基于神经网络模型的直接优化预测控制

针对具有时延的非线性系统提出了一种基于神经网络模型直接优于的预测控制。     

Optimal PID automatic voltage regulator for synchronous machines

The design and testing of an optimal PID automatic voltage regulator for synchronous machines is treated. The proposed digital PID regulator combines automatic voltage regulation with the function of a power system stabilizer. The PID and stabilizing signal parameters are optimized based on a linear quadratic performance index using the Simplex method. The design of the regulator is achieved directly without using a linearised mathematical model, and only impulse responses of the system are required during the optimization procedure. The microcomputer-based regulator with digital filters has been tested on a laboratory model turbogenerator system. Simulation and experimental results are presented, showing that the regulator provides very good performance, which is superior to that obtained with a conventional automatic voltage regulator with power system stabilizer.     

Generalized Hamilton–Jacobi–Bellman Formulation -Based Neural Network Control of Affine Nonlinear Discrete-Time Systems

In this paper, we consider the use of nonlinear networks towards obtaining nearly optimal solutions to the control of nonlinear discrete-time (DT) systems. The method is based on least squares successive approximation solution of the generalized Hamilton-Jacobi-Bellman (GHJB) equation which appears in optimization problems. Successive approximation using the GHJB has not been applied for nonlinear DT systems. The proposed recursive method solves the GHJB equation in DT on a well-defined region of attraction. The definition of GHJB, pre-Hamiltonian function, HJB equation, and method of updating the control function for the affine nonlinear DT systems under small perturbation assumption are proposed. A neural network (NN) is used to approximate the GHJB solution. It is shown that the result is a closed-loop control based on an NN that has been tuned a priori in offline mode. Numerical examples show that, for the linear DT system, the updated control laws will converge to the optimal control, and for nonlinear DT systems, the updated control laws will converge to the suboptimal control.     

Generalized hamilton-jacobi-bellman formulation -based neural network control of affine nonlinear discrete-time systems

In this paper, we consider the use of nonlinear networks towards obtaining nearly optimal solutions to the control of nonlinear discrete-time (DT) systems. The method is based on least squares successive approximation solution of the generalized Hamilton-Jacobi-Bellman (GHJB) equation which appears in optimization problems. Successive approximation using the GHJB has not been applied for nonlinear DT systems. The proposed recursive method solves the GHJB equation in DT on a well-defined region of attraction. The definition of GHJB, pre-Hamiltonian function, HJB equation, and method of updating the control function for the affine nonlinear DT systems under small perturbation assumption are proposed. A neural network (NN) is used to approximate the GHJB solution. It is shown that the result is a closed-loop control based on an NN that has been tuned a priori in offline mode. Numerical examples show that, for the linear DT system, the updated control laws will converge to the optimal control, and for nonlinear DT systems, the updated control laws will converge to the suboptimal control.     

Application of multivariable frequency response methods to control of turbogenerators

This paper deals with the design, evaluation and testing of an integrated control scheme for a turbogenerator equipped with a high-gain thyristor exciter and an electro-hydraulio governing system. Multivariable frequency response methods are used, and these are demonstrated to be eminently suitable for the design and analysis of turbogenerator controllers. It is shown that a control scheme consisting of an automatic voltage regulator with speed stabilizer, and speed governor with lead compensator, designed by the above methods, can greatly improve the dynamic and transient performance of a turbogenerator.

This was confirmed by computer simulation, and by extensive tests on a laboratory model turbogenerator. The controllers in the oxcitor and governor loops are easily implemented, and the results show significant improvements in system damping, transient stability, and post-fault recovery of terminal voltage. It has thus been established that these controllers, designed on the basis of linearized mathematical models, work well in practice, at least in a laboratory environment.     

