Design and Development of Fractional Order Convolutional Neural Network Based Fractional Order Nonlinear Reactor Power Simulator for CANDU-PHWR
DOI:
https://doi.org/10.53560/PPASA(61-2)832Keywords:
Fractional Order, Convolutional Neural Network, Nonlinear Control, MIMO System, Visual Basic, Simulator, Reactor Regulating System, CANDU-PHWRAbstract
A highly complex nonlinear Reactor Regulating System (RRS) of Canadian Deuterium Uranium Pressurized Heavy Water Reactor (CANDU-PHWR) based Nuclear Power Plant (NPP) simulated in the present research. The internal design of RRS is secured and vendor controlled which is embedded in AC-132 Programmable Logic Controller (PLC). Therefore, the problem of the identification of the RRS controller model is addressed. A data-driven Fractional Order Nonlinear MIMO Hammerstein Model (FO-NC-MIMO-HM) of NPP is identified using an Adaptive Immune Algorithm (AIA) based on a Global Search Strategy (GSS) and Auxiliary Model Recursive Least Square Method (AMRLSM). Parameters of FO-MIMO-HM are identified using Innovative Real-Time Plant Operational Data (IRTPOD). The original PLC-based controller is replaced with a new Fractional Order Convolutional Neural Network (FO-CNN) based Fractional Order Nonlinear Controller (FO-NC). Therefore, a visual Simulator is developed for detailed modeling, control, simulation, and analysis of the proposed design scheme for RRS in Visual Basic (VB) Software. The performance of the proposed design scheme is tested and validated for different modes of RRS against benchmark data obtained from Plant Data Recorder (PDR) and found in close agreement well within the design bounds.
References
D. Feng, S. Yang, and C. Tongwen. Auxiliary least squares identification methods for Hammerstein output error systems. Systems and Control Letters 56: 373-380 (2007).
L. Qinyao, and D. Feng. Auxiliary model based recursive generalized least squares algorithm for multivariate output error autoregressive systems using the data filtering. Circuits Systems and Signal Processing 38: 590-610 (2019).
Y. Dai, Y. Li, L. Wei, J. Wang, and D. Deling. Adaptive immune genetic algorithm for global optimization to multivariable function. Journal of Systems Engineering and Electronics 18(03): 655-660 (2007).
L. Qiuzhen, M. Yueping, C. Jianyong, Z. Qingling, C. A. Carlo, W. Ka-Chun, and C. Fei. An adaptive immune inspired multi-objective algorithm with multiple differential evolution strategies. Information Sciences 431: 46-64 (2018).
L. Zeng, Z. Zhuting, L. Shu, P. Cheng, and W. Yong. Subspace identification for fractional order Hammerstein systems based on instrumental variables. International Journal of Control, Automation and Systems 10(5): 947-953 (2012).
S. Oliver, K. Philipp, and H. Soren. Iterative method for online fractional order and parameter identification. IEEE Conference on Decision and Control, Republic of Korea 14-18 December, pp. 5159-5166 (2020).
Y. Zhao, Y. Li, and Y.Q. Chen. Complete parameter identification of fractional order Hammerstein systems. ICFDA’14 International Conference on Fractional Differentiation and Its Application, Catania, Italy, 23-25 June 2014 pp. 1-6 (2014).
M.R. Rahmani, and M. Farrokhi. Fractional order Hammerstein state space modeling of nonlinear dynamic systems from input-output measurements. ISA Transactions 96: 177-184 (2020).
W. Allafi, I. Zajic, K. Uddin, and K.J. Burnham. Parameter estimation of the fractional order Hammerstein-Wiener model using simplified refined instrumental variable fractional order continuous time. IET Control Theory and Applications 11(15): 2591-2598 (2017).
Q. Jin, Y. Ye, W. Cai, and Z. Wang. Recursive identification for fractional order Hammerstein model based on ADELS. Mathematical Problems in Engineering 2021: 6629820 (2021).
J.Xu, Z. Xiong, and S.P. Bhattacharyya. Real time semantic segmentation network inspired by PID controllers. Computer Vision and Pattern Recognition, Cornell University, New York, 4 June 2022 pp. 1-11 (2022).
R. Farkh, M.T. Quasim, K.A. Jaloud, S. Alhuwaimel, and S.T. Siqqiqui. Computer vision-control-based CNN-PID for mobile robot. Computers, Materials and Continua 68(1): 1065-1079 (2021).
W. An, H. Wang, Q. Sun, J. Xu, Q. Dai, and L. Zhang. A PID controller approach for stochastic optimization of deep networks. IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 18-23 June 2018 pp. 1-6 (2018).
M.M. Taresh, N. Zhu, T.A. Ali, and W. Guo. Using a novel fractional order gradient method for CNN back propagation. Computer Vision and Pattern Recognition, Cornell University, New York, 1 May 2022 pp. 1-10 (2022).
C.A. Calderon. Realization and comparative analysis of fractional order controllers for different discretization methods. IEEE International Conference on Automaton, Curicó, Chile, 24-28 October 2022 pp. 5159-5166 (2022).
T. Kan, Z. Gao, C. Yang and J. Jian. Convolutional neural networks based on fractional order momentum for parameter training. Neurocomputing 449: 85-99 (2021).
P.V. Surjagade, J. Deng, V. Vajpaeyee, V.M. Becerra, S.R. Shimjith, and A.J. Arul. Fractional order integral sliding mode control for PWR nuclear power plant. European Control Conference, London, United Kingdom, 12-15 July 2022 pp. 987-992 (2022).
