Abstract: Existing circulating dryers predominantly rely on manual operation guided by empirical experience, which often leads to suboptimal drying efficiency and inconsistent product quality—critical drawbacks in modern agricultural processing systems. These limitations are further compounded by the inherent complexity of the paddy deep-bed drying process, which exhibits strong nonlinear dynamics, significant time delays in heat and mass transfer, tight multivariable coupling among key operational parameters such as grain moisture content, airflow rate, and drying temperature, as well as high sensitivity to multiple internal and external disturbances, including fluctuations in ambient humidity, variations in initial grain moisture, and changes in feed rate. Collectively, these characteristics render the development and implementation of effective, robust, and automated control strategies a particularly challenging task. To address these challenges and achieve verifiable optimal control for paddy circulation counter-flow drying, this study presents a dedicated advanced controller based on the theoretical framework of Model Predictive Control (MPC), specifically designed to accommodate the dynamic behavior and structural constraints of this drying configuration. The core functionality of the proposed controller lies in its ability to dynamically regulate the paddy mass flow rate—the primary manipulated variable—in real time, thereby adjusting the material’s residence time within the drying chamber and guiding the entire drying trajectory toward the desired final moisture content with high precision. By utilizing an embedded predictive model, the controller forecasts future system states over a finite receding horizon and computes a sequence of optimal control actions that minimize deviations from target setpoints while respecting physical, operational, and actuator constraints. A key innovation in the design of the controller is the formulation of its objective function, which incorporates an analytically derived attenuation coefficient, denoted as ‘β’, applied to the penalty term associated with changes in the control input. This strategic modification effectively relaxes the constraint on the rate of change of the paddy flow rate, allowing for more agile and responsive adjustments during transient conditions while maintaining closed-loop stability and preventing excessive wear on actuators or mechanical components. Comprehensive simulation-based experiments were conducted to rigorously evaluate the controller’s fundamental performance attributes, including its robustness to model uncertainty, resilience against unmeasured disturbances, and accuracy in tracking time-varying setpoints. Specific case studies were designed to investigate the controller’s capability to maintain precise control under abnormal operating scenarios, such as sudden equipment malfunctions, thereby validating its suitability for real-world industrial applications. Experimental results demonstrate that the developed MPC controller significantly mitigates the adverse effects of major practical disturbances, particularly variations in the initial moisture content of incoming wet paddy and fluctuations in ambient relative humidity. Across all test conditions, the maximum absolute deviation between the actual and target moisture content of the discharged paddy remained below 0.35% on a wet basis (w.b.), indicating exceptional control accuracy and consistency. Under dynamic setpoint transitions—including step changes, linear ramp profiles, and sinusoidal reference signals—the controller achieved remarkably low Relative Average Deviations (RAD) of 0.11%, 0.07%, and 0.01%, respectively, confirming its superior tracking performance and adaptability to varying operational demands. Furthermore, in stringent anti-interference tests, artificial disturbances with peak amplitudes of ±40%, ±60%, and ±90% were introduced at the outlet stage of the drying process to simulate severe process upsets; under these extreme conditions, the average relative deviation in final moisture content was maintained at only 0.32%, 0.40%, and 0.41%, respectively. Furthermore, in strict anti-interference tests, artificial disturbances with maximum peak amplitudes of ±40%, ±60%, and ±90% were introduced at the outlet stage of the drying process to simulate severe process disturbances. Under these extreme conditions, the average relative deviation of the final moisture content remained at 0.32%, 0.40%, and 0.41%, respectively. Comparative analysis revealed that, compared with other control methods, the average relative deviation (RAD) between the actual and target moisture content of the paddy decreased significantly: compared with the uncontrolled condition, the moisture content RAD decreased by 10.7%, 18.4%, and 30.5%, respectively; compared with the traditional PID control, the RAD decreased by 5.9%, 11.1%, and 21.2%, respectively; and compared with the feedforward PID, the moisture content RAD decreased by 3.0%, 7.5%, and 14.6%, respectively. These results collectively highlight the controller's outstanding ability to attenuate the impact of unknown, variable, and potentially destructive disturbances caused by abnormal or unpredictable operating conditions. Therefore, this study established a novel, theoretically sound, and practically implementable method framework to achieve optimal parameter coordination in the rice circulation and enhance process stability.