Active stabilisation of buckling in composite laminates.
Thesis DisciplineMechanical Engineering
Degree GrantorUniversity of Canterbury
Degree NameMaster of Engineering
The active suppression of elastic buckling has the potential to significantly increase the effective strength of thin-wall structures. Despite all the interest in smart structures, the active suppression of buckling has received comparatively little attention. This research further develops analytical and experimental techniques for the optimal control of columns and plates using piezo-ceramic actuators. Previous work in this area has included numerous theoretical studies and a very limited number of experiments. Numerical models are formulated to simulate both the structure and its active control system. The inclusion of mixed continuous-discrete control simulations for active laminate design is unique to this research and provides insight into issues that arise when trying to implement a continuous control strategy for this unstable system with a discrete controller subject to sensor and noise error. Therefore, limitations such as sensor uncertainty and noise, actuator saturation and control architecture are included in the model. Three active plate strips and a pneumatic compression loading system are implemented based on simulation results and optimal controller design, to command the structure to deform in ways that interfere with the development of buckling mode shapes. Due to the importance of early detection, the relative effectiveness of active buckling control is shown to be strongly dependent on the performance of the sensing scheme, as well as on structure specific characteristics. Initial experiments highlight the difficulties involved in obtaining ideal buckling behaviour in a practical environment. Correction of initial curvature in laminates is successfully implemented, resulting in buckling curves closely resembling finite element results. Active control is combined with the constant actuator offset required to correct for initial curvature to obtain further gains in effective strength. Current experimental results show a 37% increase in the measured buckling load from 2.1 to 2.9 kN and simulations indicate that the controlled critical load can be further increased given higher actuator authority. These results are significant because they stabilise a structure approximately 30 times stiffer than in any other published results, where this high stiffness has introduced additional difficulties mainly due to the faster dynamics of the structure. They are also the first known experimental results to successfully stabilise an active laminate plate structure.