基于模型跟随的神经网络非线性重构控制

针对歼击机在结构故障下的动力学方程 ,提出了一种基于径向基函数 (RBF)神经网络的模型跟随非线性重构控制策略。该方法不必精确已知系统故障的位置及其损伤程度 ,可直接对故障系统实施重构控制 ,使其输出能精确跟踪期望参考模型的输出。该方法在模型跟随重构控制的基础上 ,引入了神经网络控制器 ,以补偿故障引起的非线性因素的影响。理论分析和仿真验证表明 ,所提方法可保证闭环系统具有良好的重构性能和很强的鲁棒性 ,且算法高效简单 ,易于计算机在线控制。     

基于Wiener模型的风力发电系统变桨距控制

风力发电系统具有非线性强和波动性大的特点，如何在风速随机变化的工况下，快速地将风电机组输出功率控制在额定范围内是风电系统控制面临的重大难题。针对5 MW风电机组的变桨距控制问题，提出一种基于Wiener模型的风力发电系统变桨距控制方案。根据Wiener模型线性和非线性环节可分离的特殊结构，通过计算非线性环节的逆，构造具有线性性质的广义风电系统被控对象，将风电系统非线性变桨距控制问题简化为线性模型控制问题。采用独立可分离信号分步辨识风电机组Wiener模型，可以更加精准地预测风电机组输出功率。仿真实例表明：该控制方法有效地提高了变桨距系统的控制效果。     

非仿射非线性系统的自适应模糊输出反馈控制

针对一类单输入单输出非仿射非线性系统,提出了基于观测器的自适应模糊输出反馈控制。该算法通过构造线性误差观测器估计系统输出误差和系统状态,利用估计状态变量和估计输出误差变量设计自适应控制律,采用模糊逻辑系统在线消除非线性系统中的不确定模型,引入监督控制消除系统逼近误差和外界干扰,并利用Lyapunov方法严格证明了在该控制律作用下闭环系统所有信号一致最终有解,闭环系统的输出能最终跟踪期望输出。仿真结果表明了该方法的有效性。     

磨矿分级过程多模型智能控制研究

提出了一种非线性多模型控制算法,在若干平衡点附近首先建立多个子模型及相应的控制器,再通过在线计算模型匹配度来适应模型参数的变化,被控对象的最终输入是各控制器输出的加权和.将其应用于磨矿分级非线性系统的控制研究,仿真结果表明,所提算法极大地改善磨矿分级系统的动态响应,并增强了系统的鲁棒性.     

基于一种新型动态神经网络的非线性自适应逆控制

根据生物神经元的机能,提出了一种具有动态激励函数的新型神经元模型,由此构成的神经网络(DAFNN)应用在非线性自适应逆控制中时只需要确定隐层神经元个数,从而克服了用NARX回归神经网络时需确定输入和输出延时阶数及隐层神经元个数等多个参数的不足。通过对单输入单输出(SISO)及多输入多输出(MIMO)非线性系统的自适应逆控制仿真研究,证实了DAFNN是一种很好的非线性系统建模和控制工具。     

基于T-S模糊模型的模型参考自适应逆控制

针对非线性系统,提出一种基于T-S模糊模型的模型参考自适应逆扰动消除控制方法。所提方法根据模糊辨识理论与模型参考自适应逆控制各自的特点,将两者相结合。首先,根据模糊系统理论,分别采用模糊对角线划分和递推最小二乘算法进行前提和结论参数辨识,离线辨识得到对象模糊模型和逆对象模糊模型。将辨识出的对象逆设为原始控制器,与被控对象串联;为了分离出系统扰动信号,将辨识出的对象模型与被控对象并联,通过被控系统与对象模型输出做比较,再通过逆对象模型反馈到系统输入端,组成扰动消除环节。用最小均方差算法在系统运行过程中在线调节逆对象模糊模型参数,使其输出误差最小。最后,使用所提方法对一混合非线性系统及输入/输出非线性系统进行仿真试验,仿真结果验证了所提方法的有效性。     

基于Elman网络的鲁棒型广义预测控制方法研究

提出一种基于Elman动态回归神经网络模型的鲁棒型广义预测控制(GPC)．该算法首先用EIman网络对非线性系统进行辨识，建立预测模型，然后在控制中将模型输出值与测量输出值进行综合，代替量测输出用于控制中，从而降低辨识器与控制器对未建模动态的敏感性，加强控制器的适应能力和鲁棒性．仿真结果证明：将本算法应用于非线性系统预测控制，对未建模动态具有很强的鲁棒性和很好的控制能力。     

