Development of New Model-Based Adaptive Predictive Control Algorithms and Their Implementation on Real-Time Embedded Systems (original) (raw)

This dissertation deals with the development of two new neural network-based model identification algorithms and two new model predictive control (MPC) algorithms which are combined to form model-based adaptive control strategies. Also, two new computer platforms for the implementation of these algorithms and their corresponding strategies are proposed. The overall strategies consist of an online model identification part and a model-based predictive control part. The proposed algorithms for the model identification are practically two new algorithms for training a dynamic feedforward neural network (DFNN) which will be considered to comprise the model of a nonlinear dynamic system. The proposed training algorithms are: the adaptive recursive least squares (ARLS) algorithm and the modified Levenberg-Marquardt algorithm (MLMA). The proposed algorithms for the predictive control parts are: the adaptive generalized predictive control (AGPC) and the nonlinear adaptive model predictive control (NAMPC). The two real-time platforms for the implementation of the combined operation of identification and predictive control algorithms with the purpose of forming an adaptive control strategy are: a service-oriented architecture (SOA) cluster network based on the device profile for web services (DPWS) and a Xilinx's Virtex-5 FPGA (field programmable gate array) development board. The proposed control strategies have been applied to control three applications, namely: the fluidized bed furnace reactor (FBFR) of the steam deactivation unit (SDU) used for preparing catalyst for evaluation in a fluid catalytic cracking (FCC) pilot plant; an activated sludge wastewater treatment plant (AS-WWTP) in accordance with the European wastewater treatment standards; and the auto-pilot control unit of a nonlinear F-16 fighter aircraft. The neural network models for these three applications were validated using one-step, five-step and tenstep ahead prediction simulations as well as with the Akaike's final prediction error (AFPE) estimate algorithm. Also, the performances of the proposed ARLS and MLMA algorithms were compared with the backprogation with momentum (BPM) and incremental backpropagation (INCBP) algorithms. Also the performances of the PID control of the identified model of the FBFR process by means of the ARLS and the MLMA network training algorithms versus the PID control of the first principles model of the same process. The AGPC and NAMPC control of the considered applications when model identification is performed by the ARLS and MLMA algorithms were implemented on a conventional mulitcore personal computer (PC) and SOA cluster of muticore PCs. Hardware-in-the-loop simulations have been performed by linking the PC or SOA implementations with MATLAB simulations of the processes. An AGPC implementation with neural networks trained using the MLMA algorithm has been made on a Xilinx Virtex-5 FPGA. The hardware-in-the-loop simulations have shown that the proposed algorithms and their SOA or FPGA implementations can have execution times shorter than other algorithms which present similar performance. Therefore, they render themselves more appropriate compared to other algorithms for use in the control of processes requiring shorter sampling time for stable operations. Acknowledgement ii ACKNOWLEDGEMENT My sincere appreciation and gratitude goes to my project supervisor, Professor George Hassapis, who conceived and supervised the work contained in this dissertation. I also thank him for his technical and financial supports, encouragements and fatherly roles throughout the course of this work. I will always remain grateful to him for his advice, suggestions, intuitive comments, patience and untiring efforts in reading through my manuscripts with necessary corrections from conception through algorithm developments, problem formulations, implementations, and several simulations and analyses which have resulted in this dissertation. I also thank Professor Alkiviadis Hatzopoulos and Associate Professor Loukas Petrou for their co-supervisory roles in this work. My sincere thanks to Associate Professor Loukas Petrou for his technical supports and comments as well as his efforts and time devoted to this work from inception to completion. I specially acknowledge and thank the Greek State Scholarships' Foundation (I.K.Y.) that provided the scholarship as well as the major funding for this research. I also thank the Federal Government of Nigeria for their financial support towards the Bilateral Educational Agreement with I.K.Y. and the Federal University of Technology, Akure-Nigeria for their financial supports which has made this scholarship a reality leading to the successful completion of my doctorate degree programme. My acknowledgment also goes to Ambassador of Nigeria to Greece, His Excellency (Dr.) Etim U. Uyie for his love, care and financial assistance. My special thanks go to the Staff of the School of Electrical and Computer Engineering, AUTH, Greece. I gratefully acknowledge Dr. Simeonidis Andreas for his comments and encouragements, and to Mr. George Voukalis for his technical assistance. I also wish to thank my colleagues at the Laboratory of Computer Systems Architecture: Maria Koukourli, Ioakeim Samaras, Babis Serenis, Manos Tsardoulias and Nikos Sismanis for their technical support, comments and contributions towards the successful completion of this project. I am highly indebted to my mother Mrs. Cecilia Andrew Akpan; my mother-in-law Mrs. Titilayo Nathaniel Oyewo, and my siblings Justine, Sylvester, Emmanuel and Justina for their sacrifices and prayers. Words are not enough to thank my wife, Mrs. Rachael Oyenike Vincent-Akpan, for all her sacrifices, financial support, prayers, and encouragements throughout the period of this study. Just know that I love you. Finally, I am most grateful to God Almighty for His infinite mercy, divine grace and sound health.