可重构飞行控制律设计的混合特征结构配置方法研究

将特征结构配置与模糊控制理论相结合对飞机侧向飞行重构控制律进行设计.首先介绍了重构控制律中特征结构配置的原理和反馈增益矩阵的算法实现,在此基础上结合带优化修正参数的无量化模糊控制方法对闭环系统进行鲁棒控制器设计.该控制策略以特征结构配置作为故障后系统的内环控制器,模糊控制器作为外环控制器,使系统获得较好的动态性能和较强的鲁棒性.仿真结果表明系统有效抑制了内部参数摄动对飞行任务的不良影响.     

基于自适应逆的鲁棒容错飞行控制方法研究

研究飞行安全控制问题,可采用动态逆与神经网络模型参考自适应控制理论相结合,对飞机纵向自适应鲁棒容错飞行控制律进行设计,同时采用改进的粒子群算法优化神经网络参数,提高了自适应算法的效率.控制策略采用内-外环的控制结构,内控制回路以逆控制消除系统的非线性性和输出耦合;外控制回路基于模型参考自适应控制的思想,利用改进粒子群优化的神经网络设计前向自适应控制器,以消除逆控制的建模误差和对参数变化敏感的缺点,可使系统获得较好的动态性能和较强的鲁棒性.仿真结果表明采用的自适应鲁棒容错飞行控制方法有效抑制了操纵面故障,消除了对飞行任务的不良影响,保证了安全性能.     

基于特征结构配置的飞翼无人机纵向控制律设计与试飞验证

针对飞翼无人机的纵向、横侧向静稳定性较差,飞机纵向短周期模态稳定性不足等问题,以某小型飞翼无人机为对象,研究了特征结构配置方法,并提出了基于特征结构配置方法的前向增益型、误差积分型控制律设计方案;利用线性模型设计了该飞机的纵向控制律,通过非线性仿真分析、飞翼无人机的试飞验证了该控制方案的可行性与有效性;仿真及试飞结果表明:采用特征结构配置方法设计的飞行控制系统具有较好的控制效果,改善了飞翼飞机的飞行品质.     

直升机横侧向通道模糊自适应控制律设计

根据飞行包线中典型状态的小扰动线性化方程,建立了直升机横侧向系统的T-S模型。采用参数映射设计方法,利用系统的部分状态信息,直接针对规定的系统性能指标,在满足要求的可用参数集中设计控制器参数。根据平行分布补偿原理,设计模糊神经网络控制器,利用各典型状态下选定的参数作为样本,训练模糊神经网络,实现所设计的控制律,使飞机能够在全包线范围内达到要求的性能指标。仿真结果表明,应用所设计的控制律,可以在飞行包线内将系统极点较准确地配置在希望极点附近,系统动态性能指标完全满足规定的要求。表明提出的设计方法可行而且有效。     

基于EA方法的飞行控制律设计与仿真

在飞行控制系统优化设计中,由于系统存在多输入、多输出的模型结构,古典控制方法无法应用到MIMO系统中,尤其现代控制方法设计过程中一次性计算出反馈控制器,导致应用过程中可靠性降低,对古典飞行控制系统控制结构进行优化设计.控制系统内回路采用EA方法,外回路适当引入误差积分控制和前馈比例控制,进而针对广义被控对象直接进行控制律设计与仿真.运用根轨迹方法对控制系统进行验证分析,并根据实际需求对控制律进行适当改进.仿真结果表明,特征结构配置方法能够较好的与古典控制理论相结合,完成指令跟踪控制律设计,且具有很好的鲁棒性和动态特性,新的控制结构应用起来方便有效.     

飞机俯仰运动自抗扰控制器设计

提出了利用自抗扰控制器在大包线范围内设计飞机俯仰运动控制器的新方法．利用二阶自抗扰控制器补偿系统模型扰动和外扰，实现了纵向运动俯仰角变量的跟踪控制．自抗扰控制器直接依据飞机的非线性模型，符合飞机动力学模型摄动大的特点，在很大的包线范围内不需要改变控制器的结构和参数，简化了飞行控制律的设计过程．大包线范围内的仿真结果表明，系统具有良好的动态和稳态性能，控制器具有很强的鲁棒性，为解决大包线范围内的飞行控制问题提供了一种有效的新途径．     

