基于神经网络的PMSM自适应滑模控制

结合滑模控制和神经网络各自的优点,对永磁同步电机(PMSM)提出了一种基于神经网络的PMSM自适应滑模控制方案.首先设计了带积分操作的滑模变结构位置控制器,通过递归神经网络的在线学习来实时估计系统参数变化和外部负载扰动等不确定性的界限,减小滑模控制器的控制量.进而,在滑模控制器中又引入饱和函数取代符号函数,进一步减弱"抖振"现象.理论分析和实验仿真对比研究的结果表明所提出方法具有优越的动态性能和鲁棒性.     

多关节机器人的模糊神经滑模控制

针对含有建模误差和不确定干扰的多关节机器人轨迹跟踪控制,提出了一种模糊神经滑模控制方法.该方法采用全局快速终端滑模面,保证了系统能够从任意初始状态在有限时间内到达滑模面和平衡点.采用模糊神经网络自适应地补偿系统的建模误差和外界干扰,保证了滑模控制在滑模面的运动.文中利用李亚普诺夫稳定性判据推导出了控制器的控制律和模糊神经网络的目标函数,通过模糊神经网络的在线学习.削弱了滑模控制的抖振.仿真结果表明了其有效性.     

基于径向基函数神经网络的机器人滑模控制

尽管滑模控制响应快,对系统参数和外部扰动呈不变性,但在保证系统的渐进稳定性上却存在很强的抖动缺点.因此,在一般滑模控制的基础上,引入了径向基函数神经网络(RBFNN).利用滑模控制的特点设定目标函数,将切换函数作为RBFNN的输入,滑模控制量作为其输出.利用RBF神经网络的在线学习功能,消除了控制的抖动,同时使系统具有很强的鲁棒性.对两连杆机械手进行了仿真研究,其结果表明,在存在模型误差和外部扰动的情况下,该方案既能达到高精度快速跟踪的目的,又能消除滑模控制的抖动问题.     

基于RBF神经网络的多关节机器人固定时间滑模控制

针对具有典型非线性特性的多关节机器人轨迹跟踪控制问题,提出一种基于径向基函数(RBF)神经网络的固定时间滑模控制方法.首先,基于凯恩方法建立包括系统模型不确定性以及外部干扰在内的多关节机器人动力学模型;然后,根据机器人动力学模型设计一种固定时间收敛的滑模控制器, RBF神经网络用来逼近系统模型中的不确定性项,并利用Lyapunov理论证明该系统跟踪误差能在固定时间内收敛;最后,对特定型号的多关节机器人虚拟样机进行仿真分析,结果表明:与基于RBF神经网络的有限时间滑模控制器相比,所提出控制器具有良好的跟踪性能且能保证系统状态在固定时间内收敛.     

多关节机器人的神经滑模控制

针对含有建模误差和不确定干扰的多关节机械臂轨迹跟踪控制,提出了一种神经滑模控制方法。该方法采用全局快速终端滑模面保证了系统状态能够在有限时间内到达滑模面和平衡点。采用径向基函数神经网络自适应地补偿系统的建模误差和外界干扰,保证了滑模控制在滑模面的运动。利用李亚普诺夫稳定性判据推导出了控制器的控制律和神经网络的目标函数,通过神经网络的在线学习,削弱了滑模控制的抖振。仿真结果表明了其有效性。     

基于函数滑模控制器的机械手轨迹跟踪控制

提出一种基于函数滑模控制器(FSMC)的控制策略,用于不确定机械手的轨迹跟踪控制。首先,由动力学模型和滑模函数得到系统的不确定项;然后,利用RBF神经网络逼近系统不确定项,由于神经网络逼近存在误差,而且在初始阶段误差较大,设计函数滑模控制器和鲁棒补偿项对神经网络逼近误差进行补偿,以克服普通滑模控制器容易引起的抖振问题,同时提高系统的跟踪控制性能。基于李亚普诺夫理论证明了闭环系统的全局稳定性,仿真实验也验证了方法的有效性。     

永磁同步电机的自适应反演滑模变结构控制

针对永磁同步电机提出一种基于反演的PMSM自适应滑模控制方案.设计基于反演的滑模变结构位置控制器,通过RBF神经网络实现系统参数变化和外部负载扰动等引起的不确定上界值的在线辨识,减小滑模控制器的控制量,并引入饱和函数来减弱系统的"抖动"现象.理论分析和仿真结果对比表明,基于RBF神经网络的自适应反演滑模控制对参数变化和外部负载扰动具有很好的鲁棒性,永磁同步电动机获得了很好的跟踪效果.     

