Parallel Active Control of Acceleration Noise 2017-32-0068

The number of people experiencing psychological discomfort due to the increasing amount of noise emanating from motor vehicles has been on the rise. Legal regulations define the permissible level of vehicle noise in a given area. Active noise control (ANC) is a noise cancellation method that reduces low-frequency sounds, such as engine noise, effectively. Furthermore, this method is suitable for controlling engine noise because the equipment necessary to perform it is small and does not require a large space for installation. Advances in digital processing technology have increased the scope of ANC’s applications, and it is no longer restricted to use in motor vehicles.

The purpose of this study is to demonstrate the effectiveness of the proposed method in reducing the motor vehicle engine noise produced during acceleration. In this study, we attempt to control the engine sounds from a vehicle with a four-cylinder four-stroke engine. In previous studies [7], we proposed using and performed experiments on a reference signal for both the fundamental frequency and its harmonic, calculated from the engine pulse and focused on harmonic structures of engine sound. However, tracking the variations in the fundamental frequency resulting from the fluctuation of the engine’s rotation during acceleration presented quite a challenge. Therefore, in the present study, a sinusoidal waves, consisting of numerous frequencies, are employed as reference signals. Each reference signal is inputted to its respective adaptive filter. In this manner, we attempt to expand the reduction range by utilizing multiple adaptive filters.

As a result, we achieve noise reductions of at least 8 dB for acceleration and at least 20 dB for constant-speed driving.