基于扩张状态观测器的车辆横摆稳定模型预测控制器设计（英文）

针对车辆横摆稳定性控制问题,本文提出一种基于扩张状态观测器的线性模型预测控制器设计方法.首先,将非线性车辆模型线性化,建立带有模型误差干扰项的线性模型,其中线性化导致的模型误差采用扩张状态观测器估计得到,并证明了观测器的稳定性.然后基于此模型设计线性预测控制器,近似实现了非线性预测控制器的控制效果,同时降低了计算量.最后,通过不同路况下的仿真实验结果,验证了所提方法的计算性能和控制效果.     

基于干扰抑制控制的飞行器姿态跟踪

本文针对飞行器姿态跟踪控制问题, 考虑系统的内部模型不确定性和外界扰动, 设计了使跟踪误差一致最终有界的控制器. 以四元数为姿态参数, 建立系统的非线性误差模型; 将控制系统分为内环观测器和外环控制器分别设计, 其中, 线性扩张状态观测器作为系统内环将实际系统补偿为标称模型, 而外环非线性控制器则用于镇定非线性标称系统. 最后, 利用Lyapunov理论得到了一致最终有界的稳定性结论, 并基于频域理论, 分析了线性扩张状态观测器阶次对系统性能的影响. 姿态跟踪实验表明, 本文设计的控制系统能够有效实现飞行器的姿态跟踪控制.     

通信延时环境下基于观测器的智能网联车辆队列分层协同纵向控制

考虑通信延时影响的车辆队列控制问题,提出一种基于观测器的分布式车辆队列纵向控制器.首先,基于分层控制策略分别设计上下层控制器,通过上层控制器优化期望加速度、下层控制器克服车辆模型非线性实现期望加速度和实际加速度的一致.上层控制器设计过程中,基于三阶线性化车辆模型,考虑观测器、车辆动态耦合特性和通信延时,提出一种通信延时环境下基于观测器的车辆队列控制器,利用观测器估计领导车辆加速度信息从而减轻通信负担.然后,利用Lyapunov-Krasovskii方法分析车辆队列的稳定性,并得出通信延时上界,同时利用传递函数方法分析了串稳定性.最后,通过数值仿真验证上层控制器的有效性和稳定性.在此基础上,利用PreScan软件中高保真车辆动态模型,验证了该分层控制策略的有效性.     

并网逆变器逆系统自学习滑模抗扰控制

针对三相并网逆变器模型的多变量、非线性、强耦合等特点,采用开关函数法建立其开关周期平均模型,在此模型的基础上采用逆系统方法实现反馈线性化和解耦控制,对伪线性系统设计自适应滑模抗扰控制器,使用非线性光滑函数设计扩张状态观测器以实现内部建模误差与外部扰动的扩张状态估计,并将非线性扩张状态观测器和跟踪微分器与自学习滑模控制器结合使用.仿真结果表明,该方法具有响应速度快、控制精度高、抗扰能力强的特性,在并网逆变器中具有较大应用价值.     

基于预测控制的高超声速飞行器容错控制器设计

针对存在升降舵面偏转角卡死故障的高超声速飞行器,提出一种基于预测控制的容错控制器设计方法.利用输入输出反馈线性化,对高超声速飞行器纵向模型进行变换;对于速度和高度的高阶导数以及故障项,设计扩张状态观测器在线观测;采用泰勒展开得到预测模型,建立连续预测控制器,分析证明闭环控制系统的稳定性和观测误差的有界性.仿真结果验证了所提方法的有效性.     

基于迭代扩张状态观测器的数据驱动最优迭代学习控制

针对一类带扰动有限时间内重复运行的离散时间非线性非仿射不确定系统,本文提出了一种基于迭代扩张状态观测器的数据驱动最优迭代学习控制方法.首先,提出了改进的迭代动态线性化方法,将被控系统线性化为与控制输入有关的仿射形式,并将不确定性合并到一个非线性项中;然后,设计了迭代扩张状态观测器对非线性不确定项进行估计,作为对扰动的补偿;最后,设计了性能指标函数,通过最优技术,提出了参数迭代更新律和最优学习控制律.本文通过数学分析,证明了跟踪误差的有界收敛性.仿真结果验证了方法的有效性.所提出的新型迭代动态线性化方法可很大程度上降低线性化后的控制增益的动态复杂性,使其易于估计.所提出的迭代扩张状态观测器可以在重复中学习,对非重复扰动可进行有效的估计.此外,本文控制器的设计与分析是数据驱动的控制方法,除了被控系统的输入输出数据以外,不需要任何其他模型信息.     

