CASIA OpenIR  > 毕业生  > 博士学位论文
基于模糊系统理论的弹性体高超声速飞行器控制方法研究
高俊龙
学位类型工学博士
导师易建强
2017-05
学位授予单位中国科学院研究生院
学位授予地点北京
关键词吸气式高超声速飞行器 模糊系统 区间二型模糊系统 自适应控制 数据驱动
摘要高超声速飞行器具备不低于5马赫的高超声速快速行进、远程响应时间短、机动性高、有效载荷高、可重复利用等优点,具备极高的军民应用前景。其中吸气式高超声速飞行器因可直接利用空气中的氧作为氧化剂而具备较高的有效载重百分比及较高性价比等优势,受到全球研究者的重点关注。由于该飞行器机体物理特性造成的不可忽视的弹性模态、耦合效应及其工作所处空间复杂多变的大气环境等因素,使其模型具有强非线性、快时变、强不确定性等特点,对控制系统的设计及鲁棒性要求带来严峻的挑战。
模糊系统及其控制方法是有效处理含有非线性、不确定性系统的方法。二型模糊系统及其控制方法作为一型模糊的一种改进,可在减少模糊规则数量的同时赋予系统更加优异的不确定性处理能力。
本论文以弹性体吸气式高超声速飞行器纵向模型轨迹跟踪控制为背景,基于模糊系统理论并重点基于区间二型模糊系统深入研究飞行过程中出现的若干关键控制问题,本文主要工作与贡献有:
(1) 针对执行器饱和及系统内部不确定条件下的高超声速飞行器巡航机动控制问题,提出了基于反馈线性化与自适应一型Mamdani模糊高阶滑模控制方案。该方案采用速度、高度滑模面高阶导数信息,推导出反馈线性化控制框架。在该框架下设计了自适应模糊高阶滑模控制器。基于人工经验设计的模糊系统根据飞行状态在线调整高阶滑模控制器参数,提升了控制器抗执行器饱和能力与被动抑制弹性模态激发的能力。
(2) 针对内、外部不确定条件下的高超声速飞行器巡航机动控制问题,提出了基于反馈线性化与间接自适应区间二型Mamdani模糊滑模的控制方案。该方案采用速度、高度滑模面一阶导数信息,推导出反馈线性化控制框架。在该框架下设计了使用多组不同飞行状态信息的自适应区间二型模糊系统用以在线估计补偿含有扰动信息的系统逆动态,并设计了积分滑模控制器作为控制系统的外环控制器。控制律与自适应律根据李雅普诺夫方法设计,保证了系统全局稳定性。
(3) 针对高超声速飞行器由未建模动态及其他不确定性导致的数学建模不精准问题,提出了一种直接采用数据驱动方式生成区间二型TSK模糊神经网络(IT2-TSK-FNN)进行建模与辨识的新方法。该方法可充分利用已有自组织一型模糊方法产生的系统模糊规则与数据集,通过两个学习阶段生成具有更优不确定处理能力及更精准建模与辨识能力的IT2-TSK-FNN。结合三个非线性基准模型建模与辨识实例与已有方法的对比实验,验证了所提出的方法的有效性。
(4) 针对高超声速飞行器关键飞行状态的区间二型TSK模糊模型在不确定扰动条件下的巡航机动控制问题,提出了基于反步法控制架构的新型区间二型TSK模糊滑模控制方案。该方案以反步法为架构,基于模型分析的结论,使用IT2-TSK-FNN建模方法对高超声速飞行器速度、航迹角与俯仰角速率三个直接受弹性模态影响的关键飞行状态开环响应数据进行学习并建立了对应的区间二型模糊模型。除对所有状态子控制系统分别设计滑模控制方案外,在上述三个飞行状态子控制系统中设计了滑模补偿控制器以提升总体控制方案的鲁棒性。子系统控制律通过李雅普诺夫方法进行设计,保证了系统的稳定性。
总体而言,本文深入研究了弹性体吸气式高超声速飞行器在不确定环境中基于模糊系统设计鲁棒控制器的若干关键问题,验证了在系统内部及外部复杂不确定条件影响下的控制系统有效性与鲁棒性,为模糊系统理论在弹性体高超声速飞行器控制应用进行了积极的理论探索。
其他摘要The hypersonic vehicle is with prospect applications in both civilian and military for the advantages such as hypersonic speed, fast remote reaction, high payload and reusabilities. Moreover, a branch called air-breathing hypersonic vehicle (AHV), which could adopt oxygen in the atmosphere directly as the oxidant, has better effective load percentage and better cost-effective to draw global researcher’s focus. However, apart from the rigid body dynamics, due to the air dynamic high nonlinearity, strong uncertainty from noticeable flexible effect coming from the airframe physical characteristics, coupling effects and the complex and varying atmospheric environments, there are existing huge challenges to design a controller with reasonable robustness for the flexible air-breathing hypersonic vehicle (FAHV).