基于模型跟随的神经网络PID飞行控制律设计

为了抑制飞行控制系统的外部扰动和建模误差,应用模型跟随自适应神经网络PID控制方法,进行飞行控制律设计。首先使用RBF神经网络进行飞行系统模型辨识,在线学习系统正向输入输出特性,辨识对象的Jacobian信息;然后应用BP神经网络实时在线调整PID参数,设计自适应神经网络PID控制器,控制飞行状态变量跟随模型输出;最后以F-8飞机纵向飞行控制系统为研究对象进行控制仿真。仿真结果表明,设计的控制器具有很强的自适应和抗干扰能力。     

基于EA／QFT的飞行控制律设计及其稳定性评估

提出特征结构配置和定量反馈理论相结合的鲁俸控制设计方法,并应用该方法对某型飞机横侧向进行控制律设计,然后通过μ分忻对控制律进行稳定性评估。仿真结果表明：该方法使某飞机不仅具有良好解耦效果和动态响应,而且具有较强的鲁棒性。     

自抗扰控制器在超机动飞行快回路控制中的应用

提出了一种利用自抗扰控制器算法设计超机动飞行快回路控制律的新方法．根据自抗扰控制器能够动态补偿系统模型扰动和外扰的特性,在超机动飞行的快回路中引入自抗扰控制器。实现了快变量的动态解耦控制．控制律设计直接依据超机动飞行的强耦合、强非线性模型。在较大的包线范围内不需要改变控制器参数．简化了设计过程．仿真结果表明,系统具有良好的动态和稳态性能,控制器具有很强的鲁棒性．     

超机动飞机的动态建模与控制律设计及仿真

建立了带推力矢量的超机动飞机非线性动态模型,重点分析了气动力、气动力矩以及发动机的建模过程.采用基于神经网络的自适应逆方法,设计了超机动飞机大迎角机动下的控制律.首先应用动态逆方法,分别设计了快慢回路的飞行控制律;然后利用BP神经网络,在线补偿飞机模型不确定性以及外界干扰.眼镜蛇机动的仿真结果表明,所设计的控制律在大迎角机动条件下具有良好的控制性能,能够保证闭环系统的稳定性.     

Mixed time- and frequency-domain robust eigenstructure assignment

This paper considers robust eigenstructure assignment for multivariable systems. The time-domain performance specifications provided by eigenstructure assignment and robust performance specifications in the frequency domain considered by H infinity control are combined to realize joint optimal robust control design. A parametric expression for state-feedback eigenstructure assignment is introduced on the basis of a set of free parameters. The mixed performance index used as the cost function consists of two parts: robustness in the frequency domain; and system eigenstructure constraints in the time domain. The analytical gradient calculation of the cost function with respect to the free parameters is derived for optimal robust eigenstructure assignment using gradientbased optimization. The mixed time- and frequency-domain robust eigenstructure assignment is demonstrated in the design of a mode-decoupled roll-yaw autopilot.     

Robust control design using eigenstructure assignment and multi-objective optimization

This paper develops a new approach to multiple objective optimization design for robust multivariable control systems, based on eigenstructure assignment and genetic algorithms. It considers various performance indices (or cost functions) in the objectives, which are individual eigenvalue sensitivity functions, and the sensitivity and the complementary sensitivity functions in the frequency domain, instead of a single performance index for a control system. Based on these performance indices, the robustness criteria are expressed by a set of inequalities. The paper will make full use of the freedom provided by eigenstructure assignment to find a controller to satisfy the robustness criteria. A numerical algorithm for multi-objective optimization using genetic algorithm approaches is developed and applied to the simulation of a distillation column control system design     

Insensitive and robust control design via output feedback eigenstructure assignment

Insensitive and robust control design using output‐feedback eigenstructure assignment for linear multivariable systems is considered in this paper. A parametric expression of closed‐loop eigenvectors and generalized eigenvectors is developed. It can cope with the case where the closed‐loop eigenvalues are multiple and/or the same as the open‐loop ones so that the system to be designed can be uncontrollable and/or unobservable. The controller designed via output‐feedback eigenstructure assignment is expressed by proposed parameter vectors. The freedom provided by output‐feedback eigenstructure assignment is used to optimize some performance functions which are used to measure the sensitivity of the closed‐loop matrix and the robustness of the closed‐loop system. Copyright © 2000 John Wiley & Sons, Ltd.     

