scholarly journals Research on Nonlinear Systems Modeling Methods Based on Neural Networks

2021 ◽  
Vol 2095 (1) ◽  
pp. 012037
Author(s):  
Ting Shi ◽  
Wu Yang ◽  
Junfei Qiao
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Nonlinear Systems ◽  
Industrial Production ◽  
Industrial Process ◽  
System Control ◽  
Nonlinear Fitting ◽  
Performance Differences ◽  
Nonlinear System Control ◽  
Process Prediction

Abstract Nonlinear systems widely exist in all fields of industrial production and are difficult to model because of complex non-linearity. Neural network is widely used in process prediction, fault detection and fault diagnosis of modern industry because of the nonlinear fitting ability. Due to various structures, there exists diversity in the performance of neural networks. However, only the appropriate network can improve the efficiency and safety in modelling nonlinear industrial process, which requires full consideration of the structure of neural network. In this study, several typical structures of neural networks are compared and analysed, and the performance differences caused by these structures are presented in detail. Finally, performance differences of neural networks with inconsistent structures are verified on several experiments. The results showed that neural networks with inconsistent structures were good at dealing with different types of nonlinear systems. Our work will provide a theoretical basis in accurately modeling the industrial production process, which is beneficial to nonlinear system control.

scholarly journals An adaptive PID neural network for complex nonlinear system control

2014 ◽  
Vol 135 ◽  
pp. 79-85 ◽  
Author(s):  
Jun Kang ◽  
Wenjun Meng ◽  
Ajith Abraham ◽  
Hongbo Liu
Keyword(s):  
Neural Network ◽  
Nonlinear System ◽  
System Control ◽  
Nonlinear System Control ◽  
Complex Nonlinear System ◽  
Pid Neural Network

Study on the Applications of Neural Networks for Processing Deformation Monitoring Data

2014 ◽  
Vol 501-504 ◽  
pp. 2149-2153 ◽  
Author(s):  
Cai Yun Gao ◽  
Xi Min Cui ◽  
Xue Qian Hong
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Prediction Models ◽  
Deformation Monitoring ◽  
Monitoring Data ◽  
Nonlinear Fitting ◽  
High Rise ◽  
High Rise Building ◽  
Good Ability ◽  
Neural Network Prediction

Accurately estimating the deformation of high-rise building is a very important work for surveyors, however it is very difficult to get an accurate and reliable predictor. In this paper, artificial neural network has been applied here because of its good ability of nonlinear fitting. On the basis of the high-rise building monitoring data, three prediction models including the BP, RBF and GRNN neural network prediction models were established, the comparative analysis for the prediction accuracy of the three models was obtained. The results show that neural network is capable for prediction, and GRNN possess higher capability in prediction and better adaptability in comparing with other two neural networks.

DYNAMIC OUTPUT FEEDBACK STABILIZATION FOR NONLINEAR SYSTEMS BASED ON STANDARD NEURAL NETWORK MODELS

2006 ◽  
Vol 16 (04) ◽  
pp. 305-317 ◽  
Author(s):  
MEIQIN LIU
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Nonlinear Systems ◽  
Network Model ◽  
Neural Network Model ◽  
Output Feedback ◽  
Control Design ◽  
Feedback Stabilization ◽  
Dynamic Output Feedback ◽  
Dynamic Output

A neural-model-based control design for some nonlinear systems is addressed. The design approach is to approximate the nonlinear systems with neural networks of which the activation functions satisfy the sector conditions. A novel neural network model termed standard neural network model (SNNM) is advanced for describing this class of approximating neural networks. Full-order dynamic output feedback control laws are then designed for the SNNMs with inputs and outputs to stabilize the closed-loop systems. The control design equations are shown to be a set of linear matrix inequalities (LMIs) which can be easily solved by various convex optimization algorithms to determine the control signals. It is shown that most neural-network-based nonlinear systems can be transformed into input-output SNNMs to be stabilization synthesized in a unified way. Finally, some application examples are presented to illustrate the control design procedures.

