The electromagnetic linear actuator is used as the core drive unit to achieve high precision and high response in the direct-drive actuation system. In order to improve the response performance and control accuracy of the linear drive unit, an improved sliding mode-active disturbance rejection control (ISM-ADRC) method was proposed. A motor model was established based on improved LuGre dynamic friction. The position loop adopts the improved integral traditional sliding mode control based on an extended state observer, and the current loop adopts PI control. The stability of the system is verified based on the Lyapunov theory. A nonlinear dilated state observer is used to effectively observe the electromagnetic linear actuator position and velocity information while estimating and compensating the internal and external uncertainty perturbations. At the same time, the saturation function sat(s) is used to replace the sign(s) and introduce the power function of the displacement error variable. The improved integral sliding mode control law further improves the response speed and control accuracy of the controller while reducing the jitter inherent in the conventional sliding mode. Simulation and experimental data show that the proposed improved sliding mode-active disturbance rejection control reduces the 8-mm step response time of the electromagnetic linear actuator by 21.9% and the steady-state error by less than 0.01 mm compared with the conventional sliding-mode control, while the system has 49.4% less adjustment time for abrupt load changes and is more robust to different loads and noise.

In this study, to further improve the dynamic performance and steady-state accuracy of the electromagnetic linear actuator (EMLA) for the direct-drive system, the adaptive non-singular terminal sliding mode controller (ANTSMC) based on an improved disturbance observer (IDOB) is proposed. Accurate modeling of the EMLA based on the LuGre friction model was obtained by identification. Aiming at the continuous disturbance of the system, a non-singular terminal sliding mode control based on improved exponential reaching law (IERL) is designed, which is combined with the advantages of fuzzy control. Adaptive adjustment of the IERL gain is achieved to ensure finite time convergence of the tracking error while suppressing system jitter. Based on the IDOB to achieve effective observation compensation under noise disturbance and further improve the system robustness. The stability of the control is analyzed based on the Lyapunov function. The results show that the proposed method suppresses the jitter and improves the tracking performance and robustness of the servo system for different desired trajectories effectively.

The hydrostatic actuation system based on linear actuators improves the complex piston force and long transmission path of the traditional electro-hydrostatic actuator (EHA). However, new nonlinear factors in the linear actuator and direct-driven piston are introduced into the system, which present challenges to system modeling and control. To improve the accuracy of system performance prediction, this paper analyzed the working characteristics of an electromagnetic direct-drive hydrostatic actuation system (EDHAS). A dynamic model of the electromagnetic linear actuator including the LuGre friction model was established. The high-pressure internal leakage of the direct-drive pump was described by an inclined eccentric leakage model. The Karnopp friction model was applied to solve the problem of switching between viscous and sliding friction in a cylinder. The hydraulic components model was established based on AMESim, and the electromagnetic linear actuator model and the system controller model were established in Matlab/Simulink, to establish a refined electromechanical–hydraulic co-simulation model of the EDHAS with electromagnetic, mechanical, hydraulic, and control coupling. A system performance test platform was built. The simulation results of the direct-drive piston displacement, the system pressure, the system flow rate, and the cylinder displacement match well with experimental results, which verifies the validity and accuracy of the refined modeling method.