  Fuzzy logic system (FLS) and fuzzy logic control (FLC) methods are the ones which can effectively handle nonlinearity and uncertainty. Moreover, as the improved version of traditional fuzzy logic methods (e.g. type-1 fuzzy logic), the interval type-2 fuzzy logic system (IT2-FLS) and the interval type-2 fuzzy logic control (IT2-FLC) have better ability in dealing with the aforementioned issues with fewer rules.
  This dissertation deeply investigated control-related key problems of the FAHV longitudinal model based on the fuzzy logic and especially the interval type-2 fuzzy logic under the background of trajectory tracking control. The main contributions are addressed as below.
(1) For the FAHV cruise tracking control problem under actuator saturation and internal uncertainty, an adaptive type-1 Mamdani fuzzy high-order sliding mode control scheme (AT1-FHOSMC) is proposed. The AT1-FHOSMC is designed under the feedback linearization framework derived from velocity and altitude sliding surfaces’ high-order derivative information. Human knowledge based type-1 FLS, which adjusts parameters of the high-order sliding mode controller online, is designed to improve the anti-saturation of actuators and the passive suppression of the flexible effects.
(2) For the FAHV cruise tracking control problem under internal and external uncertainty, an indirect adaptive interval type-2 Mamdani fuzzy sliding mode control (AIT2-FSMC) scheme is proposed. The AIT2-FSMC is designed with three adaptive IT2-FLSs which adopt different flight state combinations as the antecedent parts separately to approximate and to compensate the control system’s inverse information embedded with uncertainty. Moreover, integral sliding mode controllers are designed in velocity and altitude channels, respectively. The adaptive law and control law are derived based on Lyapunov synthesis to ensure the symmetric stability of the system.
(3) For the imprecise mathematical modeling of the FAHV due to unmodeled dynamics and other uncertainty, a method, which is used to automatically constructing an IT2 TSK fuzzy neural networks (FNN) for modeling directly from experimental data by two autonomous learning phases, is proposed. The method takes advantage of the well-behaving T1-TSK, models more precisely and behaves better in dealing with uncertainty. Three benchmark modeling and prediction examples are studied to compare with existing methods, the simulation results validate the effectiveness of the proposed method.
(4) For the FAHV cruise tracking control problem by adopting IT2-TSK nominal fuzzy models of the key states under uncertainty conditions, an interval type-2 TSK fuzzy sliding mode control scheme based on backstepping is proposed. The IT2-TSk fuzzy models of the three key states are generated through the IT2-TSK-FNN. All subsystems are designed with sliding mode controllers whereas the key subsystems are designed with a nominal model compensation controller additionally. The control and compensation laws in the subsystems are derived through Lyapunov synthesis and the system stability is ensured.
  Conclusively, this dissertation addresses comprehensive studies on several key problems in designing robust controllers based on fuzzy logic for the FHAV under different types of uncertainty. The robustness and effectiveness of the proposed methods are verified. It provides promising fuzzy logic theoretical investigations on the applications for the FAHV.
学科领域第一研究方向
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/14648
专题毕业生_博士学位论文
作者单位中国科学院自动化研究所
推荐引用方式
GB/T 7714
高俊龙. 基于模糊系统理论的弹性体高超声速飞行器控制方法研究[D]. 北京. 中国科学院研究生院,2017.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
201700601--高俊龙学位论文.p(3657KB)学位论文 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[高俊龙]的文章
百度学术
百度学术中相似的文章
[高俊龙]的文章
必应学术
必应学术中相似的文章
[高俊龙]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。