Active noise control (ANC) technology using speakers as actuators has an excellent performance, but since it generates anti-noise signals in a path completely different from the vehicle's noise source, performance can be greatly reduced at a location far from the error sensor, and this problem can get worse as the frequency to be controlled increases. Another approach is the vibration-based ANC technology that excites the body member around the engine mount and reduces acoustic error signals in a passenger compartment. This approach has the advantage of being effective even at a distance from the error sensor and reducing vibration and noise together when trying to cancel out the structure borne noise generated by the engine because the paths from the primary source and the secondary source to the interior are almost identical. This paper presents a systematic methodology for this vibration-based ANC technology to achieve high performance in vehicles. First, through noise reduction simulation, the number of error sensors, the location of actuators, the force requirements, and an appropriate system are proposed. Next, to improve performance even with a small number of actuators and limited power when applying this technology in vehicles, an algorithm using the control effort weighting parameter is introduced into the existing FxLMS algorithm and modelled on the controller. Finally, a noise reduction control test is performed in a vehicle under static and driving conditions, and the results are reviewed.
Read full abstract