Automation of friction torque identification for vane-type semi-rotary pneumatic actuators

In precise control applications, estimating the friction under operating conditions would greatly improve the control performance. In literature, especially for pneumatic actuators, the existing friction estimation methodologies bear on manual experimental methodology which is prone to user errors, has the difficulty in measurement and data processing, and in addition it takes long experimental durations. In this study, all experimental steps of friction torque estimation for semi-rotary pneumatic actuators have been automated with MATLAB/Stateflow (R) tools to process correctly more experimental data and eliminate potential user or random errors by designing a comprehensive mechatronic setup. The automated system has been integrated with a graphical user interface to facilitate the utilization of the automated setup. With the help of efficient automation, friction torque characteristics with respect to Stribeck model have been investigated under various operating pressure and speed levels. Efficiency of the system has been tested on a vane-type cylinder by constructing friction torque surface graphs. The real-time open-loop control is applied to verify the identified friction parameters at loading-free conditions. Furthermore, closed-loop cascade PID technique has been successfully implemented for under-loading case to verify the results. The outstanding performance of the cylinder at both control approach has exhibited that the estimated friction parameters obtained with automated method were quite accurate and very helpful in the control applications.

Automated summary

In this study, the friction torque estimation for semi-rotary pneumatic actuators has been automated using MATLAB/Stateflow (R) tools, which eliminates potential user or random errors and allows for more data processing. The system, integrated with a graphical user interface, enables easy utilization of the automated setup. The automated method accurately estimates the friction parameters and proves to be very helpful in control applications.