The PhD work focused on the design and the physical implementation of an Active Noise Control (ANC) system inside a vehicle. In particular, the main objective was to find the optimal microphone and loudspeaker positions for the engine noise cancellation inside the cabin. To investigate the relation between performance and transducer positions, an analytical description in terms of the enclosure normal modes has been used. The optimization has been performed for two different situations: in laboratory conditions, within an enclosure with similar dimensions to a plane, and inside a cabin van. The expected active cancellation has been calculated using a set of experimental transfer functions between all the microphone-loudspeaker couples, and the primary field measured by the error sensors. The selection of the optimal positions is a nonlinear optimization problem that may require the use of natural algorithms under certain circumstances. In this work, the best transducer combinations have been selected considering not only the cancellation performance, but their robustness to varying working conditions, using the simulating annealing algorithm. In this way, the ANC system will maintain the nominal performance against perturbations that may appear during the working life of the vehicle. Finally, the prototype developed that provides the best results with a minimum number of transducers has been installed in the van and tested under normally driving conditions.
 |
|
ANC system inside the van and detail of the loudspeakers on the roof and the microphones on the seat headrests
