Engineering with Computers

, Volume 13, Issue 1, pp 1–19 | Cite as

Upgrading automation for nuclear fuel in-core management: From the symbolic generation of configurations, to the neural adaptation of heuristics

  • Ephraim Nissan
  • Hava Siegelmann
  • Alex Galperin
  • Shuky Kimhi


FUELCON is an expert system in nuclear engineering. Its task is optimized refueling-design, which is crucial to keep down operation costs at a plant. FUELCON proposes sets of alternative configurations of fuel-allocation; the fuel is positioned in a grid representing the core of a reactor. The practitioner of in-core fuel management uses FUELCON to generate a reasonably good configuration for the situation at hand. The domain expert, on the other hand, resorts to the system to test heuristics and discover new ones, for the task described above. Expert use involves a manual phase of revising the ruleset, based on performance during previous iterations in the same session. This paper is concerned with a new phase: the design of a neural component to carry out the revision automatically. Such an automated revision considers previous performance of the system and uses it for adaptation and learning better rules. The neural component is based on a particular schema for a symbolic to recurrent-analogue bridge, called NIPPL, and on the reinforcement learning of neural networks for the adaptation.


Allocation Design Downtime Expert systems Machine learning Neural networks Nuclear engineering in-core fuel management Refueling Reload 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Galperin, A.; Kimhi, Y.; Nissan, E. (1993). FUELCON: an expert system for assisting the practice and research of in-core fuel management and optimal design in nuclear engineering, Computers and Artificial Intelligence, 12, 369–415Google Scholar
  2. 2.
    Kimhi, Y. (1992) A non-algorithmic approach to the in-core fuel management problem of a PWR core, PhD dissertation, Department of Nuclear Engineering, Ben-Gurion University of the Negev (Beer-Sheva, Israel) (in Hebrew; English abstract)Google Scholar
  3. 3.
    Galperin, A.; Kimhi, Y. (1991) Application of knowledge-based methods to in-core fuel management. Nuclear Science and Engineering, 109, 103–110Google Scholar
  4. 4.
    Galperin, A.; Yimhi, Y.; Segev, M. (1989) A knowledge-based system for optimization of fuel reload configurations, Nuclear Science and Engineering, 102, 43Google Scholar
  5. 5.
    Galperin, A.; Nissan, E. (1988) Application of a heuristic search method for generation of fuel reload configurations, Nuclear Science and Engineering, 99 (4), 343–352Google Scholar
  6. 6.
    Siegelmann, H.T. (1994) Neural programming language, Proceedings of the 12th National Conference on Artificial Intelligence, Seattle, Washington.Google Scholar
  7. 7.
    Cochran, R.G.; Tsoulfanidis, N. (1990) The Nuclear Reactor Cycle: Analysis and Management, American Nuclear Society, La Grange Park, ILGoogle Scholar
  8. 8.
    Parks, G.T.; Lewins, J.D. (1992) In-core fuel management and optimization: the state of the art, Nuclear Europe Worldscan, 12 (3/4), 41Google Scholar
  9. 9.
    Petschhat, G.R.; Rothleder, B.M.; Faught, W.S.; Eich, W.J. (1986) Interactive fuel shuffle assistant graphics interface and automation for nuclear fuel shuffle with PDQ7, Proceedings of the American Nuclear Society Topical Meeting on Advances in Fuel Management, Pinehurst, NCGoogle Scholar
  10. 10.
    Faught, W.S. (1987) Prototype fuel shuffling system using a knowledge-based toolkit, Technical Report, IntelliCorp, Mountain View, CAGoogle Scholar
  11. 11.
    Rothleder, B.M.; Petschhat, G.R.; Faught, W.S.; Eich, W.J. (1988) The potential for expert system support in solving the Pressurized Water Reactor fuel shuffling problem, Nuclear Science and Engineering, 100, 400Google Scholar
  12. 12.
    Kropaczek, D.J.; Turinski, P.J. (1991) In-core nuclear fuel optimization for Pressurized Water Reactors using simulated annealing. Nuclear Technology, 95, 9Google Scholar
  13. 13.
    Sontag, E.D. (1992) Neural nets as systems models and controllers, Proceedings of the Seventh Yale Workshop on Adaptive and Learning Systems, Yale University, New Haven, CN, pp. 73–79Google Scholar
  14. 14.
    Matthews, M. (1992) On the uniform approximation of nonlinear discrete-time fading-memory systems using neural network models, PhD dissertation, ETH No. 9635, E.T.H., ZurichGoogle Scholar
  15. 15.
    Polycarpou, M.M.; Ioannou, P.A. (1991) Identification and control of nonlinear systems using neural network models: design and stability analysis, Report 91-09-01, Department of EE/Systems, University of South California, Los AngelesGoogle Scholar
  16. 16.
    Cleeremans, A.; Servan-Schreiber, D.; McClelland, J. (1989) Finite state automata and simple recurrent networks, Neural Computation, 1, 372Google Scholar
  17. 17.
    Elman, J. L. (1990) Finding structure in time, Cognitive Science, 14, 179–211Google Scholar