基于LESO的MMC-RPC反馈线性化直接功率控制

模块化多电平铁路功率调节器作为一个耦合的多变量非线性系统, 传统PI控制的直接功率控制难以实现 对系统的精确解耦. 本文提出了一种基于线性扩张状态观测器的反馈线性化直接功率控制方法, 根据Lie导数构建 了模块化多电平铁路功率调节器(MMC-RPC)两输入/两输出功率仿射模型, 设计了精确反馈线性化功率解耦控制 器. 针对不确定因素等扰动对精确反馈线性化控制效果的影响, 设计了线性扩张状态观测器对扰动进行观测和补 偿, 实现了功率的精确跟踪控制. 最后, 通过MATLAB/Simulink平台搭建仿真模型对所提控制方法进行了验证.     

含有摩擦补偿的全方位移动机器人自抗扰控制

本文针对全方位移动机器人轨迹追踪中的摩擦补偿问题,提出了一种改进的非线性自抗扰控制器.首先建立了含有经典静态摩擦模型的全方位移动机器人动力学模型.其次,基于该模型设计非线性控制器和线性扩张状态观测器并给出了系统的稳定性分析.通过将模型已知项加入线性扩张状态观测器中得到摩擦力的估计值,并将估计值用于非线性控制器中摩擦补偿部分.为减小摩擦力对机器人低速运动轨迹追踪控制的影响,非线性控制器采用变增益控制器进行轨迹追踪控制.最后通过仿真结果验证本文提出控制器的有效性.     

一类线性时变系统的输出反馈控制

基于线性时不变系统能控能观标准型变换及非线性系统高增益观测器方法,本文研究了一类线性时变系统 的输出反馈控制问题. 通过引入时变的状态变量坐标变换,分别设计了线性时变系统的状态反馈控制器、状态观测器以及基于 状态观测器的输出反馈控制器. 进一步地,本文分别证明了观测器动态误差是渐近收敛于零的,而状态反馈控制器以及输出反馈控制器可以 保证闭环系统的渐近稳定性.     

智能车辆路径跟踪控制方法

为了提高智能车辆路径跟踪控制器的可靠性和控制精度,提出一种基于误差动力学模型的路径跟踪控制方法.基于车辆运动学模型和动力学模型建立系统误差动力学模型,并在此基础上推导出车辆路径跟踪控制的稳态控制律,利用李雅普诺夫稳定性理论验证稳态控制律的正确性.为了减小外部干扰对控制性能的影响,提高控制器的可靠性,进一步设计基于车辆侧向位移误差的瞬态控制律,并利用李雅普诺夫稳定性理论验证闭环系统的稳定性.稳态控制律和瞬态控制律构成了非线性的路径跟踪控制器.通过与车辆路径跟踪常用的线性控制器和非线性控制器对比验证所提出控制方法的有效性,线性控制器选用LQR控制器,非线性控制器选用Stanley控制器.仿真结果表明,与LQR控制器相比,所提出控制方法的路径跟踪控制精度、抗干扰性和可靠性更好.与Stanley控制器相比,所提出控制方法具有更好的路径跟踪控制精度和控制收敛速度,且在大曲率路径跟踪过程中具有更好的可靠性.     

基于扩张状态观测器的分散模型预测控制

针对复杂关联系统中分散控制方法无法有效解决子系统间的耦合和干扰问题, 提出一种基于扩张状态观测器的分散模型预测控制算法. 首先将复杂关联系统分解为多个状态维数较低、控制变量较少的子系统, 并为每个子系统设计本地预测控制器; 然后, 采用扩张状态观测器对子系统的耦合项以及干扰项进行估计, 进而利用估计值对子系统进行前馈补偿, 从而降低复杂关联系统的计算复杂度, 提高系统的稳定性和抗干扰能力; 最后, 利用液位控制系统验证了所提出算法的有效性.

