Vibration control is a topic of great interest in the Automatic Control community. For instance, in the aerospace field the high value of the ratio cost/payload suggests the use of lighter structures, with increased structural flexibility. Actually, the structure is subject to the highest accelerations during the launch phase, while in the operative phase relatively low forces are experienced by the capsule and the payload. Thus, if the control designer is able to counteract high accelerations (even for a limited amount of time, for instance during the launch), a less expensive solution is amenable.
Applications of flexible structure control are present in many engineering fields, from building in seismic areas (e.g. the skyscrapers in Japan) to aerospace applications (solar panels, robot arms, antennas are all flexible elements).
Passive methods for vibration reduction have shown to be effective only at relatively high frequencies, since at low frequency would require too thick passive layers. For this reason, the use of active methods at low frequencies is crucial.
Since one of the problem to face is to avoid the excitation of unmodeled high-frequency dynamics, a first approach has employed a limited-rate sliding control, so as to guarantee a sufficiently smooth control signal.
Next, a singular-perturbation based control strategy with state feedback has been proposed, able to operate only in specified subspaces where the structural resonance has to be reduced, while leaving the other modes unaffected, thus avoiding the so-called spillover effect, due to the orthogonality of the modes. Moreover, in order to operate an output feedback, robust sliding observers have been considered.
This methodology has been also experimentally tested on a DC9 bulkhead, available at the Dipartimento di Progettazione Aeronautica dell'Università di Napoli “Federico II”.
Later, also collocated feedback strategies have been considered, thanks to the use of piezoceramic actuators in "self-sensing" configuration, i.e. the same device acts both as actuator and sensor. In this case a second order sliding control strategy has given good results with an output feedback, thus avoiding the need for a state observer.
Finally, the case of a 2-D structures (plates) has been considered, defining a mathematical model, identifying its parameters, defining the optimal location of the piezo actuators and sensors by solving an optimization problem in order to maximize the controllability index and by considering a broadband control law, able to reject disturbances with high frequency spectrum, as shown also in the laboratory mock-up at the Second University of Napoli.