الفهرس 
Chapter 1: IntroductionOnly 14 pages are availabe for public view
 |  | from | 180 |  |  |
| | | from | 180 | | |
Abstract
Hybrid dynamical systems comprise two different types of dynamics: continuous (analog) dynamics and discrete (logical) dynamics. These two dynamics not only cohabit together inside the system but also interact and affect each other. Recently, hybrid systems have got more attention in research because most systems in industries or laboratories manifest hybrid dynamic behaviors. Optimal control design for hybrid systems gives more challenges to manage the interaction between the continuous and discrete dynamics inside the system. Moreover, Cyber-Physical Systems (CPS), that play a vital role in the fourth industrial revolution (Industry 4.0), are mainly considered as hybrid systems. Therefore, hybrid dynamical systems are found in many industrial process control systems. Tanks levels control is one of the most essential and common control problems in industrial systems. Many variations with several configurations of tank systems are found in literature as a benchmark for control algorithms such as single, double, triple, and quadruple tank systems. This thesis introduces a Multi-input Multi-output (MIMO) Double-tank Hybrid System (DTHS) with a mixed-integer actuation mechanism. The proposed DTHS aims to control the liquid level of the two tanks at the same time.
The first part of this thesis introduces the non-linear model of the Double-tank System (DTS) that is applied to generate different linear representation models. Then, Proportional-Integral-Derivative (PID) and Model Predictive Control (MPC) controllers are evaluated and designed for the control of the DTS. The dynamics of two cascaded tanks with the given parameters and dimensions are modeled a Double Integrating Plus Dead Time (DIPDT) system. The PID controllers are tuned using various empirical methods such as Ziegler-Nichols (Z-N), modified Ziegler-Nichols: Chien, Hrones and Reswick (C-H-R), Cohen-Coon (C-C), and minimum error criteria methods that include Integral of Absolute Error (IAE), Integral of Squared Error (ISE) and Integral Time Absolute Error (ITAE). Moreover, PID parameters are optimally found with Genetic Algorithm (GA) using different performance indices. The MPC is a constrained optimal control strategy with a finite receding horizon based on a state-space prediction model. The simulation and real-time results of the PID and MPC controllers investigate a comparison between the performances of the two controllers for validating the DTS.
The second part expresses DTHS by a Mixed-Logical Dynamical (MLD) model using HYSDEL Toolbox that can be translated to a Piecewise Affine (PWA) model. The hybrid model has been verified by reachability analysis using the CORA Toolbox. The hybrid model is used to synthesize Hybrid Model Predictive Control (HMPC) with Mixed-Integer Quadratic Programming (MIQP) optimization problem. An Explicit HMPC (EHMPC) has been computed to decrease the online computation in a real-time system. EHMPC is synthesized for regulation and reference tracking problems and simulated using the MPT Toolbox. Hardware-In-the-Loop (HIL) simulation is performed using the TCP/IP protocol for testing the EHMPC with a hardware real-time controller. The real-time controllers are implemented by the Node.js frameworks. The simulation and real-time results are provided to study the performance of the hybrid controller and how it works with the DTHS.