     

A hybrid model predictive control strategy for nonlinear plant-wide control

A plant-wide control strategy based on integrating linear model predictive control (LMPC) and nonlinear model predictive control (NMPC) is proposed. The hybrid method is applicable to plants that can be decomposed into approximately linear subsystems and highly nonlinear subsystems that interact via mass and energy flows. LMPC is applied to the linear subsystems and NMPC is applied to the nonlinear subsystems. A simple controller coordination strategy that counteracts interaction effects is proposed for the case of one linear subsystem and one nonlinear subsystem. A reactor/separator process with recycle is used to compare the hybrid method to conventional LMPC and NMPC techniques.     

Algorithms for improved fixed-time performance of Lyapunov-based economic model predictive control of nonlinear systems

This work presents algorithms for improved fixed-time performance of Lyapunov-based economic model predictive control (LEMPC) of nonlinear systems. Unlike conventional Lyapunov-based model predictive control (LMPC) schemes which typically utilize a quadratic cost function and regulate a process at a steady-state, LEMPC designs very often dictate time-varying operation to optimize an economic (typically non-quadratic) cost function. The LEMPC algorithms proposed here utilize a shrinking prediction horizon with respect to fixed (but potentially large) operation period to ensure improved performance, measured by the desired economic cost, over conventional LMPC by solving auxiliary LMPC problems and incorporating appropriate constraints, based on the LMPC solution, in their formulations at various sampling times. The proposed LEMPC schemes also take advantage of a predefined Lyapunov-based explicit feedback law to characterize their stability region while maintaining the closed-loop system state in an invariant set subject to bounded process disturbances. The LEMPC algorithms are demonstrated through a nonlinear chemical process example.     

Identification and predictive control of a multistage evaporator

A recurrent neural network-based nonlinear model predictive control (NMPC) scheme in parallel with PI control loops is developed for a simulation model of an industrial-scale five-stage evaporator. Input–output data from system identification experiments are used in training the network using the Levenberg–Marquardt algorithm with automatic differentiation. The same optimization algorithm is used in predictive control of the plant. The scheme is tested with set-point tracking and disturbance rejection problems on the plant while control performance is compared with that of PI controllers, a simplified mechanistic model-based NMPC developed in previous work and a linear model predictive controller (LMPC). Results show significant improvements in control performance by the new parallel NMPC–PI control scheme.     

Lyapunov-based model predictive control of stochastic nonlinear systems

In this work, we design a Lyapunov-based model predictive controller (LMPC) for nonlinear systems subject to stochastic uncertainty. The LMPC design provides an explicitly characterized region from where stability can be probabilistically obtained. The key idea is to use stochastic Lyapunov-based feedback controllers, with well characterized stabilization in probability, to design constraints in the LMPC that allow the inheritance of the stability properties by the LMPC. The application of the proposed LMPC method is illustrated using a nonlinear chemical process system example.     

Unscented Kalman filter based nonlinear model predictive control of a LDPE autoclave reactor

This paper presents a multivariable nonlinear model predictive control (NMPC) scheme for the regulation of a low-density polyethylene (LDPE) autoclave reactor. A detailed mechanistic process model developed previously was used to describe the dynamics of the LDPE reactor and the properties of the polymer product. Closed-loop simulations are used to demonstrate the disturbance rejection and tracking performance of the NMPC algorithm for control of reactor temperature and weight-averaged molecular weight (WAMW). In addition, the effect of parametric uncertainty in the kinetic rate constants of the LDPE reactor model on closed-loop performance is discussed. The unscented Kalman filtering (UKF) algorithm is employed to estimate plant states and disturbances. All control simulations were performed under conditions of noisy process measurements and structural plant–model mismatch. Where appropriate, the performance of the NMPC algorithm is contrasted with that of linear model predictive control (LMPC). It is shown that for this application the closed-loop performance of the UKF based NMPC scheme is very good and is superior to that of the linear predictive controller.     