严格反馈型随机非线性系统输出反馈滑模控制

研究了一类带有不确定性参数的严格反馈型随机非线性系统的输出反馈镇定问题,主要是采用变结构滑模控制方法(SMC)来设计镇定控制器。首先根据系统的结构和输出状态构造了随机非线性状态观测器,并针对结构参数的不确定性,采用梯度自适应调节律,然后基于系统的观测状态选择滑动模切换流形,给出了不含噪声激励的变结构控制律,并证明滑动模的可达性和闭环系统滑模运动的稳定性,最后讨论了滑模控制的抖振抑制问题。通过系统仿真,与其它鲁棒控制方法进行比较,证明了本方法不仅能够有效地镇定该类随机非线性系统,并且具有一些更好的控制性能。     

高精度陀螺稳定跟踪系统神经网络预测控制

针对高精度陀螺稳定跟踪系统，提出了一种基于小波基函数神经网络的非线性系统的一步超前预测控制算法。该方法利用小波网络学习非线性系统，利用小波神经网络模型作为系统的预测模型，控制信号直接通过极小化期望输出值与预测输出值之间的偏差来获得。通过对陀螺稳定跟踪控制系统的仿真，表明该算法具有优良的控制品质。     

电站中速磨的模糊跟踪控制及仿真

针对电站中速磨的控制难题,提出了一种模糊状态反馈跟踪控制方案。利用内模原理的特性来改善模糊跟踪控制系统的性能,使得被控非线性对象的输出变量可以渐进跟踪期望的变化轨迹。采用T-S建模方法来对中速磨进行建模并得到其T-S模糊模型。针对中速磨的T-S模糊模型设计模糊状态反馈跟踪控制系统并进行仿真,结果表明系统的输出可以很好地跟踪给定的轨迹。     

QFT control based on zero phase error compensation for flight simulator

To improve the robustness of high-precision servo systems, quantitative feedback theory (QFT) which aims to achieve a desired robust design over a specified region of plant uncertainty is proposed. The robust design problem can be solved using QFT but it fails to guarantee a high precision tracking. This problem is solved by a robust digital QFT control scheme based on zero phase error (ZPE) feed forward compensation. This scheme consists of two parts: a QFT controller in the closed-loop system and a ZPE feed-forward compensator. Digital QFT controller is designed to overcome the uncertainties in the system. Digital ZPE feed forward controller is used to improve the tracking precision. Simulation and real-time examples for flight simulator servo system indicate that this control scheme can guarantee both high robust performance and high position tracking precision.     

The further result on global practical tracking for high-order uncertain nonlinear systems

This paper extends the unknown control coefficients with lower and upper constant bounds to the ones which may take arbitrarily large and /or small values.Since the existing methods are no longer applicable and the technical obstacles caused by the extensions are essential,new control design scheme should be exploited to the global practical tracking.By the approaches of Nussbaum-gain and adding a power integrator,the authors successfully propose the design scheme of the adaptive practical tracking controller for the systems.It is shown that the designed controller guarantees that all the closed-loop system states are bounded and the tracking error becomes prescribed arbitrarily small after a finite time.     

FINITE TIME TRACKING CONTROL FOR RIGID ROBOTIC MANIPULATORS WITH FRICTION AND EXTERNAL DISTURBANCES

1.Introduction Variable structure control is well known for its robustness to system uncertainties and external disturbances and has been successfully applied to the rigid robotic manipulator systems(Utkin(1992),Zinober(1990)).Generally,a linear sliding mode is firstly designed to describe the desired system error dynamics,a robust controller drives the switching plane variables to reach the sliding mode,and the error dynamics will asymptotically converge to zero on the linear sliding mode.Si…     

基于反馈线性化的无人驾驶车辆横向位置跟踪控制

由于无人驾驶车辆横向运动学模型常被看作高非线性、时变的,所以这给控制方案的制定带来了较大的难度。提出了基于输入输出反馈线性化的横向位置跟踪控制方案。采用这种控制方案在车体正向和倒向开车的情况下,都得到较好的横向跟踪效果,动态特性好、稳态精度高。基于由Simulink和VRML语言建立起的三维动画仿真也说明了这一点。     

