By Topic

IEE Colloquium on Advances in Neural Networks for Control and Systems

25-27 May 1994

Filter Results

Displaying Results 1 - 18 of 18
  • Adaptive neurocontrol of MIMO systems based on stability theory

    Publication Year: 1994, Page(s):13/1 - 13/5
    Cited by:  Papers (2)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (353 KB)

    In this paper we prove the stability of a certain class of nonlinear discrete MIMO systems controlled by a multilayer neural net with a simple weight adaptation strategy. The proof is based on the Lyapunov formalism. The stability statement is, however, only valid if the initial weight values are not too far from their optimal values that allow perfect model matching. We therefore propose to initi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Exothermic heat estimation using fuzzy neural nets for a batch reactor temperature control system

    Publication Year: 1994, Page(s):9/1 - 9/4
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (180 KB)

    Batch processes may constantly be changing, sometimes in unpredictable ways. For this reason, the operators need to keep a close watch on things. Batch-plant control stations have traditionally been located very close to the process, because operators can do a better job of sensing abnormalities if they can see, hear and smell what is going on. The reaction process can be considered as unstable, t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning in neural networks and stochastic approximation methods with averaging

    Publication Year: 1994, Page(s):14/1 - 14/4
    Cited by:  Papers (13)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (240 KB)

    The problem of adjusting the weights (learning) in multilayer feedforward neural networks (NN) is known to be of a high importance when utilizing NN techniques in various practical applications. The learning procedure is to be performed as fast as possible and in a simple computational fashion, the two requirements which are usually not satisfied practically by the methods developed so far. Moreov... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Equalisation using non-linear adaptive clustering

    Publication Year: 1994, Page(s):17/1 - 17/3
    Cited by:  Patents (3)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (128 KB)

    The optimal equalisation of communications channels is achieved by the use of maximum likelihood sequence estimation (MLSE). Nonlinear clustering procedures extend the capabilities of MLSE in the case of channel nonlinearity or time variation. However, a problem with these techniques has been either high computational complexity or uncertainty about model order. This paper provides a technique whi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On interpolation memories for learning control

    Publication Year: 1994, Page(s):10/1 - 10/3
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (152 KB)

    The paper deals with new developments of interpolating memories as the basic element of learning control and with their possible application. It discusses learning control, interpolating memories, characteristic manifolds for automotive control, and possible future developments View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive neurofuzzy systems for difficult modelling and control problems

    Publication Year: 1994, Page(s):15/1 - 15/3
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (168 KB)

    The ability to learn is the cornerstone of current neural technology. It was responsible for their decline in the late sixties when it was shown that the perceptron training algorithm could not be easily extended to multilayered networks, and also the revival of interest in these techniques was initiated by the discovery of a multilayer network adaptation rule. Artificial neural networks attempt t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Semi-empirical modeling of non-linear dynamical systems

    Publication Year: 1994, Page(s):4/1 - 4/3
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (180 KB)

    The author is concerned with modeling problems where there are some empirical data, and some limited system knowledge available. In that case, first principles modeling may lead to an inaccurate model. Using the black-box alternative, not much of the system knowledge can be incorporated a priori. Hence, both these approaches have serious drawbacks in some cases. The proposed approach is based on t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hierarchical competitive net architecture

    Publication Year: 1994, Page(s):18/1 - 18/3
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (164 KB)

    Hypersonic aircraft require a high degree of system integration. Design tools are needed that can provide rapid, accurate calculations of complex fluid flow. Existing methods are slow. The goal of this project was to apply neural networks to the calculation of fluid flow and heat transfer in a heat exchanger panel for the National AeroSpace Plane (NASP) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Comparison of gradient based training algorithms for multilayer perceptrons

    Publication Year: 1994, Page(s):11/1 - 11/6
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (224 KB)