  18. 18.
    Giles, C.L.; Miller, C.B.; Chen, D.; Chen, H.H.; Sun, G.Z.; Lee, Y.C. (1992) Learning and extracting finite state automata with second-order recurrent neural networks,Neural Computation, 4 (3), 393–405Google Scholar
  19. 19.
    Pollack, J.B. (1990) The induction of dynamical recognizers, Report 90-JP-Automata, Department of Computer and Information Science, Ohio State UniversityGoogle Scholar
  20. 20.
    Williams, R.J.; Zipser, D. (1989) A learning algorithm for continually running fully recurrent neural networks, Neural Computation, 1, 270–280Google Scholar
  21. 21.
    Barto, A. G.; Singh, S. P. (1990) On the computational economics of reinforcement learning, Proceedings of the 1990 Connectionist Models Summer School, D. S. Touretzky, J. L. Elman, T. I. Sejnowski, and G. E. Hinton (Editors), Morgan Kaufmann, San Mateo, CAGoogle Scholar
  22. 22.
    Narendra, K.; Thathacar, M. A. L. (1989)Learning Automata: An Introduction, Prentice-Hallm, Engelwood Cliffs, NJGoogle Scholar
  23. 23.
    Barto, A. G.; Anandan, P. (1985) Pattern recognizing stochastic learning Aatomata,IEEE Transactions on System, Man, and Cybernetics,15, 360–375Google Scholar
  24. 24.
    Barto, A.G.; Jordan, M.I. (1987) Gradient following without back-propagation in layered networks, Proceedings of the First IEEE International Conference on Neural Networks, San Diego, Vol. 2, pp. 629–636Google Scholar
  25. 25.
    Hertz, J.; Krogh, A.; Palmer, R.G. (1991) Introduction to the Theory of Neural Computation, Addison-Wesley, Reading, MAGoogle Scholar
  26. 26.
    Barto, A.G.; Sutton, R.S.; Watkins, C.J.C.H. (1991) Learning and sequential decision making, In Learning and Computational Neuroscience, M. Gabriel and J.W. Moore (Editors), MIT Press, Cambridge, MAGoogle Scholar
  27. 27.
    Jordan, M.I. (1992) Forward models: supervised learning with a distal teacher, Cognitive Science, 16, 307–354Google Scholar
  28. 28.
    Siegelmann, H.T. (1993) Foundations of recurrent neural networks, PhD Dissertation, Rutgers University, New Brunswick, New JerseyGoogle Scholar
  29. 29.
    Gallant, S.I. (1988) Connectionist expert systems, Communications of the ACM, 31 (2), 152–169Google Scholar
  30. 30.
    Medsker, L.R. Editor (1991) The Synergism of Expert System and Neural Network Technologies, special issue of Expert Systems with Applications, 2 (1)Google Scholar
  31. 31.
    Medsker, L.R. (1994) Hybrid Neural Network and Expert Systems, Kluwer, Dordrecht, The NetherlandsGoogle Scholar
  32. 32.
    Bhogal, A.S.; Seviora, R.E.; Elmasry, M.I. (1991) Towards connectionist expert systems, Expert Systems with Applications, 2 (1), 3–14Google Scholar
  33. 33.
    Silverman, B.G. (1992) Survey of expert critiquing systems: practical and theoretical frontiers, Communications of the ACM, 35 (4), 106–127Google Scholar
  34. 34.
    Towell, G.G.; Shavlik, J.; Noordewier, M.O. (1990) Refinement of approximately correct domain theories by knowledge-based neural networks, Proceedings of the Eighth National Conference on Artificaial Intelligence, p. 861Google Scholar
  35. 35.
    Maclin, R.; Shavlik, J.W. Incorporating advice into agents that learn from reinforcement (journal submission)Google Scholar
  36. 36.
    Nissan, E.; Siegelmann, H.; Galperin, A.; Kimhi, S. (1994) Towards full automation of the discovery of heuristics in a nuclear engineering project, by combining symbolic and subsymbolic Computation, Proceedings of the Eighth International Symposium on Methodologies for Intelligent Systems (ISMIS'94), Charlotte, N.C., Springer-Verlag, New York (Lecture Noter in Artificial Intelligence, Vol. 869), pp. 427–436Google Scholar
  37. 37.
    Nissan, E.; Siegelmann, H.; Galperin, A. (1994) An integrated symbolic and neural network architecture for machine learning in the domain of nuclear engineering, Proceedings of the Conference on Pattern Recognition and Neural Networks, within the 12th ICPR: International Conferences on Pattern Recognition, Jerusalem, IEEE, Computer Society Press, New YorkGoogle Scholar
  38. 38.
    Galperin, A.; Kimhi, S.; Nissan, E.; Siegelmann, H.; Zhao, J. (1995)Symbolic and subsymbolic integration in prediction and rule-revision tasks for fuel allocation in nuclear reactors. Proceedings of the Third European Congress on Intelligent Techniques and Soft Computing (EUFIT'95), Aachen, Germany, (in press)Google Scholar

Copyright information

© Springer-Verlag London Limited 1997

Authors and Affiliations

  • Ephraim Nissan
    • 1
  • Hava Siegelmann
    • 1
  • Alex Galperin
    • 2
  • Shuky Kimhi
    • 2
    • 3
  1. 1.Department of Mathematics and Computer ScienceBar-Ilan UniversityRamat-GanIsrael
  2. 2.Department of Nuclear EngineeringBen-Gurion University of the NegevBeer-ShevaIsrael
  3. 3.Kernforschungszentrum KarlsruheInstitut für Neutronenphysik und ReaktortechnikKarlsruheGermany

Personalised recommendations