NN-based fuzzy control for TLP systems: A case study of practical structural parameters and wave properties

Although there are many successful applications of neural networks (NNs), however, there are still some drawbacks in using neural networks (NNs) in any control scheme. In this study an NN-based model is applied for a tension leg platform (TLP) system. A linear differential inclusion (LDI) state-space representation is constructed to represent the dynamics of the NN model. Control performance is achieved by using the parallel distributed compensation (PDC) scheme to ensure the stability of TLP systems subjected to an external wave force. In terms of the stability analysis, the linear matrix inequality (LMI) conditions are derived using the Lyapunov theory to guarantee the robustness design and stability of the TLP system. A simulation example based on practical data is given to demonstrate the feasibility of the proposed fuzzy control approach. In the end, we discuss a practical application with field data on the wave properties and structural characteristics. The results indicate the efficiency and robustness of the proposed NN based approach.     

Direct adaptive neural network controller for a class of nonlinear systems based on fuzzy estimator of the control error

A new approach of direct adaptive control of single input single output nonlinear systems in affine form using single-hidden layer neural network (NN) is introduced. In contrast to the algorithms in the literature, the weights adaptation laws are based on the control error and not on the tracking error or its filtered version. Since the control error is being expressed in terms of the NN controller, hence its weights updating laws are obtained via back-propagation concept. A fuzzy inference system (FIS) with heuristically defined rules is introduced to provide an estimate of this error based on the past history of the system behaviour. The stability of the closed loop is studied using Lyapunov theory. A fixed structure is then proposed for the FIS and the design parameters reduce to the parameters of the NN. The method is reproducible and does not require any pre-training of the network weights.     

Neural-network-observer-based optimal control for unknown nonlinear systems using adaptive dynamic programming

In this paper, an observer-based optimal control scheme is developed for unknown nonlinear systems using adaptive dynamic programming (ADP) algorithm. First, a neural-network (NN) observer is designed to estimate system states. Then, based on the observed states, a neuro-controller is constructed via ADP method to obtain the optimal control. In this design, two NN structures are used: a three-layer NN is used to construct the observer which can be applied to systems with higher degrees of nonlinearity and without a priori knowledge of system dynamics, and a critic NN is employed to approximate the value function. The optimal control law is computed using the critic NN and the observer NN. Uniform ultimate boundedness of the closed-loop system is guaranteed. The actor, critic, and observer structures are all implemented in real-time, continuously and simultaneously. Finally, simulation results are presented to demonstrate the effectiveness of the proposed control scheme.     

采样非线性系统的动态神经网络稳定自适应控制

研究了一类采样数据非线性系统的动态神经网络稳定自适应控制方法.不同于静态 神经网络自适应控制,动态神经网络自适应控制中神经网络用于逼近整个采样数据非线性系 统,而不是动态系统中的非线性分量.系统的控制律由神经网络系统的动态逆、自适应补偿项 和神经变结构鲁棒控制项组成.神经变结构控制用于保证系统的全局稳定性,并加速动态神 经网络系统的适近速度.证明了动态神经网络自适应控制系统的稳定性,并得到了动态神经 网络系统的学习算法.仿真研究表明,基于动态神经网络的非线性系统稳定自适应控制方法 较基于静态神经网络的自适应方法具有更好的性能.     

人工神经网络在汽轮发电机组故障诊断中的应用

基于ＢＰ神经网络，结合船舶汽轮发电机组故障诊断问题，提出了一个适合于复杂、非线性系统的块层化神经网络诊断模型，并开发了一个诊断系统。通过实验优化了网络参数，经在某船舶汽轮发电机组进行的运行实验，验证了诊断模型高效的自组织、自学习能力。初步试验结果表明诊断系统是有效和可行的。     

Nonlinear Adaptive Model Predictive Control of Constrained Systems with Offset‐Free Tracking Behavior

In this paper, a nonlinear model‐based predictive control strategy for constrained systems based on an adaptive neural network (NN) predictor is proposed. The proposed controller is robust against the model uncertainties and external bounded disturbances. Moreover, it provides offset‐free tracking behavior using the adaptive structure in the model. Based on the uncertainties bounds, the restriction of the system constraints causes robust feasibility and stability of the closed‐loop system. It is shown that the output of the NN predictor converges to the system output. Moreover, offset‐free behavior of the closed‐loop system is investigated using the Lyapunov theorem. Simulation results show the effectiveness of the proposed method as compared to the recently proposed model predictive control methods in the literature.     