    The training speed of batch backpropagation using steepest descent, conjugate gradient and quasi-Newton algorithm for a feedforward neural network are compared. Results illustrating the advantages of the Hessian based techniques are given and issues affecting speed discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Data analysis by means of Kohonen feature maps for load forecast in power systems

    Publication Year: 1994, Page(s):6/1 - 6/4
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (268 KB)

    Because of the clustering and dimensionality reduction abilities the Kohonen feature map (KFM) can be the preferable tool for deriving knowledge about dependencies of the load consumption in electrical energy systems (EES). This paper describes the application of the KFM for analysing and splitting extensive load databases. The objective is to get separate clusters of load shapes for making short ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive neural network control of the temperature in an oven

    Publication Year: 1994, Page(s):8/1 - 8/3
    Cited by:  Papers (2)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (172 KB)

    Many authors have shown by simulated studies, that a great number of non-linear dynamical systems could be identified and controlled by using neural network models. The authors applied these results to a real process : an oven with two inputs, one for heating and one for cooling. The output to be controlled is the temperature inside the oven. The choice of the control strategy on the one hand and ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The TACOMA learning architecture for reflective growing of neural networks

    Publication Year: 1994, Page(s):2/1 - 2/2
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (132 KB)

    One of the important problems to be solved for neural network applications is to find a suitable network structure solving the given task. To reduce the engineering efforts for the architecture design a data driven algorithm is desirable which constructs a network structure during the learning process. There are different approaches for structure adaptation with evolutionary algorithms, growth alg... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Constructive learning-industrial perspectives

    Publication Year: 1994, Page(s):16/1 - 16/4
    Cited by:  Papers (7)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (220 KB)

    The learning algorithms and structures which have become popular in the neural network community over the last few years have been successfully applied to various challenging modelling problems. In contrast to the linear modelling methods, there is still no clearly defined engineering process from which a model can reliably be created from measurements taken from a physical system. Problems for th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural networks for control of industrial processes

    Publication Year: 1994, Page(s):5/1 - 5/3
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (128 KB)

    In this contribution, neural concepts and methods for control an their use in industrial applications are discussed and illustrated. Even though there exist numerous conventional approaches for solving control tasks, their realization in practice frequently proves to be very difficult. The author pursues several approaches to using neural networks in the context of nonlinear control tasks. In iden... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improved prediction of the corrosion behaviour of car body steel using a Kohonen self organising map

    Publication Year: 1994, Page(s):7/1 - 7/3
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (164 KB)

    In a highly competitive market, an optimised processing plays a key role for the future of our industry. Intelligent processing is only feasible, when the materials and products are controlled directly during and after processing. Traditional methods for controlling corrosion are often very time consuming and destructive (e.g. corrosion under paint, delamination of paints on metals), Therefore the... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hierarchical mixtures of experts and the EM algorithm

    Publication Year: 1994, Page(s):1/1 - 1/3
    Cited by:  Papers (3)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (120 KB)

    The problem of training a mixture of experts architecture can be treated as a maximum likelihood estimation problem. Both Jacobs et al. (1991) and Jordan and Jacobs (1992) derived learning algorithms by computing the gradient of the log likelihood for their respective architectures. Empirical tests revealed that although the gradient approach succeeded in finding reasonable parameter values in par... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • ASMOD (Adaptive Spline Modelling of Observation Data): some theoretical and experimental results

    Publication Year: 1994, Page(s):3/1 - 3/7
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (408 KB)

    The ASMOD algorithm uses B-splines for identifying and representing nonlinear dependencies in multidimensional observation data. By automatically adjusting the internal structure of the model to the dependencies identified in the data, the model can be made flexible enough to model general and coupled nonlinear dependencies. At the same time the flexibility of the model with respect to other not y... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IEE Colloquium on `Advances in Neural Networks for Control and Systems' (Digest No.1994/136)

    Publication Year: 1994
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (72 KB)

    The following topics were dealt with: constructive learning algorithms; industrial applications; training methods and analysis; and related technologies View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.