摘要
针对一类存在作动器概率性故障的分布式驱动近空间飞行器的控制问题,提出了一种基于改进的蝙蝠算法的容错控制方法。首先,针对飞行器的小扰动线性化模型,设计虚拟控制器,使得系统能够较精确地跟踪给定的参考模型;然后,在考虑作动器概率性故障的情况下,结合整数规划理论,将飞行器执行阵列控制指令的设计问题转化为一类包含整数约束的优化问题,并利用结合了莱维飞行机制的蝙蝠算法进行求解;最后,通过仿真实验验证所提方法的必要性和有效性。仿真结果表明,与未考虑容错策略的控制方法相比,所提出的容错控制方法在作动器出现概率性故障的情况下具有良好的跟踪性能;且莱维飞行机制能够有效改善蝙蝠算法的搜索效率。
According to the control allocation problem of a class of distributed driving near space vehicles with actuator probabilistic faults, a fault tolerant control method based on the improved bat search algorithm is proposed. Firstly, according to the small disturbance linearization model of the flight vehicle, the virtual controller is designed for making the system track the given reference model accurately. Afterwards, considering the actuator probabilistic faults and combining with the theory of integer programming, the executive array control command design is transformed into an optimization problem including integer constraints. Then, the bat algorithm with Lévy flights is utilized to solve the problem. Finally, a simulation experiment is performed for verifying the necessity and effectiveness of the proposed method. Simulation results illustrate that compared with the control method not considering fault tolerant strategy, the proposed fault tolerant control method has better tracking performance under the actuator probabilistic faults. Meanwhile, the Lévy flights can improve the search efficiency of the bat algorithm effectively.
作者
路遥
王青
梁小辉
董朝阳
LU Yao;WANG Qing;LIANG Xiao-hui;DONG Chao-yang(Beijing Aerospace Automatic Control Institute,Beijing 100854,China;a.Department of Aeronautic Science and Engineering/b.Department of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China)
出处
《控制工程》
CSCD
北大核心
2018年第9期1656-1661,共6页
Control Engineering of China
基金
国家自然科学基金(61374012)
关键词
近空间飞行器
容错控制
整数规划
蝙蝠搜索
Near space vehicle
fault tolerant control
integer programming
bat algorithm
