Adaptive control of a PWR core power using neural networks

Abstract Reactor power control is important because of safety concerns and the call for regular and appropriate operation of nuclear power plants. It seems that the load-follow operation of these plants will be unavoidable in the future. Discrepancies between the real plant and the model used in controller design for load-follow operation encourage one to use auto-tuning and/or adaptive techniques. Neural network technology shows great promise for addressing many problems in non-model-based adaptive control methods. Also, there has been a great attention to inverse control especially in the neural and fuzzy control context. Fortunately, online adaptation eliminates some limitations of inverse control and its shortcomings for real world applications. We use a neural adaptive inverse controller to control the power of a PWR reactor. The stability of the system and convergence of the controller parameters are guaranteed during online adaptation phase provided the controller is near the plant’s real inverse after offline training period. The performance of the controller is verified using nonlinear simulations in diverse operating conditions.

[1]  Hans Christian,et al.  The controller output error method / Has Christian Asminn Andersen , 1998 .

[2]  Kwang Y. Lee,et al.  State Feedback Assisted Classical Control: An Incremental Approach to Control Modernization of Existing and Future Nuclear Reactors and Power Plants , 1990 .

[3]  Nam Zin Cho,et al.  Design of a nonlinear model-based controller with adaptive PI gains for robust control of a nuclear reactor , 1992 .

[4]  B. Widrow,et al.  Adaptive inverse control , 1987, Proceedings of 8th IEEE International Symposium on Intelligent Control.

[5]  Daniel Sbarbaro,et al.  Neural Networks for Nonlinear Internal Model Control , 1991 .

[6]  T. Teichmann,et al.  Control of Nuclear Reactors and Power Plants , 1956 .

[7]  Kwang Y. Lee,et al.  Robust optimal control of nuclear reactors and power plants , 1992 .

[8]  A. Sideris,et al.  A multilayered neural network controller , 1988, IEEE Control Systems Magazine.

[9]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[10]  Ferenc Szeifert,et al.  Takagi-Sugeno fuzzy control of batch polymerization reactors , 1997, Proceedings of IEEE International Conference on Intelligent Engineering Systems.

[11]  Nam Zin Cho,et al.  Time-optimal control of nuclear reactor power with adaptive proportional-integral-feedforward gains , 1993 .

[12]  Hossein Arabalibeik,et al.  Improved temperature control of a PWR nuclear reactor using an LQG/LTR based controller , 2003 .

[13]  Chaung Lin,et al.  Neurocontrol of Pressurized Water Reactors in Load-Follow Operations , 2000 .

[14]  Kwang Y. Lee,et al.  LQG/LTR robust control of nuclear reactors with improved temperature performance , 1992 .

[15]  Bernard Widrow,et al.  Nonlinear adaptive inverse control , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[16]  P. K. Chawdhry,et al.  Soft Computing in Engineering Design and Manufacturing , 1998, Springer London.

[17]  Enrico Zio,et al.  Neural estimation of first-order sensitivity coefficients : Application to the control of a simulated pressurized water reactor , 1999 .

[18]  Ferenc Szeifert,et al.  Inverse fuzzy-process-model based direct adaptive control , 1999 .

[19]  M. Agarwal.  A systematic classification of neural-network-based control , 1997 .

[20]  Mehrdad Nouri Khajavi,et al.  A neural network controller for load following operation of nuclear reactors , 2002 .

[21]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.