充液航天器液晃抑制及其姿态控制研究

发布时间：2018-01-08 12:08

本文关键词：充液航天器液晃抑制及其姿态控制研究　出处：《北京理工大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着航天技术的不断发展,航天器需要携带更多的液体燃料以实现更长的在轨工作时间和更精确、快速的姿态控制。在充液航天器中,燃料的消耗不仅会产生整体质心的漂移,造成推力偏心,而且部分充液贮箱内的液体在姿态机动时还会发生晃动现象,从而产生较大的耦合干扰力矩,进而对航天器的姿态稳定性造成不良影响,甚至导致航天器任务的失败。因此,研究大型充液航天器的姿态稳定性及其液体晃动的抑制算法至关重要。本课题充分考虑了充液航天器变轨期间大喷气推力的情形,针对其姿态控制系统进行研究。首先对充液航天器模型进行了详细分析,明确指出了姿态机动及液体晃动的控制目标,在此基础上,基于Lyapunov稳定性理论提出了适用于充液航天器的直接状态反馈控制律,进一步建立了降维状态观测器以重构液晃状态,实现了状态反馈的工程要求。其次,基于最优控制的设计思想,采用状态依赖黎卡提方程(State Dependence Riccati Equations,SDRE)法,在不进行线性化的前提下,设计了姿态系统的最优控制器,并且考虑外部不确定性干扰的问题,增加积分滑模变结构控制,有效的抑制了液体晃动和外部扰动。最后,本文设计了基于线性矩阵不等式(Liner Matrix Inequality,LMI)求解和高阶奇异值分解(Higer-Order Singular Value Decomposition,HOSVD)的张量积(Tensor Product,TP)模型变换算法,通过进行离线的数值采样计算,求解状态反馈控制增益阵,减少了控制器设计过程中对被控系统数学模型的依赖,不仅简化了参数设计过程,而且对内部参数不确定性具有良好的鲁棒性。仿真结果表明,基于降维状态观测器的状态反馈控制律、基于SDRE的最优积分滑模控制律和TP模型变换控制律能同时满足本文所述的姿态控制稳定性和精确性。
[Abstract]:With the continuous development of space technology, spacecraft need to carry more liquid fuel to achieve longer on-orbit working time and more accurate, rapid attitude control. The consumption of fuel will not only cause the drift of the whole center of mass, cause thrust eccentricity, but also the liquid in the liquid filling tank will be sloshing in attitude maneuvering, which will produce a large coupling disturbance torque. Thus, the attitude stability of spacecraft will be adversely affected, and even the spacecraft mission will fail. It is very important to study the attitude stability of large liquid-filled spacecraft and the suppression algorithm of liquid sloshing. Firstly, the model of liquid-filled spacecraft is analyzed in detail, and the control targets of attitude maneuver and liquid sloshing are clearly pointed out. Based on Lyapunov stability theory, a direct state feedback control law for liquid-filled spacecraft is proposed, and a reduced-order state observer is further established to reconstruct liquid state. The engineering requirements of state feedback are realized. Secondly, the design idea based on optimal control is proposed. The state Dependence Riccati equations (SDREs) method is applied to the state dependent Riccati equation without linearization. The optimal controller of attitude system is designed and the integral sliding mode variable structure control is added to reduce the liquid sloshing and external disturbance. In this paper, we design a liner Matrix Inequality based on linear matrix inequality (LMI). LMI) and higher-order Singular Value Decomposition. The tensor product Tensor Product (TP) model transformation algorithm of HOSVD is used to solve the state feedback control gain matrix by off-line numerical sampling calculation. It not only simplifies the process of parameter design, but also has good robustness to the uncertainty of internal parameters. The simulation results show that. Based on the state feedback control law of reduced order state observer, the optimal integral sliding mode control law based on SDRE and the TP model transformation control law can satisfy the stability and accuracy of attitude control described in this paper.
【学位授予单位】：北京理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：V448.2

【相似文献】