Pneumatic actuators find widespread use in industry when motion between two end-points is required, given their high power to weight ratio and low maintenance requirements. However, classical PID control of pneumatic actuators may present several undesired features, such as large steady-state errors. In this work, a two servo-valve architecture was developed for the position control of a servo-pneumatic system. With this architecture, the two servo-valves are independently controlled—the one connected to the charging chamber is controlled so as to maintain an approximately constant pressure in the discharging chamber, while the other handles motion control. The use of this control architecture is justified through analysis of the system model. By using this architecture with linear PID-family controllers, the aim is to enhance motion smoothness and improve the steady-state errors usually obtained with PID controllers in classical architectures, where the control actions are applied symmetrically to each servo-valve. Both simulation and experimental results show that the newly developed architecture compares very favorably to the classical one in terms of motion smoothness, steady-state positioning errors, and robustness to load variations.
The International Journal of Advanced Manufacturing Technology – Springer Journals
Published: Jun 5, 2018