Modelling and Predictive Control of Electromechanical Actuators for All-Electric Nose Landing Gear Systems

This study investigates the modelling and predictive control methods for electromechanical actuators (EMAs) used in the retraction and extension system of all-electric nose landing gears. By integrating predictive control theory, the proposed approach aims to enhance control performance and system reliability. A discrete-time EMA model is developed to establish the relationship between predicted current and actuator dynamics. A cost function minimisation algorithm is constructed using switching states, predicted current and measured current values to determine the optimal switching sequence, thereby generating the voltage vectors required for motor operation. To address potential faults such as unbalanced loads, magnetic interference or environmental factors, this study employs a fault diagnosis method based on feedback current, predicted current and adaptive thresholds. Upon detecting actuator failure, a secondary control loop enables emergency gear release. This dual-loop strategy ensures routine and emergency functionality, delivering over 2000 N of thrust at operating speeds ranging from 5 mm/s to 8 mm/s. A prototype EMA system was developed, and experimental results confirm its feasibility, accuracy and robustness, providing a reliable solution for all-electric landing gear applications.