分别建立基于模糊性原理的PTMD与基于模糊性原理的消能减震结构运动模糊微分方程，运用模糊微分方程理论求得其解，分析各自存在的问题。 2)建立PTMD与消能减震相结合的混合控制的运动模糊微分方程，求其解；分别与PTMD、消能减震系统相比较；建立PTMD与消能减震相结合的混合控制系统的设计理论和方法。 3)建立基于神经元网络的抗震结构混合控制的理论和方法；进行基于神经元网络技术的消能减震器的优化设置研究，分析其有效控制的振型数量。 4)运用模糊控制器的设计原理，进行PTMD模糊控制器的设计。 5)分别进行PTMD与基于神经元的消能减震系统相结合的混合控制、SATMD(带模糊控制器或神经-模糊控制器的PTMD)与消能减震相结合的混合控制、SATMD(带神经-模糊控制器的PTMD)与基于神经元的消能减震系统相结合的混合控制的研究，并与已有的试验结果对比，确立各自的优缺点，建立各自的设计理论和方法。 6)从广义混合控制入手，利用本课题的研究成果，建立抗震结构智能混合控制方案选择模糊专家系统的基本理论。 7)针对1、2、3、5情况，建立基于MATLAB语言的抗震结构智能混合控制的仿真系统。 本文的创新之处在于： 1）国内外首次提出了抗震结构智能控制、狭义混合控制、广义混合控制等概念，并提出了理论上主动控制的效果最好，混合控制是一种实践上最好的控震形式的新论断。 2）国内外首次研制开发了PTMD模糊控制器，并应用于PTMD中，改善了PTMD制频范围。 3）国内外首次提出了对粘弹性阻尼结构进行神经网络优化设置的新方法，以考虑不同地震动特性的影响。该方法是对传统优化方法的进一步发展。 4）国内外首次提出了基于弹塑性时程分析进行抗震结构模糊建模及模糊控制规则自动提取的方法。该方法可以推广到利用所有成熟程序自动提取模糊控制规则以进行结构模糊控制。 5）国内外首次提出了应用神经网络技术形成减震结构模糊控制规则的关系生成方法和推理合成算法，从而实现了神经网络驱动的混合控制结构体系模糊推理。 6）编制了一套基于模糊的广义混合控制结构体系的弹塑性时程分析计算机程序，该套程序能同时考虑了消能器非线性和结构非线性。 7）建立了抗震结构智能混合控制的方案选择模糊专家系统的基本理论；并建立了基于MATLAB语言的抗震结构广义混合控制的仿真系统。 8）国内外首次提出了P皿0（或SATM切与消能减震（或基于神经元的消能减震）相结合的广义混合控制设计理论和方法。更多还原

【Abstract】 Structure vibration control is a new kind anti-earthquake technique; it not only applies widely, but also has a very good effect of earthquake control. So at home and abroad, more and more experts study on this aspect, and acquire many achievements. But in sum, now at home and abroad, seismic reduction technique of hybrid control for aseismatic structure is still placed in experiment and theories research stage. Proceeding the study on hybrid control of aseismatic structure will carve out an all-new realm of structural control by applying fuzzy-neural network control technique, and relevant problem in the realm will be solved or consummated.Fuzzy-neural network control technique is applied, and the major aim of this paper is to study on the design theory and method of generalized hybrid control, which is made up of PTMD system (or PTMD system with fuzzy controller) combined with the energy dissipation system (or the energy dissipation system based on neuron). The following work is done:1) Fuzzy motion differential equations are established for the PTMD (Passive Tuned Mass Damper) system based on fuzzy theory and the energy dissipation structural system based on fuzzy theory respectively. With applying the theory of fuzzy differential equations, their solutions are acquired. Their problems that existed are analyzed respectively.2) Fuzzy motion differential equations are established for hybrid control system, which is made up of the PTMD system combined with the energy dissipation system. Their solutions are acquired. By contrasting them with the PTMD system and the energy dissipation system respectively, the design theory and method is established for hybrid control system, which is made up of the PTMD system combined with the energy dissipation system.3) The design theory and method is established for hybrid control system of aseismatic structure based on neural network. Optimum installation study for energy dissipation damper based on neuralnetwork technique is carried through, and the quantity of modes of vibration controlled effectively is analyzed.4) By applying the design principle of fuzzy controller, PTMD fuzzy controller is designed.5) The following hybrid control systems are studied respectively: i) PTMD system combined with the energy dissipation system based on neuron, ii) SATMD system (PTMD system with fuzzy controller or neural-fuzzy controller) combined with the energy dissipation system, iii) SATMD system ( PTMD system with neural-fuzzy controller) combined with the energy dissipation system based on neuron. And then by contrasting theoretical research results of these systems with test results existed, merits and shortcomings of these systems are found out respectively, and in the meantime the design theories and methods of these systems are also established respectively.6) By starting with generalized hybrid control, and by making use of research results of this paper, general theories of fuzzy expert system on the choice of project for intelligent hybrid control of aseismatic structure are established.7) By aiming at K 2> 3^ 5 circumstances, simulation systems based on MATLAB language for intelligent hybrid control of aseismatic structure are established.In this paper, groundbreaking theories are shown in the following:1) The concept of intelligent control of aseismatic structure is put forward at home and abroad for the first time. The concept of generalized hybrid control and that of the opposite are put forward at home and abroad for the first time. Seismic reduction effect of active control is best theoretically, but as a kind of seismic reduction form, hybrid control is best practically, this new judgment is put forward at home and abroad for the first time.2) PTMD fuzzy controller is developed at home and abroad for the first time, and the scope of frequency control get improved by applying it to the PTMD system.3) A new kind of optimum installation method of structure with viscoelastic damper applying neural network technique, which the influence of更多还原