Neural Networks For Nonlinear Systems Research Articles

Event-triggered model-free adaptive control for nonlinear systems using intuitionistic fuzzy neural network: simulation and experimental validation

This article presents model-free adaptive control based on an intuitionistic fuzzy neural network for nonlinear systems with event-triggered output. Essentially, model-free adaptive control (MFAC) is constructed by establishing an online approximate model of the controlled system using the pseudo-partial derivative (PPD) form. By the proposed scheme, first, an intuitionistic fuzzy neural network (IFNN) is developed as an estimator for time-varying PPD in both compact-form dynamic linearization (CFDL) and partial-form dynamic linearization (PFDL) for the MFAC technique. Second, two periodic event-triggered output methods are integrated with the proposed IFNN-based MFAC in both forms to save communication resources and reduce the computation burden and energy consumption. Based on the Lyapunov theory and BIBO stability approach, necessary conditions are established to guarantee the convergence of the adaptive law of the IFNN controller and the boundary of the tracking error of the closed loop system. Third, regarding the feasibility and the effectiveness of the developed control method, two simulation examples including the continuous stirred-tank reactor (CSTR) system and the heat exchanger system are given. Finally, the practical validation of the proposed data-driven control method is conducted via the speed control of a DC motor.

Iterative Learning Model Predictive Control With Fuzzy Neural Network for Nonlinear Systems

Due to the existence of strong nonlinearity and external disturbances, the controller design of complex nonlinear systems is a challenging problem. Therefore, it is necessary to design an effective robust predictive controller for this issue. In this paper, based on a fuzzy neural network, an iterative learning model predictive control (FNN-ILMPC) is designed for complex nonlinear systems. Firstly, a dynamic linearization technique (DIT) is used to establish a data-driven model, which only relies on input and output data. Since the established model contains an unknown disturbance term that may have an impact on the control performance, an FNN is used to evaluate the disturbance so that the uncertainty of the system is captured. Subsequently, based on the above data-driven model, an FNN-ILMPC strategy, considering the impact of external disturbances, is developed to eliminate the influence of disturbances. Then, it is proved that the designed controller can make both modeling error and tracking error decrease gradually and ensure the closed-loop system stability. Finally, experimental results verify the effectiveness and superiority of the designed controller.

Real time adaptive PID controller based on quantum neural network for nonlinear systems

In this paper, an adaptive proportional-integral-derivative (PID) controller based on a quantum neural network (APIDC-QNN) is introduced. In the proposed neural network structure, three neural layers are used namely: input layer, hidden layer, and output layer. The input and hidden layers use three and six quantum-processing neurons respectively. In contrast, in the output layer, the multiplication processing is used as an activation function free of the output weights which aims to reduce the number of tunable parameters. The output layer with three neurons represents the adaptive PID controller parameters. In this work, the tunable parameters are updated using the Lyapunov stability criterion to guarantee optimization and learning stability. In addition, the plant identification based on the Wiener-type model and quantum neural network is introduced to estimate the sensitivity of the plant output to the control signal. The proposed neural network structure yields only 30 tunable parameters for the proposed controller and 13 tunable parameters for the proposed identifier while the conventional neural network structure in other works requires 90 tunable parameters just for the controller. To evaluate the performance of the proposed APIDC-QNN, it is practically operated on a non-holonomic two-wheel mobile robot (NHWMR). Moreover, to investigate the robustness, superiority, and convergence speed of the proposed APIDC-QNN, some experimental tasks are discussed. The simulation and practical results show the powerful processing and the superiority of the proposed APIDC-QNN over that designed based on conventional neural networks by recording the minimum values of the performance indices.

Tunable Predefined Time Nonsingular Terminal Sliding Mode Control Based on Neural Network for Nonlinear Systems

Tunable Predefined Time Nonsingular Terminal Sliding Mode Control Based on Neural Network for Nonlinear Systems

Intelligent computing with Levenberg-Marquardt artificial neural networks for nonlinear system of COVID-19 epidemic model for future generation disease control.

The aim of this work is to design an intelligent computing paradigm through Levenberg–Marquardt artificial neural networks (LMANNs) for solving the mathematical model of Corona virus disease 19 (COVID-19) propagation via human to human interaction. The model is represented with systems of nonlinear ordinary differential equations represented with susceptible, exposed, symptomatic and infectious, super spreaders, infection but asymptomatic, hospitalized, recovery and fatality classes, and reference dataset of the COVID-19 model is generated by exploiting the strength of explicit Runge–Kutta numerical method for metropolitans of China and Pakistan including Wuhan, Karachi, Lahore, Rawalpindi and Faisalabad. The created dataset is arbitrary used for training, validation and testing processes for each cyclic update in Levenberg–Marquardt backpropagation for numerical treatment of the dynamics of COVID-19 model. The effectiveness and reliable performance of the design LMANNs are endorsed on the basis of assessments of achieved accuracy in terms of mean squared error based merit functions, error histograms and regression studies.

