INVESTIGATIONS ON AUTOMATED AND NON LINEAR CONTROL SYSTEMS

Abstract

A control system can range from a single home heating controller to large industrial control systems for controlling different processes or devices. Control theory evolution leads to the development of analytical controller design techniques and an automatic decision-making process that improves the performance of control benchmark problems in real time. Lately, automation has been gaining traction globally owing to its huge social, economic and environmental benefits. However, most of the industrial plants have hard non-linearities, which makes it difficult to design and analyse closed-loop control systems as compared to linear ones. However, a significant portion of the literature exists for traditional control through distributed control systems (DCS) and field-bus technologies to monitor and control all the processes of industrial automation control systems. But none of them have resulted in an intelligent and highly productive real-time control of the plant. Also, the current systems used for industrial monitoring and control have various drawbacks related to expensive hardware and difficulty of integration with other network protocols. Very little research is focused on Programmable Logic Controller (PLC) and Supervisory Control and Data Acquisition (SCADA) based real-time automation systems to accelerate industrial processes and perform industrial operations remotely. Additionally, nonlinearity plays an important role in the controlled process or controllers in nonlinear control systems. Nonlinear control systems are used in numerous engineering systems like aerospace control, industrial process control, autonomous robots, position control, trajectory tracking, path planning, etc. A nonlinear controller through robustness or adaptability can handle the consequences due to hard nonlinearities, and model uncertainties. Non-linear systems behave differently at different operating points and hence it is not preferred for industrial applications due to its complexity, non-flexibility in tuning and poor reconfigurability. Thus, there is a need for an adaptive method to resolve the issues of non-linearities in the traditional control of industrial automation systems and improve the performance and stability of non-linear control systems like inverted pendulum and ball balancer systems. The objective of this thesis is to investigate automated and non-linear control systems and propose robust, reliable, and efficient methods for various automated and non-linear control systems. In this research, a variety of techniques are applied to automated and non-linear control ix systems. Some newly introduced applications of automation and ratio control strategy in the Industry 4.0 concept are also studied. Various methods have been designed and applied to improve the performance and stability of non-linear control systems along with the implementation of various optimization techniques to these non-linear control systems. The contributions presented in this thesis are outlined below: • A PLC and SCADA-based control framework is proposed to automate the process industry plant and monitor all the processes using a single screen Human Machine Interface (HMI). The proposed method addresses a dynamic real-world problem for the mixing of raw materials, filling of final product composition, capping, labelling and sorting of containers on the basis of both size and type using a single assembly line by applying the developed model in a real-life case study of an assembly line from a chemical process industry supplier in north India to verify the design for effectively balancing a real-world assembly line in a process industry. • A PLC-based control framework is developed to automate the processes in the heat exchanger plant and monitor all the processes using SCADA. A PLC and SCADA based control framework is designed to control the temperature of a heat exchanger system through Proportional Integral Derivative (PID) and Fractional Order PID (PI λD μ ) controllers and the performance of the controllers is optimized using Genetic algorithm (GA), Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) techniques. • Intelligent controllers are introduced for the real-time balancing and position-tracking control of a 2-DOF ball balancer using i. PID with Integral ANTI-WINDUP controller with different velocity setpoints values ii. Neuro Integrated Fuzzy PID controllers with different 𝑉𝑠𝑤 values • A continuous Linear Quadratic Gaussian (LQG) controller is proposed for optimal control of a rotary inverted pendulum to swing up the rotary inverted pendulum upward from its stationary downward position and maintain an equilibrium in the vertical upward position. • A PID and Fractional Order PID (PI λD μ ) controller with Integral ANTI-WINDUP technique is designed for swing-up and stabilization control of rotary inverted pendulum system by using metaheuristic optimization techniques. Simulation and real- x time experimentation analysis of the pendulum angle, rotary arm angle, controlled input voltage, transient response and steady-state response have been done for the proposed controllers. The proposed method is efficient in i. keeping the pendulum balanced and preserving some degree of tolerance for a vertical upright position. ii. bringing the pendulum out of its state of passive equilibrium to achieve balance. iii. both reference tracking and rejection of external disturbances in comparison to the existing literature Simulation and real-time experimental analysis have been done extensively to prove the efficacy of the developed solutions.