一类非线性系统的自适应预测函数控制

针对一类具有输出反馈耦合的离散非线性系统，将过程的非线性状态空间模型等效为线性时变状态空间模型；然后利用最小二乘法辨识系统参数，并通过在目标函数中引入系统状态的变化给出一种具有类似离散PI最优调节器结构的新型自适应预测函数控制器．由于引入了新的优化目标函数，该控制器控制效果与鲁棒性要优于仅考虑预测输出误差的传统预测函数控制器．仿真结果表明，该控制器优于经典离散PI最优调节器．     

一类非线性系统的扩展状态空间预测控制

针对一类具有输出反馈耦合的离散非线性系统，将过程模型表示为扩展状态空间模型，并利用反馈回路非线性函数在参考轨迹附近的局部线性表示，导出一种具有类似离散PI最优调节器结构并具有未来P步设定值前馈控制的新型预测控制器．因其目标函数中考虑了系统状态的变化，所以控制效果要优于仅考虑预测输出误差的情况．仿真结果表明了它的有效性．     

直驱小惯量表贴式永磁同步电机的转速脉动抑制

非线性电流测量误差和电压源逆变器(VSI)非线性畸变电压造成了直驱小惯量表贴式永磁同步电机(SPMSN)的转速脉动.本文将q轴非线性电流测量误差等效为扰动负载电流,提出一种复合PI(CPI)调速器抑制电机转速脉动.该调速器由传统PI调节器与偏差补偿器并联构成,偏差补偿器用以抑制非线性负载电流.同时,用分段线性函数建立IGBT关闭时间模型,并推导了VSI非线性畸变电压表达式.引入积分型模型预测控制(MPC)作为电流环控制器,利用MPC的滚动时域最优预测特性抑制VSI的非线性畸变电压,消除了零电流钳位现象.最后,通过仿真分析验证了所提控制策略的有效性.     

一种具有PI结构的时滞系统APFC控制

提出了一种具有PI结构的时滞系统自适应预测函数控制（APFC），该算法通过在性能指标函数中引入预测输出误差和预测输出增量误差，使最终形成的控制器具有比例-积分（PI）结构。结合了预测控制具有较好的处理时滞、快速跟踪和PID调节器抗干扰性强、鲁棒性强的优点，并且模型参数是通过带遗忘因子的递推最小二乘算法在线辨识得到。在具有时滞的工业过程数学模型上进行仿真研究，结果表明，该算法具有较好的跟踪效果、抗干扰能力和鲁棒性。     

带翼水下机器人的非线性拟合分段控制方法

S面控制方法可较好地解决水下机器人的运动控制问题,但由于其参数是固定的,无法达到全局最优．不同航渡速度段,采取不同的控制参数值,可保证水动力不同阶段控制输出的最优;但在速度变化的分界点,控制器输出有跳变,不利于系统的全局稳定性．利用T-S模糊系统逼近非线性连续函数的能力,采用非线性的S面函数作为模糊系统的后件,设计了基于T—S模型的S面控制器．通过T-S模型的引入,避免了控制器输出的跳变,增强了系统稳定性．将该方法应用于带翼水下机器人的深度控制,水池试验和湖中实验均证明了算法的有效性．     

反激变换器的输出反馈模型预测控制仿真研究

针对反激变换器的输出电压稳定性问题,提出一种输出反馈模型预测控制方法.建立双线性模型,设计状态反馈和输出反馈模型预测控制器,增加Luenberger-type型观测器确保状态估测动态误差是全局指数稳定;外环PI控制旨在消除电压偏移误差.给出系统存在输入电压和存在输入约束条件下的电感电流全局收敛的闭环稳定性分析.仿真结果表明,相比于PID控制,该方法在保证系统全局稳定性的同时,可以实现系统误差在有限时间内收敛,控制精度能达0.22％,提高了系统的鲁棒性和快速性.     

输出概率密度函数的最优跟踪控制:均方根B-样条模型

在分析均方根B样条模型在实现输出概率密度函数最优跟踪控制时存在的问题的基础上,提出了将最优跟踪控制转化为非线性状态约束下的跟踪误差最优调节器,然后依据非线性状态约束和系统模型的特点分别设计了鲁棒变结构控制器及非线性观测器,并利用误差补偿控制来保证非线性观测器误差的有界性.仿真结果表明了提出的转换控制策略的有效性.     

