Affine Radial Basis Function Neural Networks for the Robust Control of Hyperbolic Distributed Parameter Systems

Authors: Eleni Aggelogiannaki, Haralambos Sarimveis

Abstract:

In this work, a radial basis function (RBF) neural network is developed for the identification of hyperbolic distributed parameter systems (DPSs). This empirical model is based only on process input-output data and used for the estimation of the controlled variables at specific locations, without the need of online solution of partial differential equations (PDEs). The nonlinear model that is obtained is suitably transformed to a nonlinear state space formulation that also takes into account the model mismatch. A stable robust control law is implemented for the attenuation of external disturbances. The proposed identification and control methodology is applied on a long duct, a common component of thermal systems, for a flow based control of temperature distribution. The closed loop performance is significantly improved in comparison to existing control methodologies.

Keywords: Hyperbolic Distributed Parameter Systems, Radial Basis Function Neural Networks, H∞ control, Thermal systems.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1075376

References:

[1] E.M. Hanczyc and A. Palazoglu, "Nonlinear Control of a Distributed Parameter Process: The Case of Multiple Characteristics," Ind. Eng. Chem. Res., vol. 34, pp. 4406-4412, 1995.
[2] E.M. Hanczyc and A. Palazoglu, "Sliding Mode Control of Nonlinear Distributed Parameter Chemical Processes," Ind. Eng. Chem. Res., vol. 34, pp. 557-566, 1995.
[3] P. Christofides and P. Daoutides, "Feedback control of hyperbolic PDE systems," AIChe J., vol. 42, no.11, pp. 3063-3086, 1996.
[4] P. Christofides and P. Daoutides, "Robust control of hyperbolic PDE systems," Chemical Engineering Science, vol. 53, no. 1, pp. 3063-3086, 1998.
[5] P. Christofides, Nonlinear and Robust Control of PDE Systems: Methods and Applications to transport reaction processes. Boston: Birkhäuser, 2001.
[6] S. Alotaibi, M. Sen, B. Goodwine and K.T. Yang, "Flow based control of temperature in long ducts," International Journal of Heat and Mass Transfer, vol. 47, pp. 4995-5009, 2004.
[7] H. Shang, J.F. Forbes and M. Guay, "Feedback control of hyperbolic distributed parameter systems," Chemical Engineering Science, vol. 60, pp. 969 - 980, 2005.
[8] I. Karafyllis and P. Daoutidis, "Control of hot spots in plug flow reactors," Computers & Chemical Engineering, vol. 26, no. 7-8, pp. 1087-1094, 2002.
[9] A.A. Patwardhan, G.T. Wright and T.F. Edgar, "Nonlinear model predictive control of distributed parameter systems," Chemical Engineering Science, vol. 47, no. 4, pp. 721-735, 1992.
[10] S. Dubljevic, P. Mhaskar, N.H. El-Farra and P. Christofides, "Predictive control of transport-reaction processes," Computers and Chemical Engineering, vol. 29, pp. 2335-2345, 2005.
[11] H. Shang, J.F. Forbes and M. Guay, "Model predictive control of Quasilinear Hyperbolic Distributed Parameter Systems," Ind. Eng. Chem. Res., vol. 43, pp. 2140-2149, 2004.
[12] R. Gonzáles-García, R. Rico-Martínez, I. Kevrekidis, "Identification of distributed parameter systems: A neural net based approach," Comp. Chem. Eng., vol. 22, pp. 965-968, 1998.
[13] R. Padh, S. Balakrishnan and T. Randolph, "Adaptive-critic based optimal neuro control synthesis for distributed parameter systems," Automatica, vol. 37, pp. 1223-1234, 2001.
[14] E. Aggelogiannaki, H. Sarimveis, "Model predictive control for distributed parameter systems using RBF neural networks," in: Proc. 2nd International Conference on Informatics in Control, Automation and Robotics, Barcelona, 2005.
[15] L. Magni, G. De Nicolao, R. Scattolini, F. Allgöwer, "Robust model predictive control for nonlinear discrete-time systems," International Journal of Robust and Nonlinear Control, vol. 13, pp. 229-246, 2003.
[16] W. Lin and C.I. Byrnes, "Discrete-time Nonlinear H∞ Control with Measurement Feedback", Automatica, vol. 31, pp. 419-434, 1995.
[17] W. Lin and L. Xie, "A link between H∞ control of a dicrete time nonlinear system and its linearization", International Journal of control, vol. 69, no. 2, pp. 301-314, 1998.
[18] H. Sarimveis, A. Alexandridis, G. Tsekouras and G. Bafas, "A fast and efficient algorithm for training radial basis function neural networks based on a fuzzy partition of the input space," Industrial Engineering Chemistry Research, vol. 41, pp. 751-759, 2002.