Skip to Main Content
Maintenance of membrane structures geometry is crucial for them to be utilized in space missions, for instance, the synthetic aperture radar (SAR) missions, for which a membrane antenna system is being developed at the Canadian Space Agency. An active flatness control system is developed to control and maintain the required surface flatness of the membrane antenna. Taking a mechatronics approach, twenty shape memory alloy (SMA) actuators are installed around the membrane boundary to apply tension forces to the membrane. Wrinkling-related surface deviations have a highly nonlinear relationship with applied tension forces, which can have different forms when the membrane is subject to different thermal disturbances. In this paper, genetic algorithms are used to search for tension force combinations reducing thermally induced wrinkles to the required level. To overcome the premature convergence problem during search processes, adaptive rules are devised to regulate GA parameters so that tension force combinations could be found faster and more robustly, named as an adaptive genetic algorithm (AGA). Through experimental studies, it is demonstrated that the AGA can expedite search process and prevent premature convergence. This intelligent mechatronics solution can potentially be used in real-time active flatness control.