Development and Experimental Application of a Pre-Design Tool for Active Noise and Vibration Reduction Systems

Abstract

Typical challenges in the design of active systems for noise and vibration reduction applied to large lightweight structures are complex material systems under distributed and stochastically varying loading. Having a limited set of actuators and sensors available, it is required in the pre-design process to derive an optimal system performance with respect to above mentioned boundary conditions. Therefore, a pre-design tool has been developed that numerically computes an actuator placement based on a structural model derived from either finite-element models or experimental system identification. It furthermore accounts for flat and simply curved plates with loading of multiple point forces or diffuse sound fields. A model for piezoceramic actuators is implemented and a modal correction method applied to account for its effect to the structural model. Subsequently, a genetic algorithm is utilized for actuator placement with respect to sound radiation or average vibration level as well as the actuators power consumption. Based on the derived actuator placement, the patches are applied on a stiffened carbon fiber reinforced plastic panel mounted in a transmission test facility. Control objective is the reduction of sound transmission by manipulating vibrations of the panel. Different feed-forward and feedback control concepts are considered and implemented.