Fuzzy and Recurrent Neural Network Motion Control among Dynamic Obstacles for Robot Manipulators

An integration of fuzzy controller and modified Elman neural networks (NN) approximation-based computed-torque controller is proposed for motion control of autonomous manipulators in dynamic and partially known environments containing moving obstacles. The fuzzy controller is based on artificial potential fields using analytic harmonic functions, a navigation technique common used in robot control. The NN controller can deal with unmodeled bounded disturbances and/or unstructured unmodeled dynamics of the robot arm. The NN weights are tuned on-line, with no off-line learning phase required. The stability of the closed-loop system is guaranteed by the Lyapunov theory. The purpose of the controller, which is designed as a neuro-fuzzy controller, is to generate the commands for the servo-systems of the robot so it may choose its way to its goal autonomously, while reacting in real-time to unexpected events. The proposed scheme has been successfully tested. The controller also demonstrates remarkable performance in adaptation to changes in manipulator dynamics. Sensor-based motion control is an essential feature for dealing with model uncertainties and unexpected obstacles in real-time world systems.     

Neural network learning from demonstration and epipolar geometry for visual control of a nonholonomic mobile robot

The control of a robot system using camera information is a challenging task regarding unpredictable conditions, such as feature point mismatch and changing scene illumination. This paper presents a solution for the visual control of a nonholonomic mobile robot in demanding real world circumstances based on machine learning techniques. A novel intelligent approach for mobile robots using neural networks (NNs), learning from demonstration (LfD) framework, and epipolar geometry between two views is proposed and evaluated in a series of experiments. A direct mapping from the image space to the actuator command is conducted using two phases. In an offline phase, NN–LfD approach is employed in order to relate the feature position in the image plane with the angular velocity for lateral motion correction. An online phase refers to a switching vision based scheme between the epipole based linear velocity controller and NN–LfD based angular velocity controller, which selection depends on the feature distance from the pre-defined interest area in the image. In total, 18 architectures and 6 learning algorithms are tested in order to find optimal solution for robot control. The best training outcomes for each learning algorithms are then employed in real time so as to discover optimal NN configuration for robot orientation correction. Experiments conducted on a nonholonomic mobile robot in a structured indoor environment confirm an excellent performance with respect to the system robustness and positioning accuracy in the desired location.     

Model predictive control of resonant systems using Kautz model

The scope of this paper broadly spans in two areas: system identification of resonant system and design of an efficient control scheme suitable for resonant systems. Use of filters based on orthogonal basis functions (OBF) have been advocated for modelling of resonant process. Kautz filter has been identified as best suited OBF for this purpose. A state space based system identification technique using Kautz filters, viz. Kautz model, has been demonstrated. Model based controllers are believed to be more efficient than classical controllers because explicit use of process model is essential with these modelling techniques. Extensive literature search concludes that very few reports are available which explore use of the model based control studies on resonant system. Two such model based controllers are considered in this work, viz. model predictive controller and internal model controller. A model predictive control algorithm has been developed using the Kautz model. The efficacy of the model and the controller has been verified by two case studies, viz. linear second order underdamped process and a mildly nonlinear magnetic ball suspension system. Comparative assessment of performances of these controllers in those case studies have been carried out.     

Direct adaptive control for a class of MIMO nonlinear systems using neural networks

In this note, direct adaptive neural network (NN) control is studied for a class of multiple-input-multiple-output nonlinear systems based on input-output discrete-time model with unknown interconnections between subsystems. By finding an orthogonal matrix to tune the NN weights, the closed-loop system is proven to be semiglobally uniformly ultimately bounded. The control performance of the closed-loop system is guaranteed by suitably choosing the design parameters.     

