Output Feedback Control of Dual-Valve Electro-Hydraulic Valve Based on Cascade Structure Extended State Observer Systems with Disturbance Compensation

In the development trend of intelligent and high-performance construction machinery, the dual-spool electro-hydraulic valve, as a new-generation core control element, directly affects the operation accuracy and energy-efficiency level of construction machinery. The standard linear extended state observer (LESO) produces relatively serious peaks as the system order increases, which leads to the degradation of the observer’s performance and affects the controller’s accuracy. To solve this problem, this paper innovatively proposes an output feedback control strategy for a cascaded structure observer for the dual-spool electro-hydraulic valve. This paper designs an output feedback controller based on the cascaded structure observer. The uniform exponential stability (USE) criterion ensures that the tracking error of the observer for the system state is bounded. The expected load pressure is constructed based on the expected trajectory to replace the actual load pressure, avoiding the influence of the nonlinear coupling between the load pressure and the input signal on the control system. Finally, a stable output feedback controller is obtained based on the backstepping control method and Hurwitz polynomial stability analysis. This study first applies the cascaded structure observer to the field of dual-spool electro-hydraulic valve control, providing a new theoretical framework and technical path for the high-precision control of the hydraulic system of construction machinery. Theoretical analysis shows that compared with the standard LESO, the cascaded structure observer can significantly reduce the online computational burden and effectively suppress the peak phenomenon, providing stronger estimation ability. Finally, a large number of simulation examples verify the effectiveness and superiority of the output feedback controller based on the cascaded structure observer. In all four test scenarios, the average tracking error of C1 (the output feedback controller designed based on the cascaded structure linear extended state observer) is about 5.1%, the average tracking error of C2 (the output feedback controller designed based on the standard structure linear extended state observer) is about 7.8%, and the average tracking error of C3 (the high-gain PID controller) is about 19.2%. The average control accuracy of the designed C1 controller is improved by 2.7% and 14.1% compared with C2 and C3, respectively. In terms of the estimation of external disturbances, the average error of C1 is 14% and the average error of C2 is 29.6%. The estimation accuracy of the former is improved by 15.6% compared with the latter.