Robust Adaptive Recurrent Cerebellar Model Neural Network for Non-linear System Based on GPSO

A robust adaptive recurrent cerebellar model articulation controller (RARC) neural network for non-linear systems using the genetic particle swarm optimization (GPSO) algorithm is presented in this study. The RARC is used as the principal tracking controller and the robust compensation controller is designed to recover the residual of the approximation error. In the RARC neural network, the steepest descent gradient method and the Lyapunov function are used for deriving the adaptive law parameter of the system. Besides, the learning rates play an important role in these adaptive laws and they have a great effect on the functions of control systems. In this paper, the combination of the genetic algorithm with the mutation particle swarm optimization algorithm is applied to seek for the optimal learning rates of the RARC adaptation laws. The numerical simulations about the inverted pendulum system as well as the robot manipulator system are given to confirm the effectiveness and practicability of the GPSO-RARC-based control system. Compared with other control schemes, the proposed control scheme is testified to be reliable and can obtain the optimal parameter about the learning rates and the minimum root mean square error for non-linear systems.

An improved adaptive switching control based on quasi-ARX neural network for nonlinear systems

In this paper, an improved switching mechanism based on quasi-linear auto regressive exogenous (quasi-ARX) neural network (QARXNN) is presented for the adaptive control of nonlinear systems. The proposed switching control is composed of a QARXNN-based prediction model and an improved switching mechanism using two new adaptive control laws, first is moving average filter law and second is new switching law. Since the control result of nonlinear predictor is better than the linear predictor in most of the time, the adaptive control with a simple switching mechanism has many useless switching during the processing. Hence, the improved smooth switching mechanism is proposed to replace the original switching mechanism; it can improve the performance by reducing the useless switching while guaranteeing stability of the system control. The simulations show that the efficiency of the proposed control method is satisfied in stability, improve control accuracy and robustness.

An improved elman neural network controller based on quasi‐ARX neural network for nonlinear systems

AbstractAn improved Elman neural network (IENN) controller with particle swarm optimization (PSO) is presented for nonlinear systems. The proposed controller is composed of a quasi‐ARX neural network (QARXNN) prediction model and a switching mechanism. The switching mechanism is used to guarantee that the prediction model works well. The primary controller is designed based on IENN using the backpropagation (BP) learning algorithm with PSO. PSO is used to adjust the learning rates in the BP process for improving the learning capability. The adaptive learning rates of the controller are investigated via the Lyapunov stability theorem. The proposed controller performance is verified through numerical simulation. The method is compared with the fuzzy switching and 0/1 switching methods to show its effectiveness in terms of stability, accuracy, and robustness. © 2014 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Tracking Control Based on Recurrent Neural Networks for Nonlinear Systems with Multiple Inputs and Unknown Deadzone

This paper deals with the problem of trajectory tracking for a broad class of uncertain nonlinear systems with multiple inputs each one subject to an unknown symmetric deadzone. On the basis of a model of the deadzone as a combination of a linear term and a disturbance‐like term, a continuous‐time recurrent neural network is directly employed in order to identify the uncertain dynamics. By using a Lyapunov analysis, the exponential convergence of the identification error to a bounded zone is demonstrated. Subsequently, by a proper control law, the state of the neural network is compelled to follow a bounded reference trajectory. This control law is designed in such a way that the singularity problem is conveniently avoided and the exponential convergence to a bounded zone of the difference between the state of the neural identifier and the reference trajectory can be proven. Thus, the exponential convergence of the tracking error to a bounded zone and the boundedness of all closed‐loop signals can be guaranteed. One of the main advantages of the proposed strategy is that the controller can work satisfactorily without any specific knowledge of an upper bound for the unmodeled dynamics and/or the disturbance term.

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.

Fuzzy neural networks for nonlinear systems modelling

A technique for the modelling of nonlinear systems using a fuzzy neural network topology is described. The input space of a nonlinear system is initially divided into a number of fuzzy operating regions within which reduced order models are able to represent the system. The complete system model output, the global model, is obtained through the conjunction of the outputs of the local models. The fuzzy neural network approach to nonlinear process modelling provides a way of opening up the purely 'black box' approach normally seen in neural network applications. Process knowledge is used to identify appropriate local operating regions and as an aid to initialising the network structure. Fuzzy neural network models are also easier to interpret than conventional neural network models. The weights in a trained fuzzy network model can be interpreted in terms of process information. This technique has been applied to model the nonlinear dynamic behaviour of a pH reactor and two static nonlinear systems. Correlation based tests are used to assess the fuzzy network model validity for nonlinear dynamic systems.

Nonlinear Modeling Using Neural Networks with Multiresolution Representations

This paper discusses neural networks that use multiresolution representation signals as their input, and the use of those neural networks for modeling nonlinear chemical systems. Decomposing an original signal into a multiresolution representation enables its analysis at different resolutions. By using a multiresolution representation as its input, neural networks can substantially expand the time range of a signal. According to wavelet theory, these decomposition signals can be used to easily reconstruct an original signal, and certain types of feedforward neural networks soon are able to learn this reconstruction. We have studied the modeling with this type of neural network for a nonlinear system with a highly underdamping system and a continuous polymerization reactor (CSTR).