Water Pump Control: A Hybrid Data-Driven and Model-Assisted Active Disturbance Rejection Approach

Abstract

Water pump control, prevalent in various industrial plants, such as wastewater treatment and steam generator facilities, plays a significant role in maintaining economic efficiency and stable plant operation. Due to its slow dynamics, strong nonlinearity, and various disturbances, it is also widely studied as a typical benchmark problem in process control. The current control strategies can be categorized into two aspects: one branch resorts to model-based design and the other to data-driven design. To merge the merits and overcome the deficiencies of each paradigm, this paper proposes a hybrid data-driven and model-assisted control strategy, namely modified active disturbance rejection control (MADRC). The model information regarding water dynamics is incorporated into an extended state observer (ESO), which is used to estimate and mitigate the limitations of slow dynamics, strong nonlinearity, and various disturbances by analyzing the real-time data. The tuning formula is given in terms of the desired closed-loop performance. It is shown that MADRC is able to produce a satisfactory control performance while maintaining a low sensitivity to the measurement noise under general parametric setting conditions. The simulation results verify the clear superiority of MADRC over the proportional-integral (PI) controller and the conventional ADRC, and the results also evidence its noise reduction effects. The experimental results agree well with the simulation results based on a water tank setup. The proposed MADRC approach is able to improve the control performance while reducing the actuator fluctuation. The results presented in this paper offer a promising methodology for the water control loops widely used in the water industry.

Active Disturbance Rejection Control for Robot Manipulator

“…The ADRC is based on an extension of the system model with an additional and fictitious state variable, representing those elements of the system dynamics that the user does not include in the mathematical description of the plant. These virtual states (sum of internal and external disturbances, sometimes denoted as a total disturbance) are estimated online and used in the control loop to decouple the system from the actual perturbation acting on the plant (10) [34], [35], [36], [37], [38], [39] as…”

“…The principle in the ADRC has supported consists of the estimation and later cancellation of 𝑓. Assuming that 𝑓 is derivable, the (10) of the system can be represented in (11) as an extended state form [6], [10], [33], [34], [35], [36], [37], [38], [39], [40], [41].…”