永磁同步风电系统转速环自抗扰控制器设计北大核心CSCD

转速环调节器是永磁同步风力发电系统的核心控制环节。将自抗扰控制器引入永磁同步风力发电系统转速环,对自抗扰控制器的跟踪微分器、扩张状态观测器和非线性误差反馈控制律3个组成部分进行了研究和设计,提出了一种结合抗抖振因子函数的改进型Gfal函数来提高自抗扰控制器的性能,将风力发电系统的各项扰动纳入新增状态变量进行了估计与补偿。仿真和实验结果表明,与传统PI调节器相比,转速环采用自抗扰调节器的永磁同步发电系统直流母线电压动态建立过程无超调、更接近稳态运行控制所需目标,系统发电性能得到了提高。     

基于事件触发的离散 MIMO 系统自适应评判容错控制

本文针对具有执行器故障的一类离散非线性多输入多输出(Multi-input multi-output, MIMO)系统, 提出了一种基于事件触发的自适应评判容错控制方案. 该控制方案包括评价和执行网络. 在评价网络里, 为了缓解现有的非光滑二值效用函数可能引起的执行网络跳变问题, 利用高斯函数构建了一个光滑的效用函数, 并采用评价网络近似最优性能指标函数. 在执行网络里, 通过变量替换将系统状态的将来信息转化成关于系统当前状态的函数, 并结合事件触发机制设计了最优跟踪控制器. 该控制器引入了动态补偿项, 不仅能够抑制执行器故障对系统性能的影响, 而且能够改善系统的控制性能. 稳定性分析表明所有信号最终一致有界且跟踪误差收敛于原点的有界小邻域内. 数值系统和实际系统的仿真结果验证了该方案的有效性.     

基于RBF神经网络的改进多变量预测控制

针对一类多输入多输出非线性被控对象,提出一种基于单神经网络的预测控制算法,应用RBF神经网络对非线性系统进行辨识,并计算被控系统多步预测输出值.该方法通过对传统预测目标函数加以改进,给出一种带微分项的多步预测目标函数,通过迭代寻优实时给出优化控制量.该方法实时性好,简化了传统预测控制算法,加快了滚动寻优的速度,有效地抑制了系统惯性和输入时滞所带来的超调,减小了模型误差、干扰及不确定性对控制器的影响.仿真及应用结果表明了该方法的有效性.     

Neural network based iterative learning predictive control design for mechatronic systems with isolated nonlinearity

The paper presents a new nonlinear predictive control design for a kind of nonlinear mechatronic drive systems, which leads to the improvement of regulatory capacity for both reference input tracking and load disturbance rejection. The nonlinear system is first treated into an equal linear time-variant system plus a nonlinear part using a neural network, then an iterative learning linear predictive controller is developed with a similar structure of PI optimal regulator and with setpoint feed forward control. Because the overall control law is a linear one, this design gives a direct and also effective multi-step prediction method and avoids the complicated nonlinear optimization. The control law is also an accurate one compared with traditional linearized method. Besides, changes of the system state variables are considered in the objective function with control performance superior to conventional state space predictive control designs which only consider the predicted output errors. The proposed method is compared with conventional state space predictive control method and classical PI optimal control method. Tracking performance, robustness and disturbance rejection are enlightened.     

基于模糊模型的CMP过程智能R2R预测控制方法

针对化学机械研磨(CMP)过程非线性、时变和产品质量不易在线测量的特性,提出了一种基于T-S模糊模型的CMP过程智能run-to-run(R2R)预测控制器FIPR2R;通过G-K聚类算法和最小二乘法对CMP过程的T-S模糊预测模型离线辨识,解决了复杂CMP过程难以建立精确数学模型的难题和提高了模型预测精度;通过双指数加权移动平均(dEWMA)中对过程扰动及漂移进行估计的方法实现反馈校正和基于克隆选择算法的滚动优化求取最优控制律,提高了控制精度;性能分析结果表明,FIPR2R控制器的控制性能优于dEWMA方法,有效抑制了过程扰动和漂移的影响。     

Model Predictive Control of Nonlinear Systems： Stability Region and Feasible Initial Control

This paper proposes a new method for model predictive control (MPC) of nonlinear systems to calculate stability region and feasible initial control profile/sequence, which are important to the implementations of MPC. Different from many existing methods, this paper distinguishes stability region from conservative terminal region. With global linearization, linear differential inclusion (LDI) and linear matrix inequality (LMI) techniques, a nonlinear system is transformed into a convex set of linear systems, and then the vertices of the set are used off-line to design the controller, to estimate stability region, and also to determine a feasible initial control profile/sequence. The advantages of the proposed method are demonstrated by simulation study.     

