A Hammerstein Recurrent Neurofuzzy Network With an Online Minimal Realization Learning Algorithm

This paper presents a Hammerstein recurrent neurofuzzy network associated with an online minimal realization learning algorithm for dealing with nonlinear dynamic applications. We fuse the concept of states in linear systems into a neurofuzzy framework so that the whole structure can be expressed by a state-space representation. An online minimal realization learning algorithm has been developed to find a controllable and observable state-space model of minimal size from the input–output measurements of a given system. Such an idea can simultaneously resolve the problem of the determination of a minimal structure and the difficulty of network stability analysis. The advantages of our approach include: 1) our recurrent network is capable of translating the complicated dynamic behavior of a nonlinear system into a minimal set of linguistic fuzzy dynamical rules and into state-space representation as well and 2) an online minimal realization learning algorithm unifies an order determination algorithm, a hybrid parameter initialization method, and a recursive recurrent learning algorithm into a systematic procedure to identify a minimal structure with satisfactory performance. Performance evaluations on benchmark examples as well as real-world applications have successfully validated the effectiveness of our approach.      

Control of a direct internal reforming molten carbonate fuel cell system using wavelet network-based Hammerstein models

Inspired by a direct internal reforming molten carbonate fuel cell (DIR-MCFC) coupled with complicated nonlinear dynamics, the identification and control design of the Hammerstein model is presented. Through the sequential identification procedure, the static nonlinearity block is considered as the wavelet network which is trained and validated by the on-line learning algorithm, and the linear dynamic block is described by the state-space model in which parameters are estimated by the recursive least square algorithm. Using the numerical interpolation technique to approximate the implicit nonlinear function, we present a composite control framework consists of a nonlinear inversion and linear control. Through the closed-loop simulation tests, the nonlinear inversion design for the nonlinearity cancellation of a class of nonlinear systems is validated.     

Neural-network predictive control for nonlinear dynamic systems with time-delay

A new recurrent neural-network predictive feedback control structure for a class of uncertain nonlinear dynamic time-delay systems in canonical form is developed and analyzed. The dynamic system has constant input and feedback time delays due to a communications channel. The proposed control structure consists of a linearized subsystem local to the controlled plant and a remote predictive controller located at the master command station. In the local linearized subsystem, a recurrent neural network with on-line weight tuning algorithm is employed to approximate the dynamics of the time-delay-free nonlinear plant. No linearity in the unknown parameters is required. No preliminary off-line weight learning is needed. The remote controller is a modified Smith predictor that provides prediction and maintains the desired tracking performance; an extra robustifying term is needed to guarantee stability. Rigorous stability proofs are given using Lyapunov analysis. The result is an adaptive neural net compensation scheme for unknown nonlinear systems with time delays. A simulation example is provided to demonstrate the effectiveness of the proposed control strategy.     

On-line training of recurrent neural networks with continuoustopology adaptation

This paper presents an online procedure for training dynamic neural networks with input-output recurrences whose topology is continuously adjusted to the complexity of the target system dynamics. This is accomplished by changing the number of the elements of the network hidden layer whenever the existing topology cannot capture the dynamics presented by the new data. The training mechanism is based on the suitably altered extended Kalman filter (EKF) algorithm which is simultaneously used for the network parameter adjustment and for its state estimation. The network consists of a single hidden layer with Gaussian radial basis functions (GRBF), and a linear output layer. The choice of the GRBF is induced by the requirements of the online learning. The latter implies the network architecture which permits only local influence of the new data point in order not to forget the previously learned dynamics. The continuous topology adaptation is implemented in our algorithm to avoid memory and computational problems of using a regular grid of GRBF'S which covers the network input space. Furthermore, we show that the resulting parameter increase can be handled "smoothly" without interfering with the already acquired information. If the target system dynamics are changing over time, we show that a suitable forgetting factor can be used to "unlearn" the no longer-relevant dynamics. The quality of the recurrent network training algorithm is demonstrated on the identification of nonlinear dynamic systems.     

Dynamics identification and control of nonlinear MIMO coupled plant using supervised neural gas and comparison with recurrent neural controller

The dynamics identification and subsequent control of a nonlinear system is not a trivial issue. The application of a neural gas network that is trained with a supervised batch version of the algorithm can produce identification models in a robust way. In this paper, the neural model identifies each local transfer function, demonstrating that the local linear approximation can be done. Moreover, other parameters are analyzed in order to obtain a correct modeling. Furthermore, the algorithm is applied to control a nonlinear multi-input multi-output system composed of tanks. In addition, this plant is a coupled system where the manipulated input variables are influencing all the output variables. The aim of the work is to demonstrate that the supervised neural gas algorithm is able to obtain linear models to be used in a state space design scenario to control nonlinear coupled systems and guarantee a robust control method. The results are compared with the common approach of using a recurrent neural controller trained with a dynamic backpropagation algorithm. Regarding the steady-state errors in disturbance rejection, reference tracking and sensitivity to simple process changes, the proposed approach shows an interesting application to control nonlinear plants.