Multi-dimensional Taylor network modelling and optimal control of SISO nonlinear systems for tracking by output feedback

2019 ◽  
Vol 37 (3) ◽  
pp. 699-717 ◽  
Author(s):  
Qi-Ming Sun ◽  
Hong-Sen Yan
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Nonlinear Systems ◽  
Real Time ◽  
Output Feedback ◽  
Tracking Error ◽  
Real Time Control ◽  
Network Modelling ◽  
Single Input Single Output ◽  
Time Control

Abstract In this paper, a multi-dimensional Taylor network (MTN) output feedback tracking control of nonlinear single-input single-output (SISO) systems in discrete-time form is studied. To date, neural networks are generally used to identify unknown nonlinear systems. However, the neuron of neural networks includes the exponential function, which contributes to the complexity of calculation, making the neural network control unable to meet the real-time requirements. In order to identify the controlled object whose model is unknown, the MTN, which requires only addition and multiplication, is utilized for successful real-time control of the SISO nonlinear system based on only its output feedback. Lyapunov analysis proves that output signals in the closed-loop system remain bounded and the tracking error converges to an arbitrarily small neighbourhood around the origin. In contrast to the back propagation (BP) neural network self-adaption reconstitution controller, the edge of the scheme is that the MTN optimal controller promises desirable response speed, robustness and real-time control.

Iterative learning radial basis function neural networks control for unknown multi input multi output nonlinear systems with unknown control direction

2019 ◽  
Vol 41 (12) ◽  
pp. 3452-3467 ◽  
Author(s):  
Tarek Bensidhoum ◽  
Farah Bouakrif ◽  
Michel Zasadzinski
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Nonlinear Systems ◽  
Radial Basis Function ◽  
Basis Function ◽  
Rbf Neural Network ◽  
Iterative Learning ◽  
Unknown Control Directions ◽  
Control Scheme ◽  
Unknown Control

In this paper, an iterative learning radial basis function neural-networks (RBF NN) control algorithm is developed for a class of unknown multi input multi output (MIMO) nonlinear systems with unknown control directions. The proposed control scheme is very simple in the sense that we use just a P-type iterative learning control (ILC) updating law in which an RBF neural network term is added to approximate the unknown nonlinear function, and an adaptive law for the weights of RBF neural network is proposed. We chose the RBF NN because it has universal approximation capabilities and can approximate any continuous function. In addition, among the advantages of our controller scheme is the fact that it is applicable to deal with a class of nonlinear systems without the need to satisfy the global Lipschitz continuity condition and we assume, only, that the unstructured uncertainty is norm-bounded by an unknown function. Another advantage of the proposed controller and unlike other works on ILC, we do not need any prior knowledge of the control directions for MIMO nonlinear system. Thus, the Nussbaum-type function is used to solve the problem of unknown control directions. In order to prove the asymptotic stability of the closed-loop system, a Lyapunov-like positive definite sequence is used, which is shown to be monotonically decreasing under the control design scheme. Finally, an illustrative example is provided to demonstrate the effectiveness of the proposed control scheme.

scholarly journals Adaptive Control Using Neural Networks and Approximate Models for Nonlinear Dynamic Systems

2020 ◽  
Vol 2020 ◽  
pp. 1-13 ◽  
Author(s):  
Khadija El Hamidi ◽  
Mostafa Mjahed ◽  
Abdeljalil El Kari ◽  
Hassan Ayad
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Adaptive Control ◽  
Nonlinear Systems ◽  
Moving Average ◽  
Approximate Model ◽  
Autoregressive Moving Average ◽  
Reference Trajectory ◽  
Set Up ◽  
Nonlinear Autoregressive

In this research, a comparative study of two recurrent neural networks, nonlinear autoregressive with exogenous input (NARX) neural network and nonlinear autoregressive moving average (NARMA-L2), and a feedforward neural network (FFNN) is performed for their ability to provide adaptive control of nonlinear systems. Three dynamical nonlinear systems of different complexity are considered. The aim of this work is to make the output of the plant follow the desired reference trajectory. The problem becomes more challenging when the dynamics of the plants are assumed to be unknown, and to tackle this problem, a multilayer neural network-based approximate model is set up which will work in parallel to the plant and the control scheme. The network parameters are updated using the dynamic backpropagation (BP) algorithm.

Sign in / Sign up

Export Citation Format

Share Document