Model-Free Adaptive Time-Delay-Based Estimation Control for Input-Saturated Coupled Tank System: Experimental Validation

Coupled tank systems are widely employed in various industries, including wastewater treatment, water desalination, pharmaceuticals, and petrochemicals. Precise liquid level regulation is a critical challenge in these applications, especially under system uncertainty and disturbance. This paper introduces a novel model-free, robust adaptive time delay-based estimation control to address this challenge, with disturbances and input saturation. The proposed controller employs a time delay-based estimation technique to estimate unknown system dynamics using the immediate past input and output data. This estimated information is then integrated into a composite control scheme to compensate for unknown dynamics and disturbances. A new adaptive-based proportional controller is proposed, which avoids overestimating the adaptive gains. The closed-loop analysis through Lyapunov theory guarantees uniformly ultimately boundedness of all the internal states. Experimental results validate the better performance of the proposed controller compared to existing schemes under different operating conditions. Note to Practitioners—Controlling liquid levels in coupled tank systems is crucial in many industries, from wastewater treatment to pharmaceuticals. This paper presents a new, practical control strategy to address the challenge of precise level regulation, especially when dealing with system uncertainties, disturbances, and limitations on the control input. The proposed approach is “model-free,” meaning it doesn’t require the mathematical model of the tank system, making it easier to implement in real-world scenarios. The controller only requires the past input and output data to estimate the system’s behavior and then incorporates this estimate into a robust control scheme that compensates for unknown dynamics and disturbances. Another key feature is the application of a new adaptive controller, which avoids the problem of overestimating the control gains, thus reducing excessive control effort. The effectiveness of the proposed scheme is experimentally demonstrated by controlling the water levels in a coupled tank system. The experimental results elucidate that the new controller outperforms traditional methods under various operating conditions, offering improved performance for practitioners dealing with coupled tank systems.