Journal ArticleDOI
Neural networks for nonlinear internal model control
Kenneth J. Hunt,D. Sbarbaro +1 more
- Vol. 138, Iss: 5, pp 431-438
Reads0
Chats0
TLDR
In this paper, a novel technique, directly using artificial neural networks, is proposed for the adaptive control of nonlinear systems, where the ability of neural networks to model arbitrary nonlinear functions and their inverses is exploited.Abstract:
A novel technique, directly using artificial neural networks, is proposed for the adaptive control of nonlinear systems. The ability of neural networks to model arbitrary nonlinear functions and their inverses is exploited. The use of nonlinear function inverses raises questions of the existence of the inverse operators. These are investigated and results are given characterising the invertibility of a class of nonlinear dynamical systems. The control structure used is internal model control. It is used to directly incorporate networks modelling the plant and its inverse within the control strategy. The potential of the proposed method is demonstrated by an example.read more
Citations
More filters
Proceedings ArticleDOI
Inversion of recurrent neural networks for control of non-linear systems
C. Kambhampati,R. Craddock +1 more
TL;DR: Using the theoretical results, it is shown how an inverse controller can be produced from a neural network model of the plant, without the need to train an additional network to perform the inverse control.
Book ChapterDOI
An On-Line LearningAlgorithm of Parallel Mode for MLPN Models
TL;DR: An on-line learning algorithm in parallel mode for multi-layer perceptron network (MLPN) model is proposed, which is able to learn the non-linear dynamic behaviour of unknown time-varying systems and perform multi-step-ahead prediction for control purpose.
Proceedings ArticleDOI
Comparisons between classical and neural schemes of process control: applications to a continuous reactor
R. Lamanna,Y. Alcocer,E. Samper +2 more
TL;DR: In this paper, the authors use feed-forward neural networks to implement some control schemes on a continuous stirred-tank reactor (CSTR), in order to compare their behaviour with that of a classical PI controller.
Posted Content
Recurrent neural network-based Internal Model Control of unknown nonlinear stable systems.
TL;DR: In this article, a first gated RNN is used to learn a model of the unknown input-output stable plant, and then, another gated RLNN approximating the model inverse is trained.
Journal ArticleDOI
Parâmetro de Exatidão para Aproximacão de Funcões Utilizando Multilayer Perceptrons nos Domínios Real, Complexo e de Clifford
TL;DR: The authors propose a metodologia for abordagem de dois pontos de interesse: 1) estimar a parâmetro de exatidão for as saídas of the Redes Neurais Artificiais (RNA) Multilayer Perceptrons, with the goal of determining the confiáveis of the SAIs.
References
More filters
Journal ArticleDOI
Approximation by superpositions of a sigmoidal function
TL;DR: It is demonstrated that finite linear combinations of compositions of a fixed, univariate function and a set of affine functionals can uniformly approximate any continuous function ofn real variables with support in the unit hypercube.
Book
Feedback Systems: Input-output Properties
TL;DR: In this paper, the Bellman-Gronwall Lemma has been applied to the small gain theorem in the context of linear systems and convolutional neural networks, and it has been shown that it can be applied to linear systems.
Book
Robust process control
TL;DR: A state-of-the-art study of computerized control of chemical processes used in industry is presented in this article for chemical engineering and industrial chemistry students involved in learning the micro-macro design of chemical process systems.
Journal ArticleDOI
Internal model control. A unifying review and some new results
Carlos E. García,Manfred Morari +1 more
Journal ArticleDOI
A multilayered neural network controller
TL;DR: A modified error-back propagation algorithm, based on propagation of the output error through the plant, is introduced, for learning several learning architectures for training the neural controller to provide the appropriate inputs to the plant.