Predictive neuro-control of uncertain systems: design and use of a neuro-optimizer

We consider the problem of predictive control of uncertain stochastic discrete I/O systems. Given a model identification procedure able to give accurate output system estimates, e.g. a neural network approximation, we use another feedforward neural network to generate at each time step a constrained optimal control. Dynamic backpropagation is used to improve when necessary the controller network parameters. Both system and controller neural structures are first selected off-line by a statistical Bayesian procedure in order to make the predictive control minimizing process more efficient. The issue of stochastic stability of the closed-loop is considered. We developed this approach for the tracking control of such uncertain systems as biotechnological processes. Actual and simulated predictive neuro-control case studies in this field of application are proposed as illustrations. A comparison with a more classic quasi-Newton-based approach is also proposed, showing the interest of this neuro-control approach.     

机械手臂的多模型预测控制

针对机械手臂的非线性特点,提出了基于隶属度函数的多模型预测控制方法。该方法首先根据机械手臂的特点,选择合适的调度变量,将机械手臂的工作空间划分为若干个工作子空间,在每个子空间内的平衡点处对机械手臂进行线性化处理,得到相应的线性子模型,从而得到机械手臂的多模型表示;其次针对每个线性子模型设计局部预测控制器,使其在相应的子空间内达到控制要求;最后选择梯形隶属度函数与局部预测控制器进行加权求和,获得全局多模型预测控制器,以对机械手臂进行控制。仿真结果表明,当机械手臂的工作条件在大范围内变化时,全局多模型预测控制器的控制性能远优于常规PD控制器,达到了预期的控制目的。     

Nonlinear PI control for variable pitch wind turbine

Wind turbine uses a pitch angle controller to reduce the power captured above the rated wind speed and release the mechanical stress of the drive train. This paper investigates a nonlinear PI (N-PI) based pitch angle controller, by designing an extended-order state and perturbation observer to estimate and compensate unknown time-varying nonlinearities and disturbances. The proposed N-PI does not require the accurate model and uses only one set of PI parameters to provide a global optimal performance under wind speed changes. Simulation verification is based on a simplified two-mass wind turbine model and a detailed aero-elastic wind turbine simulator (FAST), respectively. Simulation results show that the N-PI controller can provide better dynamic performances of power regulation, load stress reduction and actuator usage, comparing with the conventional PI and gain-scheduled PI controller, and better robustness against of model uncertainties than feedback linearization control.     

A mixed PI/VI design method for nonlinear H∞ Control

This paper proposes a new mixed policy iteration and value iteration (PI/VI) design method for nonlinear H_∞ control based on the theories of polynomial optimization and Lasserre's hierarchy. The design of a mixed PI/VI controller can be carried out in four steps: firstly, initialize design parameters and expand nonlinear system matrices; secondly, obtain a polynomial matrix inequality for policy improvement; thirdly, obtain the Lasserre's hierarchy of a global polynomial optimization problem for value improvement; fourthly, perform the mixed PI/VI algorithm to approximate the optimal nonlinear H_∞ control law. The novelty of this work lies in that the problem of designing a nonlinear H_∞ controller is translated into a polynomial global optimization problem, which can be solved by Lasserre's hierarchy directly, and then, the mixed PI/VI algorithm is presented to approximate the optimal nonlinear H_∞ control law by updating global optimizers iteratively. The main results of this paper consist of the mixed PI/VI algorithm and the related three theorems, which guarantee robust stability and performance of the closed‐loop nonlinear system. Numerical simulations show that the mixed PI/VI algorithm converges very fast and achieves good robust stability and performance in transient behavior, disturbance rejection, and enlarging the domain of attraction of the close‐loop system.     

基于神经网络模型的空燃比非线性模型预测控制

采用基于径向基神经网络（RBFNN）模型的非线性模型预测控制方法,被控对象选择火花塞点火（SI）发动机的空燃比（AFR）高度非线性复杂系统,利用渐消记忆最小二乘法实现基于RBFNN的SI发动机AFR系统建模以及参数在线自适应更新。针对非线性模型预测控制中寻优问题,运用序列二次规划滤子算法对最优控制序列进行求解,并加入滤子技术避免了罚函数的使用。在相同的实验环境下,与PI控制算法和Volterra模型预测控制方法进行仿真对比实验,结果表明,所提算法的控制效果明显优于其他两种方法。