An integrated approach to hypersonic entry attitude control

This paper presents an integrated approach based on dynamic inversion(DI)and active disturbance rejection control(ADRC)to the entry attitude control of a generic hypersonic vehicle(GHV).DI is frstly used to cancel the nonlinearities of the GHV entry model to construct a basic attitude controller.To enhance the control performance and system robustness to inevitable disturbances,ADRC techniques,including the arranged transient process(ATP),nonlinear feedback(NF),and most importantly the extended state observer(ESO),are integrated with the basic DI controller.As one primary task,the stability and estimation error of the second-order nonlinear ESO are analyzed from a brand new perspective:the nonlinear ESO is treated as a specifc form of forced Li′enard system.Abundant qualitative properties of the Li′enard system are utilized to yield comprehensive theorems on nonlinear ESO solution behaviors,such as the boundedness,convergence,and existence of periodic solutions.Phase portraits of ESO estimation error dynamics are given to validate our analysis.At last,three groups of simulations,including comparative simulations with modeling errors,Monte Carlo runs with parametric uncertainties,and a six degrees-of-freedom reference entry trajectory tracking are executed,which demonstrate the superiority of the proposed integrated controller over the basic DI controller.     

An approach to improve active disturbance rejection control

ABSTRACT

In order to reduce the error and phase delay of the classical extended state observer (ESO) in estimating the system state and disturbance, in this paper, we combine ESO and tracking differentiator (TD) to construct a tracking differential extended state observer (TDESO). The observation error and observation speed of TDESO are also discussed. Then a nonlinear active disturbance rejection control system improved by TDESO for a linear plant is transformed into a Lurie system. Moreover, the circular criterion is used to analyse the absolute stability of the transformed Lurie system. Finally, TDESO is optimised and an improved linear state error feedback (PLSEF) is proposed to improve the rapidity of the system by using simulation and time domain analysis. And a second-order system is used to illustrate the performance of the proposed scheme. The simulation results show that our algorithm is effective.     

Control of the common rail pressure in gasoline engines through an extended state observer based MPC

In this paper, a model predictive control (MPC) solution, assisted by extended state observer (ESO), is proposed for the common rail pressure control in gasoline engines. The rail pressure dynamic, nonlinear with large uncertainty, is modeled as a simple first order system. The discrepancy of the model from the real plant is lumped as ``total disturbance'', to be estimated in real-time by ESO and then mitigated in the nonlinear MPC, assuming the total disturbance does not change in the prediction horizon. The nonlinear MPC problem is solved using the Newton/generalized minimum residual (GMRES) algorithm. The proposed ESO-MPC solution, is compared with the conventional proportional-integral-differential (PID) controller, based on the high-fidelity model provided in the benchmark problem in IFAC-E-CoSM. Results show the following benefits from using ESO-MPC relative to PID (benchmark): 1) the disturbance rejection capability to fuel inject pulse step is improved by 12% in terms of recovery time; 2) the transient response of rail pressure is improved by 5% in terms of the integrated absolute tracking error; and 3) the robustness is improved without need for gain scheduling, which is required in PID. Additionally, increasing the bandwidth of ESO allows reducing the complexity of the model implemented in MPC, while maintaining the disturbance rejection performance at the cost of high noise-sensitivity. Therefore, the ESO-MPC combination offers a simpler and more practical solution for common rail pressure control, relative to the standard MPC, which is consistent with the findings in simulation.     

未知死区输入非线性系统的双观测器动态面控制

针对一类含死区输入的严格反馈非线性系统,提出基于双观测器的自适应鲁棒控制算法.动态面的每一步设计中,第1观测器即跟踪信号观测器对指令信号进行观测,并得到指令信号的差分信号,消除传统动态面控制中计算复杂问题.第二观测器即扰动观测器在线估计高阶动态面控制系统中每一步的不确定模型,与跟踪信号观测器实现双反馈控制,提高控制效果.通过李雅普诺夫方法分析了自适应鲁棒控制器的输入输出有界的特性.最后,实验验证基于双观测器的自适应鲁棒控制器的性能,结果表明本文方法控制效果较好.且采用反双曲正弦函数建立观测器,参数调节少,利于工业应用.