     

A context layered locally recurrent neural network for dynamic system identification

This paper puts forward a novel recurrent neural network (RNN), referred to as the context layered locally recurrent neural network (CLLRNN) for dynamic system identification. The CLLRNN is a dynamic neural network which appears in effective in the input–output identification of both linear and nonlinear dynamic systems. The CLLRNN is composed of one input layer, one or more hidden layers, one output layer, and also one context layer improving the ability of the network to capture the linear characteristics of the system being identified. Dynamic memory is provided by means of feedback connections from nodes in the first hidden layer to nodes in the context layer and in case of being two or more hidden layers, from nodes in a hidden layer to nodes in the preceding hidden layer. In addition to feedback connections, there are self-recurrent connections in all nodes of the context and hidden layers. A dynamic backpropagation algorithm with adaptive learning rate is derived to train the CLLRNN. To demonstrate the superior properties of the proposed architecture, it is applied to identify not only linear but also nonlinear dynamic systems. The efficiency of the proposed architecture is demonstrated by comparing the results to some existing recurrent networks and design configurations. In addition, performance of the CLLRNN is analyzed through an experimental application to a dc motor connected to a load to show practicability and effectiveness of the proposed neural network. Results of the experimental application are presented to make a quantitative comparison with an existing recurrent network in the literature.     

Decentralized adaptive recurrent neural control structure

This paper presents a novel decentralized variable structure neural control approach for large-scale uncertain systems, which is developed using recurrent high-order neural networks (RHONN). It is assumed that each subsystem belongs to a class of block-controllable nonlinear systems whose vector fields includes interconnection terms, which are bounded by nonlinear functions. A decentralized RHONN structure and the respective learning law are proposed in order to approximate online the dynamical behavior of each nonlinear subsystem. The control law, which is able to regulate and to track the desired reference signals, is designed using the well-known variable structure theory. The stability of the whole system is analyzed via the Lyapunov methodology. The applicability of the proposed decentralized identification and control algorithm is illustrated via simulations as applied to an interconnected double inverted pendulum.     

Neurocontrol of nonlinear dynamical systems with Kalman filtertrained recurrent networks

Although the potential of the powerful mapping and representational capabilities of recurrent network architectures is generally recognized by the neural network research community, recurrent neural networks have not been widely used for the control of nonlinear dynamical systems, possibly due to the relative ineffectiveness of simple gradient descent training algorithms. Developments in the use of parameter-based extended Kalman filter algorithms for training recurrent networks may provide a mechanism by which these architectures will prove to be of practical value. This paper presents a decoupled extended Kalman filter (DEKF) algorithm for training of recurrent networks with special emphasis on application to control problems. We demonstrate in simulation the application of the DEKF algorithm to a series of example control problems ranging from the well-known cart-pole and bioreactor benchmark problems to an automotive subsystem, engine idle speed control. These simulations suggest that recurrent controller networks trained by Kalman filter methods can combine the traditional features of state-space controllers and observers in a homogeneous architecture for nonlinear dynamical systems, while simultaneously exhibiting less sensitivity than do purely feedforward controller networks to changes in plant parameters and measurement noise.     

基于多面体终端域的Hammerstein-Wiener非线性预测控制

许多实际系统都可以表示成一种中间为线性动态子系统、输入输出端为非线性静态子系统的Hammerstein-Wiener型非线性模型. 针对输入和输出受约束的Hammerstein-Wiener型非线性系统, 提出一种基于多面体终端域的预测控制综合算法. 离线设计时, 通过构造一系列多面体不变集, 扩大了终端域; 在多面体不变集内, 设计非线性控制律, 减少了常规线性控制律设计的保守性. 在线计算时, 通过求解有限个线性矩阵不等式(Linear matrix inequalities, LMIs)优化问题, 不仅可以满足实时性要求, 而且能够改善控制性能. 仿真结果表明了采用多面体不变集的优越性.     

