【作者】 程华强；

【导师】 罗尧治；

【作者基本信息】 浙江大学 ， 结构工程， 2014， 博士

【摘要】 本文在智能张力结构和张弦梁结构的发展和应用基础上,提出一类具有广泛工程应用前景的智能张力结构：自适应张弦梁结构(ABSS),即在张弦梁结构的基础上,用可调伸缩杆替换原结构中的竖向撑杆或拉索作为作动器,并附加测试结构工作状态的传感器和分析处理测试信息的控制器形成的智能结构。为保证该类结构的正常工作,本文综合运用随机搜索、线性力法、非线性有限元法和模糊决策等理论与数值方法,针对该类结构控制理论、静力性能、优化设计以及传感器优化布置等四个方面中存在的共性科学问题及关键技术挑战展开深入研究和探讨。采用结构力法,推导了自适应张弦梁结构的理论控制模型,给出结构在恒荷载作用下的几何构形控制方程,以及在均布活荷载和风荷载作用下的工作状态控制方程。采用有限单元法,推导了考虑作动器耦合效应的结构有限元方程,并通过灵敏度分析法建立了结构几何构形和工作状态线性控制的优化模型,采用模拟退火算法搜索得到线性控制的最优解,在此基础上,引入一个非线性有限元计算重新评估控制效果并加以修正的迭代过程,实现自适应张弦梁结构的单目标非线性优化控制。根据决策者对内力、位移以及能耗等控制目标的偏好要求,建立了考虑目标优先级要求的自适应张弦梁结构多目标控制模型。采用模糊决策的理论,用满意度和满意度序来描述控制模型中目标的优化程度和目标之间的优先级,来辅助决策者在相同的量纲上评估结构的控制目标。根据分层交互优化的思路,通过基于目标满意度序和目标加法模型的交互式模糊优化算法来辅助决策者进行先验偏好信息和局部偏好信息的确定和引入,实现自适应张弦梁结构的多目标控制。自适应张弦梁结构根据作动器的布置方式可划分为撑杆式和拉索式,且考虑拉索式自适应张弦梁结构中拉索的可连续或不连续。研究两类结构在均布荷载和非均布荷载作用下的静力性能,发现撑杆式自适应张弦梁结构的调控效果和调控效率均最优。以撑杆式自适应张弦梁结构为研究对象,分析结构静力性能随几何构形参数、截面特性参数和作动器参数等参数的变化规律,并以此为设计依据,提出基于结构性能的优化设计方法,对结构进行设计。以自适应张弦梁结构的调控性能最大为优化目标,建立考虑结构可控性的作动器优化布置模型,并通过序列法和随机搜索算法求解获得作动器的最优布置方案。通过分析作动器优化布置模型的计算复杂度,建立学习策略为结构非线性控制算法选择合适的初始解,以减小算法的迭代次数来提高算法的运算效率；根据计算机多核运算的思路,采用操作并行和试验并行策略对模拟退火算进行改进,加快了搜索算法的搜索速度。本文的研究工作完善和推进了自适应张弦梁结构控制理论和优化设计理论的发展,为该类结构的工程应用提供了理论基础和技术支持,同时本文的研究是基于自适应张弦梁结构,但又不依赖于该结构,研究成果也适用于其他类型智能张力结构。更多还原

【Abstract】 Based on the progress and applications of smart tensile structure and beam string structure (BBS), the adaptive BSS (ABBS) is put forward as a type of widely used smart tensile structure. Based on BBS, ABBS is a smart structure which replaces the vertical poles or cables of the original structure with adjustable bars used as actuators, and adding additional sensors testing the status of the structure and controllers processing test information. In order to guarantee the normal working condition of the structure, the thesis has carried an intensive research and investigation towards the common scientific problems and key technical challenges of control theory, static performance, optimization design and optimal arrangement of sensors with theory and numerical methods such as stochastic search algorithm, linear force method, nonlinear finite element method and fuzzy decision.With the force method, the theory control model of ABSS, the geometric configuration equations under dead loads and the working-status control equations under uniform live load and wind load have been put forward. With the finite element method, finite element equation of the structure with consideration of actuator coupling effect is deduced. An optimized model of geometry configuration and working-status linear control is established with the analysis of sensitivity and achieve the optimal solution of the linear control using simulated annealing algorithms. The nonlinear finite element method is introduced to reevaluate the control effect and modify the iterative process in order to achieve the single-objective nonlinear optimization control of ABSS.A multi-objective control model of ABSS considering the priority of the objectives s is developed. The objectives include internal force, displacement and energy consumption. Using fuzzy decision theory, the satisfaction degree and the order of satisfaction degree are introduced to describe the priority of the objectives. The two parameters can help to evaluate the control objectives of the structure in the same dimension. According to hierarchical interactive optimization, the interactive fuzzy optimization algorithm based on the satisfaction degree order and the additive model of the objectives is proposed to aid the decision-maker to decide and introduce the priori and local preference information and finally achieve the multi-objective control of ABBS.According to the arrangement of actuators, ABSS be divided into bar type and cable type and the bar type can be continuous and discontinuous. The control effect and control efficiency of the bar type is found to be best with the study of the static performance of the two types of structures under uniform load non-uniform load. The changing law of the static performance with parameters such as geometric configuration parameters, section characteristic parameters and actuator parameters is analyzed. The optimization design method based on the structure performance is put forward with the above work.To optimize objective as the best controllability performance, an optimal arrangement model of actuators is set up in consideration of the controllability of the structure. The model is solved with the sequence method and random search algorithm. The appropriate initial solution is chosen with the learning strategy and the analysis of the computational complexity of the optimal arrangement model. The initial solution can reduce the number of iterations and improve the efficiency of algorithm. Some improvements to accelerate the speed of search algorithm are proposed according to the idea of multi-core computer operation for improving the parallel operating of simulated annealing algorithm and test paralleling strategy.The study has advanced the development of the control theory of ABSS and the theory of optimization design, and has provided theoretical basis and technical support for the engineering application of this kind of structure. Meanwhile, the study can also be applied to other types of smart tensile structures though it is based on ABBS.更多还原