【作者】 曾庆华；

【导师】 张为华；

【作者基本信息】 国防科学技术大学 ， 航空宇航科学与技术， 2003， 博士

【摘要】 由于微型飞行器(Micro Air Vehicle，简称MAV)受到体积和重量限制，在设计过程中如何从系统角度出发实现多学科综合与集成，显得尤为重要。而飞行控制系统作为飞行器设计中学科间联系最为紧密的子系统，对飞行器性能影响很大。飞行控制系统设计存在两大关键技术问题：(1)如何在飞行控制系统设计中考虑学科间的相互影响，保证控制系统性能的鲁棒性：(2)如何在多学科设计优化研究中考虑控制学科的作用，保证飞行器系统整体性能最优。本文以MAV飞行控制器设计方法研究为背景，以鲁棒控制理论为指导，以多学科集成为基础，开展了MAV飞行控制器多学科设计方法研究。主要研究内容和成果概括如下： 建立了MAV多学科系统分析模型，分析了多学科设计参数之间的数据耦合关系；从MAV非线性微分方程组描述形式出发，研究了状态方程区间模型中不确定性表示方法，推导得到MAV广义对象一般表达式：针对MAV中存在的学科间参数耦合、参数变化、未建模动态等一系列不确定性问题，提出了MAV飞行控制器多学科设计方法。 针对固定翼MAV，开展了以尺寸最小为目标函数的多学科设计优化研究，研制了MAV原型样机，通过理论分析和飞行试验对不同布局MAV飞行性能进行了研究。提出了在MAV多学科集成环境中利用试验设计理论建立MAV不确定性模型的方法，进行了MAV纵向和横侧向通道的鲁棒飞行控制器设计与评估。 研究提出控制学科在回路的多学科设计优化方法。从响应面近似模型出发，推导得到将多学科设计参数直接作为摄动量的不确定性模型，减小了不确定性描述的保守性，降低了控制学科计算量，提高了优化设计算法的收敛性，将鲁棒控制优化和并行子空间优化方法相结合，实现了控制学科与其它学科的并行设计与优化，成功将该方法应用到某无尾布局MAV气动、控制学科优化设计中。 以MAV形变控制鼓包装置几何配置优化问题为例，将MAV中气动学科和控制学科研究相结合，在MAV多学科设计框架中，利用遗传算法解决了MAV气动、控制学科并行设计中的离散优化问题，开展了新型气动控制方案探索性研究。 针对MAV飞行控制器特点，研究了飞行控制器在回路的数学仿真和半实物仿真方国防科技人学研究生院学位论文法，建立了MA入/飞行控制器仿真验证平台框架。 论文研究形成的基于多学科综合与优化的M户入z飞行控制器设计方法，能够实现控制学科和其它学科的并行设计与优化，从而有效保证M户y飞行控制器的鲁棒性和飞行器整体性能最优，为今后深入开展M户L\/飞行控制问题研究提供了实用、有效的理论和方法，具有一定应用前景。关键词:微型飞行器多学科设计优化飞行控制鲁棒控制第11页更多还原

【Abstract】 Because of strict limitation to volume and weight, it is especially important for Micro Air Vehicle(MAV) to synthesize and integrate multi disciplines from the point of system integration during design process. As a sub-discipline which has most coupling and most close connection among disciplines, flight control system has extremely important influence on aircraft’s performance. There exist two key problems in multidiscipline optimization design. One is how to consider mutual influence among disciplines to ensure robustness of control performance; The other is how to take control discipline into account to ensure optimization of overall performance. With MAV flight control system design as background, robust control theory as guidance and multidiscipline synthesis as basis, the research of multidiscipline optimization design method is carried out in this paper. The main contents and results are as follows:Firstly, the MAV multidisciplinary system analysis model is built. And the coupling relationship among multidisciplinary design parameters is analized. Beginning with MAV nonlinear differential equations, this paper studied the uncertainty expression method in status space model and the general expression of MAV generalized object is deduced. To the question of strucrual and nonstructual uncertain problems such as parameter coupling, parameter variation and unmodelled dynamics, the multidisciplinary design method for MAV flight control system is brought forward in this paper.Taking fixed-wing MAV as object, the multidiscipline optimization research is carried out with the aim of minimum size. MAV prototypes are developed. Flight performance of different configure MAV is studied with both theory analysis and flight test. Furthermore, method and process of founding the MAV uncertainty modeling using experimental design method under multidisciplinary integration environments is illustrated. The longitudinal and lateral channel robust flight controllers are designed and evaluated.In this paper, multidisciplinary design optimization method with control discipline in the loop is studied. Beginning with response surface approach model, the uncertainty model usingmultidisciplinary design parameters as perturbations is deduced, which not only decreases conservative characteristics of uncertainty expression, but also reduces computational complexity for control discipline, thus improves the convergency degree of optimization algorithum. After that, the parallel design and optimization between control discipline and other disciplines is realized and applied successfully to parallel optimization design of aerodynamics and control discipline for a given tailess MAV.Taking the problem of MAV geometrical optimization placement controlled by shape-change blisters as example, the problem of discret optimization in the process of parallel design is solved using genetic algorithum by combining aerodynamics and control discipline in the frame of MAV multidisciplinary design. Research of a new aerodynamic control scheme is also carried out in this paper.Considering characteristics of MAV flight controller, method of both mathematic simulation and hardware in the loop simulation is studied. Framework of simulation validation platform for MAV flight controller is also built.Finally, a MAV flight controller design method based on multidisciplinary collabration and optimization is developed. Such method realize the cooperation and optimization design between control discipline and other disciplines, which not only ensure the opimal integrative performance and robustness of MAV flight control system.but also provide an effective and practical theory and method for further studying of MAV flight control problem with good prospective for use.更多还原