Output feedback adaptive control of multivariable nonlinear systems using Nussbaum gain method

1. INTRODUCTION Output feedback adaptive control of single input and single output (SISO) nonlinear systems with unknown constant parameters has been well developed via backstepping design[1]. In order to achieve asymptotic tracking of a known reference trajectory, a restrictive assumption is the sign of the high frequency gain to be known. Some results for relaxing this assumption have also been proposed [2]. It has been shown that if the sign of the high frequency gain is unknown, the …     

基于鲁棒故障诊断LPV模型的舰载QUAV跟踪控制

设计适用于外部干扰抑制的无人机鲁棒故障检测与跟踪控制系统。首先,将舰载四旋翼直升机(quad-roto unmanned aerial vehicle, QUAV)建模为线性变参数系统(linear variable parameter system, LPVs),作为目标来进行系统设计。同时,为了进行故障检测和隔离,设计了一个强大的LPVs观测器,利用一组可行线性矩阵不等式(linear matrix inequalities, LMI)给出保证系统渐近稳定和抗干扰鲁棒性的充分条件。此外,基于LMI区域的极点配置考虑观测器的增益设计。然后,通过观测器库考虑故障检测和隔离方案,检测和隔离传感器故障。其次,通过考虑比较器集成控制方案对反馈控制器进行设计。目标是设计鲁棒控制器,使QUAV无偏差跟踪设定位置。最后,通过在模拟模型中,对算法的有效性进行实验验证。     

Output feedback based adaptive robust fault-tolerant control for a class of uncertain nonlinear systems

An adaptive robust approach for actuator fault-tolerant control of a class of uncertain nonlinear systems is proposed. The two chief ways in which the system performance can degrade following an actuator-fault are undesirable transients and unacceptably large steady-state tracking errors. Adaptive control based schemes can achieve good final tracking accuracy in spite of change in system parameters following an actuator fault, and robust control based designs can achieve guaranteed transient response. However, neither adaptive control nor robust control based fault-tolerant designs can address both the issues associated with actuator faults. In the present work, an adaptive robust fault-tolerant control scheme is claimed to solve both the problems, as it seamlessly integrates adaptive and robust control design techniques. Comparative simulation studies are performed using a nonlinear hypersonic aircraft model to show the effectiveness of the proposed scheme over a robust adaptive control based faulttolerant scheme.     

一类非仿射受扰混沌系统的自适应模糊控制

针对一类带非仿射输入的不确定受扰混沌系统,提出了带监督控制项的自适应模糊控制。该方法采用模糊逻辑系统逼近虚拟控制中的未知函数,仅要求逼近误差范数有界。模糊参数采用σ自适应律,并给出参数的有界性证明。构造Lyapunov函数证明闭环系统所有信号一致有界,跟踪误差一致渐进稳定。将Duffing-Holmes系统、Genesio系统和Sprott电路混沌系统作为仿真对象,仿真结果验证了算法的有效性。     

Robust adaptive neural network control with supervisory controller

The problem of direct adaptive neural network control for a class of uncertain nonlinear systems with unknown constant control gain is studied in this paper. Based on the supervisory control strategy and the approximation capability of multilayer neural networks (MNNs), a novel design scheme of direct adaptive neural network controller is proposed.The adaptive law of the adjustable parameter vector and the matrix of weights in the neural networks and the gain of sliding mode control term to adaptively compensate for the residual and the approximation error of MNNs is determined by using a Lyapunov method. The approach does not require the optimal approximation error to be square-integrable or the supremum of the optimal approximation error to be known. By theoretical analysis, the closed-loop control system is proven to be globally stable in the sense that all signals involved are bounded, with tracking error converging to zero.Simulation results demonstrate the effectiveness of the approach.     

基于神经网络的非线性系统的间接自适应控制

针对一类不确定非线性动态系统 ,利用多层神经网络系统MNNs的逼近能力 ,提出了一种间接鲁棒自适应神经网络控制器的设计方案。该方案不仅能够保证闭环系统的所有信号有界 ,而且理论分析证明了闭环系统的跟踪误差渐近收敛到零。仿真试验表明本控制算法是有效的。