Modeling and controller design of a flotation column

Abstract

An improved technique for the design of decentralised dynamic decoupled Proportional Integral (PI) controllers to control numerous variables of column flotation is developed and implemented in this thesis. This thesis was motivated by challenges when working with Multiple Inputs Multiple Outputs (MIMO) systems that are not controllable by conventional linear feedback controllers. Conventional feedback control design consists of various drawbacks, especially with the introduction of complex industrial processes. The introduction of nonlinear controller design, decentralization, and decoupling of a system overcome these drawbacks. The reason these advanced controllers are needed is because of the complex interaction that is required by the system. Therefore, designing controllers or control systems that mitigate stability is important. In this thesis different innovative control design methods and algorithms which are based on decentralized coupled and decentralized dynamic decoupled systems are developed. This thesis first focused on the mathematical modeling of the column flotation system. The column flotation system model and dynamic characteristics were analysed to achieve a good understanding of the system’s behaviour. The system's dynamic behavior is assessed based on multiple changes in the input circumstances. The analysis of the open-loop and closed-loop systems under study was performed based on the Matlab/Simulink simulation environment. The Column Flotation process was modelled by a 2x2 and 3x3 multivariable system and simulated in Matlab/Simulink. Through several evaluations, it was noted that the most critical constraints were the maximum value of wash water (Qw), the minimum of the froth layer height, and the minimum of the gas holdup. The new improved decentralized controller was thoughtfully developed using single-loop parings. Relative Gain Array (RGA) method was deployed to reduce the effects of process interactions when designing the decentralized controller. The design technique adopted was using Internal Model Controller-based (IMC) PID feedback control for set-point tracking. Set-point tracking control was achieved, and the effects of various disturbances on the behaviour of the designed closed-loop systems were investigated and analysed. The developed strategies were then deployed by a special transformation process from Simulink to a Beckhoff PLC via the functional block programming language TwinCAT 3.1. Beckhoff CX5020 PLC together with TwinCAT 3 software was used for the implementation of the decentralized coupled and decentralized dynamic decoupled model-based controllers. The technique used and implemented deployed a Programmable Logic Controller (PLC) as it allows the model transformation from Matlab/Simulink to be implemented directly for industrial application purposes. v The effectiveness of set-point tracking control and disturbance rejection was assessed in this thesis. The desired variables were achieved in run-time mode using the TwinCAT 3 functional blocks module, which was then downloaded to the Beckhoff CX5020 PLC for real-time implementation. One of the reasons for using the Beckhoff PLC CX5020 as an implementation environment was motivated by the reliability of this platform and Beckhoff CX5020 that is built according to new industry standards and allowing transformation which makes it more advantageous to use more than any other Programmable Logic Controllers.