Sensitivity properties of multirate feedback control systems, basedon eigenstructure assignment

Some sampled-data systems, e.g., fly-by-wire control schemes, have a necessarily multirate structure, various input and/or outputs sampled at different rates. When considering a multirate system which has parameter uncertainty, it is important to examine ways in which the full freedom of the multivariable design can be utilized to minimize the sensitivity to parameter variations. Given the accompanying problems induced by intersample ripple disturbance. This note examines the design capabilities of a class of multirate systems with multiple input and fixed state sampling rates (MIFS), based on eigenstructure assignment. Although the use of eigenstructure assignment for continuous and single rate discrete systems is well understood, the eigenstructure assignment for the design of multirate feedback systems is an open topic of research. Accepting that the problems of intersample ripple are often magnified through multirate control, there are advantages in terms of increased freedom for minimizing sensitivity and optimizing robustness to parameter variations. A special feature of the MIFS class of multirate systems is the ability to introduce extra design freedom in the eigenproblem by a suitable choice of eigenstructure assignment and sample rates. The criteria for the selection of minimum sample rates to produce this extra freedom, and the implication that this has on the eigenstructure assignment problem, are outlined. The improved insensitivity properties are demonstrated using an example comparing the performance of multirate and corresponding single rate designs     

Robust eigenstructure assignment combining time- and frequency-domain performance specifications

This paper is concerned with robust eigenstructure assignment for multivariable systems. It combines time-domain performance specifications provided by eigenstructure assignment and robust performance specifications in the frequency domain considered by H^∞ control to realize joint optimal robust control design. A unified parametric solution for state-feedback eigenstructure assignment is derived for both the case where the sets of closed- and open-loop eigenvalues do not intersect and the case where these sets do intersect. This is based on a set of free parameters. All complex operations are converted into the real field so that the algorithm which is developed for the controller design can be easily implemented on computers. It uses a robustness index defined in the frequency domain as the cost function to be optimized. The analytical gradient calculation of the cost function with respect to the free parameters is given. Using gradient-based optimization, the robustness index is minimized by making full use of the freedom provided by eigenstructure assignment. © 1998 John Wiley & Sons, Ltd.     

矩阵二阶系统的鲁棒极点配置

直接在矩阵二阶框架下,利用特征结构配置参数化方法,研究矩阵二阶线性系统的鲁棒极点配置问题.将两种测量闭环特征值灵敏度方法有机地结合起来,给出一个新的优化性能指标,此指标的优化过程完全依赖于特征结构配置中的设计自由度.为进一步提高闭环系统的鲁棒稳定性,闭环极点也作为设计自由度的一部分参与优化.数值例子分析结果表明了该方法的有效性.     

一种稳定的分散特征结构配置方法

针对大系统分散控制中的全局稳定性问题,将分散控制问题与优化问题相结合．提出了一种新的分散控制方法。该方法利用参数特征结构配置方法将分散控制转化为最优问题．通过求解最优问题得到子系统特征结构的参数,从而得到子系统的局域控制器。该控制器不但使子系统具有鲁棒性,而且保证了大系统的全局稳定性。由于分散控制还可以进行并行计算,因此可以提高计算速度。最后对空间站的数值进行了仿真,仿真结果表明,此算法是可行的。     

Induction machine control using robust eigenstructure assignment

Control of induction machines is well known to be difficult owing to the fact that the models of induction motor are highly non-linear and time variant. In this paper, we propose a very simple approach based on linear eigenstructure assignment and eigenstructure projection to design an efficient control law. This is the first step to a more global approach using multimodel eigenstructure assignment and self-scheduling. The controller, which is proposed here, is not scheduled (with the rotor speed) but still has correct performance over the entire operating range of the induction motor. A Kalman filter is used to estimate the flux vector. The simulation is based on the non-linear model and the application is made on an experimental bench.     

航空发动机被动容错控制系统鲁棒性设计

针对航空发动机发生故障时系统的容错能力问题,设计了基于特征结构配置方法的容错控制系统。首先分析了被动容错控制的特点和优点,然后采用特征结构配置的方法,在配置系统极点的同时,还配置系统的特征向量,并对系统进行重新调节以获得整个系统在故障发生后的稳定性与可靠性,给出了特征结构配置方法的具体设计步骤。最后,针对某型航空发动机的设计工况点模拟系统发生故障,即参数发生摄动时,对所设计的容错系统鲁棒性进行了分析计算。仿真算例结果表明,所设计的容错系统具有较好的抗干扰能力,即较好的鲁棒性。