Hybrid partial least squares and neural network approach for short-term electrical load forecasting

Intelligent systems and methods such as the neural network (NN) are usually used in electric power systems for short-term electrical load forecasting. However, a vast amount of electrical load data is often redundant, and linearly or nonlinearly correlated with each other. Highly correlated input data can result in erroneous prediction results given out by an NN model. Besides this, the determination of the topological structure of an NN model has always been a problem for designers. This paper presents a new artificial intelligence hybrid procedure for next day electric load forecasting based on partial least squares (PLS) and NN. PLS is used for the compression of data input space, and helps to determine the structure of the NN model. The hybrid PLS-NN model can be used to predict hourly electric load on weekdays and weekends. The advantage of this methodology is that the hybrid model can provide faster convergence and more precise prediction results in comparison with abductive networks algorithm. Extensive testing on the electrical load data of the Puget power utility in the USA confirms the validity of the proposed approach.     

Hybrid partial least squares and neural network approach for short-term electrical load forecasting

Intelligent systems and methods such as the neural network （NN） are usually used in electric power systems for short-term electrical load forecasting. However, a vast amount of electrical load data is often redundant, and linearly or nonlinearly correlated with each other. Highly correlated input data can result in erroneous prediction results given out by an NN model. Besides this, the determination of the topological structure of an NN model has always been a problem for designers. This paper presents a new artificial intelligence hybrid procedure for next day electric load forecasting based on partial least squares （PLS） and NN. PLS is used for the compression of data input space, and helps to determine the structure of the NN model. The hybrid PLS-NN model can be used to predict hourly electric load on weekdays and weekends. The advantage of this methodology is that the hybrid model can provide faster convergence and more precise prediction results in comparison with abductive networks algorithm. Extensive testing on the electrical load data of the Puget power utility in the USA confirms the validity of the proposed approach.     

基于遗传算法的神经网络自适应控制器的研究

提出了一种基于遗传算法的神经网络自适应控制方法。该方法是针对ＢＰ算法训练神经网络控制系统时收敛速度慢、动态特性不够理想等不足，用改进的遗传算法来优化神经网络辨识器与控制器的参数，以提高控制系统的性能，仿真实验表明该控制器对于非线性、时变、滞后等对象都具有很好的控制精度、鲁棒性和动态特性。     

弧焊机器人焊缝跟踪神经网络控制器

介绍了一种能提高高弧焊机器人焊缝跟踪精度的神经网络控制器,通过神经网络的补偿作用,弥补了由于无法知道机器人精确模型所造成的控制上的误差,不同于机器人控制中传统的网络控制器,本文提出并应用了基于笛卡尔空间轨迹控制的机器人焊缝跟踪神经网络,大大简化了控制算法,计算机模拟及实验表明,该控制器非常适用于只的实际焊接,对于现有机器人,无须改变其控制器内部结构,即可应用该技术,与常用的机器人关节力矩控制法相比     

Multiple Lyapunov functions‐based adaptive neural network tracking control of uncertain switched nonlinear systems

In this paper, the problem of adaptive neural network (NN) tracking control of a class of switched strict‐feedback uncertain nonlinear systems is investigated by state‐feedback, in which the solvability of the problem of adaptive NN tracking control for individual subsystems is unnecessary. A multiple Lyapunov functions (MLFs)–based adaptive NN tracking control scheme is established by exploiting backstepping and the generalized MLFs approach. Moreover, based on the proposed scheme, adaptive NN controllers of all subsystems and a state‐dependent switching law simultaneously are constructed, which guarantee that all signals of the resulting closed‐loop system are semiglobally uniformly ultimately bounded, and the tracking error converges to a small neighborhood of the origin. The scheme provided permits removal of a technical condition in which the adaptive NN tracking control problem for individual subsystems is solvable. Finally, the effectiveness of the design scheme proposed is shown by using two examples.