Investigations on Controller Design Using data Driven Control Approach for Non Linear Systems

Abstract

The study of non-linear systems has become increasingly important in newlinemany industries due to the fact that many industrial processes are inherently newlinenon-linear and involve strong interactions between different variables. In newlineparticular, the control of liquid level in tanks is an essential part of many process newlineindustries, as maintaining the correct level is crucial for ensuring efficient and newlinesafe operation. newlineIf the level of liquid in a tank is too high, it can cause problems such as newlineupsetting the reaction equilibrium, damaging equipment or causing spillage of newlinevaluable or hazardous material. On the other hand, if the level is too low, it can newlinealso have adverse effects on the sequential operation of the process. Therefore, newlinecontrolling the flow and level of liquid in industrial processes is a critical task newlineand the development of advanced control techniques using cutting-edge newlinetechnology is essential for ensuring safe and efficient operation. The work newlinediscusses the control of liquid levels in tanks, particularly in the conical and newlinespherical shaped tanks that are widely used in various process industries. These newlinetanks are nonlinear and can pose a challenge for level control, which can affect newlinethe output of the process. To address this issue, modelling and controller design newlinewere carried out on a single conical tank using traditional modelling, system newlineidentification and data-based methods. The study found that data-based newlinecontroller design was the most effective approach. The work is also carried out newlineon the control of liquid levels in spherical tanks, which is similarly challenging newlinedue to variations in the area of the transverse section. newline