基于函数滑模控制器的机械手轨迹跟踪控制

提出一种基于函数滑模控制器(FSMC)的控制策略,用于不确定机械手的轨迹跟踪控制。首先,由动力学模型和滑模函数得到系统的不确定项;然后,利用RBF神经网络逼近系统不确定项,由于神经网络逼近存在误差,而且在初始阶段误差较大,设计函数滑模控制器和鲁棒补偿项对神经网络逼近误差进行补偿,以克服普通滑模控制器容易引起的抖振问题,同时提高系统的跟踪控制性能。基于李亚普诺夫理论证明了闭环系统的全局稳定性,仿真实验也验证了方法的有效性。     

基于自适应模糊滑模控制的船舶航向控制器设计

针对船舶运动系统中固有的非线性、模型不确定性和风、浪、流等的干扰.提出了自适应模糊滑模控制(AFSMC)策略解决船舶的航向控制问题.通过采用模糊逻辑系统逼近系统未知函数,将滑模控制技术与自适应模糊控制技术相结合,设计了船舶航向AFSMC控制器.在滑模边界层内应用PI (proportional-integral)控制代替滑模控制中的切换项,削弱了滑模控制带来的抖振现象.借助李亚普诺夫函数证明了船舶运动系统中的信号都一致有界并利用Barbalat引理证明了跟踪误差渐近收敛到零.在参数摄动和外界干扰情况下进行了航向保持与改变仿真试验,采用AFSMC控制器得到了与无摄动和无干扰情况下相似的输出响应.实验结果表明,所提控制器能有效地处理系统不确定性和外界干扰,控制性能良好,具有很强的鲁棒性.     

船舶航向控制的多滑模鲁棒自适应设计

针对带有未知虚拟控制增益和常参数不确定的非匹配不确定船舶航向非线性控制问题,设计了一种新的多滑模鲁棒自适应控制算法.该算法利用神经网络来逼近系统模型的不确定性;应用逐步递推的多滑模控制算法降低了控制器的复杂性;尤其是采用Nussbaum函数处理系统中符号未知的问题,避免了可能存在的控制器奇异值问题;然后借助Lyapunov稳定性分析方法,理论分析证明了所得闭环系统全局一致最终有界,且跟踪误差收敛到零.仿真试验结果表明,该方法具有较好的控制效果.     

A board system for high-speed image analysis and neural networks

Two ANNA neural-network chips are integrated on a 6U VME board, to serve as a high-speed platform for a wide variety of algorithms used in neural-network applications as well as in image analysis. The system can implement neural networks of variable sizes and architectures, but can also be used for filtering and feature extraction tasks that are based on convolutions. The board contains a controller implemented with field programmable gate arrays (FPGA's), memory, and bus interfaces, all designed to support the high compute power of the ANNA chips. This new system is designed for maximum speed and is roughly 10 times faster than a previous board. The system has been tested for such tasks as text location, character recognition, and noise removal as well as for emulating cellular neural networks (CNN's). A sustained speed of up to two billion connections per second (GC/s) and a recognition speed of 1000 characters per second has been measured.     

网络诱导延时对基于网络学习的神经网络PID控制系统的影响及分析

首先介绍了网络学习控制系统,并用 MATLAB 对以神经网络作为网络学习控制器的控制系统进行实例仿真。仿真结果表明网络诱导延时越大,系统控制性能越差。最后对仿真结果做了简单分析。     

Improved training of neural networks for the nonlinear activecontrol of sound and vibration

Active control of sound and vibration has been the subject of a lot of research, and examples of applications are now numerous. However, few practical implementations of nonlinear active controllers have been realized. Nonlinear active controllers may be required in cases where the actuators used in active control systems exhibit nonlinear characteristics, or in cases when the structure to be controlled exhibits a nonlinear behavior. A multilayer perceptron neural-network based control structure was previously introduced as a nonlinear active controller, with a training algorithm based on an extended backpropagation scheme. This paper introduces new heuristical training algorithms for the same neural-network control structure. The objective is to develop new algorithms with faster convergence speed and/or lower computational loads. Experimental results of active sound control using a nonlinear actuator with linear and nonlinear controllers are presented. The results show that some of the new algorithms can greatly improve the learning rate of the neural-network control structure, and that for the considered experimental setup a neural-network controller can outperform linear controllers.     

Hybrid intelligent control scheme of a polymerization kettle for ACR production

Polymerization kettle is the key controlled plant in ACR (Acrylate Copolymer Resin) production, which is a nonlinear time-delay system with parametric variance. However, modeling difficulties make the plant dynamic model poorly defined. A hybrid intelligent control scheme including an intelligent predictor is designed for this complex plant based on time-delay compensation theory. It consists of a Smith neural-network predictor and a self-adjusting-scaling-factor fuzzy logic controller. The simulation experiments verified the performance of our proposed system in two scenarios: one with invariant parameters and the other with time-varying parameters. Moreover, the comparison to other three typical control methods including Smith PID, Smith neural-network PID and Smith fuzzy logic control is also presented, which demonstrates that the proposed control scheme has satisfactory effect. Even when the system parameters vary with time, the proposed system still gives superior performance and improved robustness.     