非线性系统的回归网络辨识

针对未知非线性系统的辨识问题,本文提出了一种新型的回归网络模型,证明了该网络模型在一定条件下能够逼近非线性系统的输入输出关系,提出了训练网络前向连接和反向连接权值的动态反向传播算法,伪真结果验证该方法的有效性。     

A new neural network-based approach for self-tuning control of nonlinear SISO discrete-time systems

This article presents a new neural network-based approach for self-tuning control of nonlinear single-input single-output (SISO) discrete-time dynamic systems. According to the approach, a neural network ARMAX (NN-ARMAX) model of the system is identified and continuously updated, using an online training algorithm. Control design is accomplished by solving an optimal discrete-time linear quadratic tracking problem using an observer-type linear state-space Kalman innovation model, which is built from the parameters of a local linear version of the NN-ARMAX model. The state-feedback control law is implemented using the Kalman state, which is calculated without estimating the noise covariance properties. The proposed control approach is shown to be very effective and outperforms the self-tuning control approach based on a linear ARMAX model on two simulation examples.     

On-line system identification of complex systems using Chebyshev neural networks

This paper proposes a computationally efficient artificial neural network (ANN) model for system identification of unknown dynamic nonlinear discrete time systems. A single layer functional link ANN is used for the model where the need of hidden layer is eliminated by expanding the input pattern by Chebyshev polynomials. Thus, creation of nonlinear decision boundaries in the multidimensional input space and approximation of complex nonlinear systems becomes easier. These models are linear in their parameters and nonlinear in the inputs. The recursive least squares method with forgetting factor is used as on-line learning algorithm for parameter updation. The good behaviour of the identification method is tested on Box and Jenkins Gas furnace benchmark identification problem, single input single output (SISO) and multi input multi output (MIMO) discrete time plants. Stability of the identification scheme is also addressed.     

Identification of continuous-time Hammerstein systems by simultaneous perturbation stochastic approximation

This paper proposes an identification method for Hammerstein systems using simultaneous perturbation stochastic approximation (SPSA). Here, the structure of nonlinear subsystem is assumed to be unknown, while the structure of linear subsystem, such as the system order, is assumed to be available. The main advantage of the SPSA-based method is that it can be applied to identification of Hammerstein systems with less restrictive assumptions. In order to clarify this point, piecewise affine functions with a large number of parameters are adopted to approximate the unknown nonlinear subsystems. Furthermore, the linear subsystems are supposed to be described in continuous-time. Though this class of systems closely reflects the actual systems, there are few methods to identify such models. Hence, the SPSA-based method is utilized to identify the parameters in both linear and nonlinear subsystems simultaneously. The effectiveness of the proposed method is evaluated through several numerical examples. The results demonstrate that the proposed algorithm is useful to obtain accurate models, even for high-dimensional parameter identification.     

Adaptive recurrent neural network control using a structure adaptation algorithm

This paper proposes an adaptive recurrent neural network control (ARNNC) system with structure adaptation algorithm for the uncertain nonlinear systems. The developed ARNNC system is composed of a neural controller and a robust controller. The neural controller which uses a self-structuring recurrent neural network (SRNN) is the principal controller, and the robust controller is designed to achieve L ² tracking performance with desired attenuation level. The SRNN approximator is used to online estimate an ideal tracking controller with the online structuring and parameter learning algorithms. The structure learning possesses the ability of both adding and pruning hidden neurons, and the parameter learning adjusts the interconnection weights of neural network to achieve favorable approximation performance. And, by the L ² control design technique, the worst effect of approximation error on the tracking error can be attenuated to be less or equal to a specified level. Finally, the proposed ARNNC system with structure adaptation algorithm is applied to control two nonlinear dynamic systems. Simulation results prove that the proposed ARNNC system with structure adaptation algorithm can achieve favorable tracking performance even unknown the control system dynamics function.     

An adaptive recurrent fuzzy system for nonlinear identification

This paper describes the architecture and training procedure of a recurrent fuzzy system (RFS). The RFS is composed of a fuzzy inference system (FIS) and a delayed feedback connection. The recurrent property comes from feeding the FIS output back to the FIS input via an adjustable feedback parameter. Both the on-line and off-line training procedures based on the backpropagation-through-time (BPTT) algorithm have been investigated. The adjoint model of the RFS is obtained and used to compute the gradients. It is shown that the off-line training is insufficient to adapt to changes in system dynamics. So, an on-line training procedure is derived. In this procedure, a first in first out stack is used to store a certain history of the input–output data to perform a truncated BPTT algorithm. A quasi-Newton optimization method with a line search algorithm is used to adjust the RFS parameters. The performance of the developed RFS is demonstrated by applying to the identification of nonlinear dynamic systems. The simulation studies show that the proposed identification model has the ability to learn dynamics of highly nonlinear systems and compensate system uncertainties. The results are promising for the further application in the area of control and modeling.     

