【作者】 王浩；

【导师】 范启富；

【作者基本信息】 上海交通大学 ， 控制理论与控制工程， 2009， 硕士

【摘要】 多学科设计优化技术(MDO)是针对复杂工程问题的优化提出的一种方法,目前主要应用在航空航天业的复杂产品的设计优化问题上。浮空器一般是指比重轻于空气的、依靠大气浮力升空的飞行器。随着科技的不断发展,浮空器在全球的宽带连接、空间观测、大气测量、全球资源监控以及军事侦察等方面的需求将会数不胜数。本文对多学科设计优化技术体系和方法应用进行了研究,并在此基础上将多学科设计优化技术和方法引入自动控制系统设计领域,实现了PID控制系统和浮空器飞控系统的多学科设计优化,有效地提高了系统的性能。本文首选综述了国内外关于多学科设计优化技术方法的理论及工程应用的现状,对多学科设计优化技术体系做了详细的介绍,论述了多学科设计优化技术体系的定义、组成及实质,重点总结了目前存在的多学科优化方法,并对每个方法的流程和特点进行了分析。然后选取最为常用的两类多学科优化方法应用于某导弹设计优化方案,分析了优化过程和优化结果,验证了学科分解,系统解耦等多学科优化方法的可行性和有效性。同时使用iSIGHT中不同的优化算法对设计方案进行了探索优化,并对结果进行分析,取得了良好的效果。最后将matlab和Simulink与多学科优化软件iSIGHT集成,实现了对任意复杂三阶系统的PID控制系统的优化仿真,并且能通过该仿真系统寻找特定优化目标下的最优PID参数,实现了自动寻优。文章最后研究了六自由度的浮空器飞行控制系统数学模型,并对其进行降阶简化并仿真,运用已得到的matlab/iSIGHT联合仿真优化系统对其进行探索性的优化设计,取得了一定的优化效果,进一步实现了自动控制系统设计的自动化。更多还原

【Abstract】 Multidisciplinary design optimization (MDO) is a method developed for complicated engineering problem, which is mainly used in aeronautics and astronautics. It has developed rapidly and has attracted lot of interest in domestic and international scale. Airship is often used to represent a kind of aircraft which is lighter than the air and ascends depending on buoyancy. With the development of advanced technology, airship is going to be used in more and more occasions such as the global internet、atmosphere measuring、global resource monitoring and military scout. This dissertation focuses on the usage of MDO, and combines MDO with control system design, achieve the optimization goal for a PID control system and a flight control system for airship.The paper first introduces the research on MDO nowadays and its use on different realm、defines the method、summarizes all the MDO technology today and analysis their features. Afterwards, it uses two wildly used MDO methods to optimize a design case on a kind of rocket. Therefore, the feasibility and validity of MDO is proved. At the end of this part, compares different optimization searching method in iSIGHT and comes into some useful conclusions. Then the dissertation combines matlab、Simulink in iSIGHT platform, uses this platform and MDO method to search best design of a PID system. At last, the dissertation introduces an airship model of six degrees of freedom and uses the platform given to optimize it, comes into some useful conclusions.更多还原