Extensive reliance on manual controller tuning in closed-loop material testing applications makes such applications good candidates for self-tuning control implementation. The current work deals with the modeling, on-line identification, and self-tuning control of electrohydraulic servomechanisms used for closed-loop control of stroke, load, or strain involving low signal frequencies. The application to load and strain control results in servovalve operation close to null and increases the significance of nonideal aspects of servovalve operation such as valve mechanical null offset, valve overlap and radial clearance. Modifications in process modeling and on-line identification necessary to accommodate these aspects are described and shown to be effective by simulation and experiment. A pole-placement controller design approach is used for controller adaptation, the control algorithm being chosen based on analysis of robustness of the control system. The self-tuning control system is shown to be effective, by simulation and experiment.
Monitoring procedure for intelligent control: on-line identification of maximum closed-loop log modulus