Identification Recurrent Type 2 Fuzzy Wavelet Neural Network and L2‐Gain Adaptive Variable Sliding Mode Robust Control of Electro‐Hydraulic Servo System (EHSS)

An electro‐hydraulic servo system (EHSS) is a kind of system with the characteristics of time‐variant, serious nonlinearity, parameter and structural uncertainty, and uncertain load disturbance in most cases. These characteristics make it very difficult to realize highly accurate control by conventional methods. In order to solve the above problems, this paper introduces a recurrent type 2 fuzzy wavelet neural network to approximate the unknown nonlinear functions of the dynamic systems through tuning by the desired adaptive law. Based on the identification by recurrent type 2 fuzzy wavelet neural network, a L2 gain design method, combining gain adaptive variable sliding mode control with H infinity control, is proposed for load disturbance, thereby accommodating uncertainties that are the main factors affecting system stability and accuracy in EHSS. In this algorithm, a recurrent type 2 fuzzy wavelet neural network is employed to evaluate the unknown dynamic characteristics of the system and gain adaptive variable sliding mode control to compensate for evaluating errors, and H infinity control to suppress the effect on system by load disturbance. The experiment results show that the proposed system L2 gain design method can make the system exhibit strong robustness to parameter variation and load disturbance.     

Adaptive Control Scheme for Large‐Scale Interconnected Systems Described by Hammerstein Models

This paper proposes an adaptive algorithm for the online control of discrete‐time large‐scale nonlinear systems, which reduces the noise effects acting on the system output (regulation problem) and allows the system output to keep track of a time‐varying trajectory (tracking problem). We consider a large‐scale nonlinear system that can be decomposed into single‐input single‐output (SISO) interconnected nonlinear subsystems with known structure variables (orders, delays) and unknown time‐varying parameters. Each interconnected subsystem is described by block‐oriented models, specifically a discrete‐time Hammerstein model. Parameter adaptation is performed using a recursive parametric estimation algorithm based on the adjustable model method and the least squares techniques. Simulation results of an interconnected petroleum process are provided to demonstrate the effectiveness of the developed control scheme.     

Recursive identification of errors-in-variables Wiener systems

This paper considers the recursive identification of errors-in-variables (EIV) Wiener systems composed of a linear dynamic system followed by a static nonlinearity. Both the system input and output are observed with additive noises being ARMA processes with unknown coefficients. By a stochastic approximation incorporated with the deconvolution kernel functions, the recursive algorithms are proposed for estimating the coefficients of the linear subsystem and for the values of the nonlinear function. All the estimates are proved to converge to the true values with probability one. A simulation example is given to verify the theoretical analysis.     

A Neural Net Predictive Control for Telerobots with Time Delay

This paper extends the Smith Predictor feedback control structure to unknown robotic systems in a rigorous fashion. A new recurrent neural net predictive control (RNNPC) strategy is proposed to deal with input and feedback time delays in telerobotic systems. The proposed control structure consists of a local linearized subsystem and a remote predictive controller. In the local linearized subsystem, a recurrent neural network (RNN) with on-line weight tuning algorithm is employed to approximate the dynamics of the time-delay-free nonlinear plant. The remote controller is a modified Smith predictor for the local linearized subsystem which provides prediction and maintains the desirable tracking performance. Stability analysis is given in the sense of Lyapunov. The result is an adaptive compensation scheme for unknown telerobotic systems with time delays, uncertainties, and external disturbances. A simulation of a two-link robotic manipulator is provided to illustrate the effectiveness of the proposed control strategy.     

Adaptive neural control for a class of stochastic nonlinear time‐delay systems with unknown dead zone using dynamic surface technique

This paper investigates the problem of adaptive control for a class of stochastic nonlinear time‐delay systems with unknown dead zone. A neural network‐based adaptive control scheme is developed by using the dynamic surface control (DSC) technique and the minimal learning parameters algorithm. The dynamic surface control technique, which can avoid the problem of ‘explosion of complexity’ inherent in the conventional backstepping design procedure, is first extended to the stochastic nonlinear time‐delay system with unknown dead zone. The unknown nonlinearities are approximated by the function approximation technique using the radial basis function neural network. For the purpose of reducing the numbers of parameters, which are updated online for each subsystem in the process of approximating the unknown functions, the minimal learning parameters algorithm is then introduced. Also, the adverse effects of unknown time‐delay are removed by using the appropriate Lyapunov–Krasovskii functionals. In addition, the proposed control scheme is systematically derived without requiring any information on the boundedness of the dead zone parameters and avoids the possible controller singularity problem in the approximation‐based adaptive control schemes with feedback linearization technique. It is shown that the proposed control approach can guarantee that all the signals of the closed‐loop system are bounded in probability, and the tracking errors can be made arbitrary small by choosing the suitable design parameters. Finally, a simulation example is provided to illustrate the performance of the proposed control scheme. Copyright © 2015 John Wiley & Sons, Ltd.