Liquid Level Control Performance Study of Conventional and Advanced Model-Based Controller for a Quadruple Tank System

Liquid level control in a quadruple tank system (QTS) is crucial for various industrialapplications due to its multivariable nature. Conventional controllers frequently encounterdifficulties maintaining performance in complex applications, showing transient and steadystateresponse difficulties. This study evaluates and compares the effectiveness of ModelPredictive Control (MPC) and Proportional-Integral-Derivative (PID) controller inregulating the liquid level of a quadruple tank system to improve control precision, stabilityand overall performance. The investigation utilized controller tuning to evaluateperformance metrics across different operational scenarios and constraints, including risetime, settling time, and overshoot. The results demonstrate that MPC has supersededperformance compared to PID controller, resulting in faster response time, shorter settlingtime, and decreased overshoot from the desired set point. As a result, constrained MPCachieved a 15% decrease in overshoot and a 20% decrease in settling time compared to thePID controller. To validate the robustness of MPC, set point changes were applied comparedwith PID controller performance metrics. MPC transitioned smoothly between 12 and 18 cmtargets without overshoot and steady-state error. In contrast, PID incurred a persistent 0.5cm error. This research contributes to the fact that MPC is a more effective method ofcontrolling QTS. MPC demonstrates better performance in both transient and steady-stateresponse.