Low-order model based smooth control of proportional valve controlled hydraulic system with pressure compensation

To address the issues of strong impact and low accuracy in position control of hydraulic construction machinery due to large load variations and limited controller computational capabilities, a low-order model based smooth control method for a pressure-compensated proportional valve-controlled hydraulic system was proposed. Firstly, a pressure-compensated proportional valve-controlled system, which can be approximated as a linear system was proved. A novel low-order modeling approach was introduced by using a piecewise linear stochastic system for a global approximation of the original system, and the quantitative covariance matrices were given. Secondly, a steady-state Kalman velocity observer and a dual-loop controller reliant on velocity observation were introduced. This approach utilizes precise velocity measurements and internal control loops to enhance system damping, thereby effectively reducing oscillations. Finally, the proposed method was experimentally validated on the lifting arm of a shield segment assembly machine, involving both AMESim-Simulink co-simulation experiments and real-machine experiments controlled by PLC. The results show that the proposed low-order model ensures that the relative error of accuracy under typical valve parameters remains below 3% by comparing with accurate model, and the proposed control method achieves low-impact, high-precision control of step position targets on low-computational-power controllers. It improves system damping, smoothes the velocity curve and reduces steady-state errors. The method reduces peak force impact by 21.6 kN and amplitude by about 39%, showing higer engineering potential. © 2024 Central South University of Technology. All rights reserved.