| Summary: | Maintaining precise levels of liquid in storage tanks is crucial across various
industries, including chemical manufacturing, water treatment, food and beverage production, and climate control for buildings. This dissertation explores
the application of the Model Predictive Control approach to a Coupled Tank
System provided by Kentridge Instruments Pte Ltd. This system was chosen
for its ability to present a complex control problem due to the interaction
between the tanks and the constraints on water levels and flow rates, making
it an ideal candidate for Model Predictive Control application.
The system modeling is based on the First Principles approach, using the
mass balance equation to describe water volume in each tank. This method
derives dynamic differential equations and accounts for water flow, identifying essential system parameters for an approximate state-space model of the
Coupled Tank System.
Model Predictive Control (MPC) predicts system behavior to compute optimal control actions over a time horizon. MATLAB and Simulink simulated
the system, focusing on flow rate, water level, and other constraints parameters. The results showed MPC enhances control in multiple input/output
systems, improving operational efficiency and performance reliability in liquid
management applications.
|
|---|