Neural network design with genetic learning for control of a single link flexible manipulator

The application of neural networks for active control of lightly damped systems is considered in this article. The training process of the neural-network controller is based on the genetic learning algorithm. The schemes imitates nature's cleansing phenomena of natural selection and survival of the fittest to generate individual controllers withe best fitness values. It essentially incorporates an exhaustive search in the weight-space governed by the rituals of crossover and mutation to seek the optimum neural-network weights to satisfy certain performance criteria. Several appropriate modifications of the classical genetic algorithm for neural-network control purposes are discussed. The genetic-trained neural-network controller is applied for tip position tracking and vibration suppression of a single-link flexible arm. Simulation studies are presented to validate the effectiveness of the advocated algorithms.     

Stable neural controller design for unknown nonlinear systems using backstepping

We propose, from an adaptive control perspective, a neural controller for a class of unknown, minimum phase, feedback linearizable nonlinear system with known relative degree. The control scheme is based on the backstepping design technique in conjunction with a linearly parametrized neural-network structure. The resulting controller, however, moves the complex mechanics involved in a typical backstepping design from off-line to online. With appropriate choice of the network size and neural basis functions, the same controller can be trained online to control different nonlinear plants with the same relative degree, with semi-global stability as shown by the simple Lyapunov analysis. Meanwhile, the controller also preserves some of the performance properties of the standard backstepping controllers. Simulation results are shown to demonstrate these properties and to compare the neural controller with a standard backstepping controller.     

双馈风力发电机系统的自抗扰神经网络的励磁控制

在双馈发电机传统控制方式的基础上, 将自抗扰控制技术和BP神经网络相结合结合, 应用于双馈风力发电机并网运行的控制上, 提出了一种新的双馈风力发电机并网运行控制方案. 该控制方案具有内外两个控制环, 内环通过BP神经网络实现双馈风力发电机的转子d-q轴电流控制, 外环通过自抗扰技术实现双馈风力发电机定子侧的有功、无功控制. 由于自抗扰控制器利用一阶跟踪微分器和扩张状态观测器对系统扰动进行动态跟踪补偿, 在此基础上输出双馈电机转子交--直轴电流的参考值, 然后将该参考值作为BP神经网络训练样本的输入, 训练后的BP神经网络可以更好地逼近实际转子电压输出量. 论文设计并实现了该方案的具体控制算法. 仿真测试表明: 该控制方案具有优良的动态性能, 对系统的内外扰动具有较强的鲁棒性, 在没有精确的发电机参数情况下依然可实现并网系统的稳定运行.     

Adaptive non-linear compensation control based on neural networks for non-linear systems with time delay

This article investigates a new adaptive non-linear compensation controller for a class of time-delay non-linear systems with partly known dynamics. First, a non-linear neural-network(NN)-based identification model that includes a prior knowledge about the plant dynamics is discussed by using the approximation capabilities of NNs. Then, the adaptive non-linear compensation controller is developed to produce the desired tracking performance. The proposed controller based on the NN can reduce the effect of modelling uncertainties and provide the time-delay compensation, while stability of the closed-loop system is guaranteed. The effectiveness of the proposed scheme is demonstrated through the application to the control of a continuous stirred tank reactor.     

Intelligent control of a constant turning force system with fixed metal removal rate

The maintenance of a constant cutting force operation via control of the turning systems can increase the metal removal rate (MRR) and tool life. However, an increase in cutting depth reduces feed rate during the constant cutting force operation, resulting in lower productivity for the machine tool. To eliminate the problem, this study proposed an MRR scheme to assist a turning system in constructing a constant turning force system with fixed MRR. This study also presented a self-organizing fuzzy controller (SOFC) for manipulating such a system to maintain a constant turning force operation and improve the productivity of the machine tool. Nevertheless, it is difficult to select a suitable learning rate and an appropriate weighting distribution for the design of an SOFC. To overcome the difficulty, this study developed a hybrid self-organizing fuzzy and radial basis-function neural-network controller (HSFRBNC) for such turning systems. The HSFRBNC uses a radial basis function neural-network to adjust in real time the learning rate and the weighting distribution parameters of the SOFC to appropriate values, rather than obtaining the parameters by trial and error. This strategy solves the problem of determining appropriate parameters for designing an SOFC. Simulation results showed that the HSFRBNC achieved better control performance than the SOFC when it came to the control of a constant turning force system with fixed MRR.     

Application of neural networks to software quality modeling of avery large telecommunications system

Society relies on telecommunications to such an extent that telecommunications software must have high reliability. Enhanced measurement for early risk assessment of latent defects (EMERALD) is a joint project of Nortel and Bell Canada for improving the reliability of telecommunications software products. This paper reports a case study of neural-network modeling techniques developed for the EMERALD system. The resulting neural network is currently in the prototype testing phase at Nortel. Neural-network models can be used to identify fault-prone modules for extra attention early in development, and thus reduce the risk of operational problems with those modules. We modeled a subset of modules representing over seven million lines of code from a very large telecommunications software system. The set consisted of those modules reused with changes from the previous release. The dependent variable was membership in the class of fault-prone modules. The independent variables were principal components of nine measures of software design attributes. We compared the neural-network model with a nonparametric discriminant model and found the neural-network model had better predictive accuracy.