Application of Neural Networks in Reactor Diagnostics and Monitoring

  • I. Pázsit
  • N. S. Garis
  • P. Lindén
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 38)


This chapter gives an account of the use of neural network techniques in reactor diagnostics and monitoring through a few concrete examples of successful practical applications. Diagnostic problems require the solution of a so-called inverse task, namely to determine the normal or abnormal values of some system parameters (“noise sources’), that cannot be directly measured, by observing their effect on other, measurable parameters (”reactor noise’). In the past, such inversion or unfolding techniques were possible to perform only if the direct task, i.e., calculation of the induced noise from the noise sources, could be made with a compact analytical solution. This condition hindered wide-spread practical applications. The use of artificial neural networks (ANN) presents a very powerful solution to the unfolding problem, since it does not require an analytical relationship between the cause and the reason. ANNs can be trained on simulated data. In the nuclear industry very powerful and accurate numerical methods exist to calculate process variables for operating plant, thus ANNs trained on simulated data can be used in real applications. This will be demonstrated through examples in this chapter.


Transfer Function Neutron Detector Trained Network Pressurize Water Reactor Neural Network Technique 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    I. Pâzsit and M. Kitamura, The role of neural networks in reactor diagnostics and control. Advances in Nuclear Science and Technology 24, 95–130 (1996)CrossRefGoogle Scholar
  2. 2.
    N. S. Garis, I. Pâzsit, U. Sandberg and T. Andersson, Determination of PWR control rod position by core physics and neural network methods. Nuclear Technology 123, pp 278–295 (1998)Google Scholar
  3. 3.
    I. Pâzsit, Development of Core Diagnostic Methods and their Application at Swedish BWRs and PWRs. Submitted to J. Nucl. Sci. Technol. (1999)Google Scholar
  4. 4.
    M. Edenius, K. Ekberg, B. H. Forssén and D. Knott, “CASMO-4 - A Fuel Assembly Burnup Program”, User’s Manual. Studsvik of America, Studsvik/SOA-93/1. (Draft 2 ) (1993)Google Scholar
  5. 5.
    J. A. Umbarger and A. S. Digiovine, “SIMULATE-3 - Advanced Three-Dimensional Two-Group Reactor Analysis Code”, User’s Manual. Studsvik of America, Studsvik/SOA-92/01 (1992)Google Scholar
  6. 6.
    J. Hertz, A Krogh and R. G. Palmer, Introduction to the theory of neural computing, Addison-Wesley, Reading, Massachusetts (1991)Google Scholar
  7. 7.
    S. Haykin, Neural Networks: A Comprehensive Foundation. 211d Edition, Prentice-Hall Inc., Upper Saddle River, New Jersey (1999.)Google Scholar
  8. 8.
    D. N. Fry, Experience in reactor malfunction diagnosis using on-line noise analysis. Nuclear Technology. 10, p. 273, (1971)Google Scholar
  9. 9.
    I. Pâzsit and O. Glöckler, On the neutron noise diagnostics of PWR control rod vibrations III. Application at power plant. Nucl. Sci. Eng. 99, pp. 313–328 (1988)Google Scholar
  10. 10.
    J. A. Thie, Power Reactor Noise. ANS monograph, New York (1983)Google Scholar
  11. 11.
    I. Pâzsit, N. S. Garis and O. Glöckler, On the neutron noise diagnostics of PWR control rod vibrations IV: Application of neural networks. Nucl. Sci. Eng. 124, pp. 167–177 (1996)Google Scholar
  12. 12.
    J. K-H Karlsson, A two-group dynamic transfer function for inhomogeneous reactors. To be submitted to Ann. Nucl. Energy (1999)Google Scholar
  13. 13.
    N. S. Garis and I. Pâzsit, Control rod localisation with neural networks using a complex transfer function. Prog. Nucl. Energy 34, pp. 87–98 (1998)CrossRefGoogle Scholar
  14. 14.
    P. Lindén, G. G.osshög and I.Pâzsit, FlowAct, Flow Rate Measurements in Pipes with the Pulsed Neutron Activation Method. Nuclear Technology, 124, pp. 31–51 (1998)Google Scholar
  15. 15.
    P. Lindén, A Study of Using Pulsed Neutron Activation for Accurate Measurements Water in Pipes, CTH-RF-136, Chalmers University of Technology, Göteborg, Sweden (1998)Google Scholar
  16. 16.
    J. Briesmeister (editor), MCNP - A general Monte Carlo N-particle transport code, Version 4A, LA-12625-M (1993)Google Scholar
  17. 17.
    P. Lindén and I. Pazsit, Study of the Possibility of Determining the Mass Flow • of Water from Neutron Activation Measurements with Flow Simulations and Neural Networks. Kerntechnik 63, pp 188–196 (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • I. Pázsit
    • 1
  • N. S. Garis
    • 2
  • P. Lindén
    • 1
  1. 1.Department of Reactor PhysicsChalmers University of TechnologyGöteborgSweden
  2. 2.Swedish Nuclear Power InspectorateStockholmSweden

Personalised recommendations