Reference signal estimation in active noise reducing system of a washing machine

Noise generated by devices is an important problem in numerous environments. It can be successfully reduced by controlling vibration of their casings. Efficiency of this method has been confirmed by the authors in previous publications. A feedforward control structure is usually employed in an active control system. However, a useful reference signal is sometimes impossible to acquire. Fortunately, for the active casing problem often the dominant noise components are generated by rotating device elements, and thus they are tonal or multi-tonal. In such a case, alternatively to feedforward control, a feedback control strategy in the Internal Model Control (IMC) architecture can be used, where error microphones are employed to estimate the reference signal. The purpose of this paper is to investigate the impact of error microphone selection for reference signal estimation, which may affect the final performance of the control system. Different error signals are used for reference signal estimation and obtained results are compared and discussed. A real device is used for active control experiment, namely, a market-available and unmodified washing machine. The Filtered-x Least Mean Square (FxLMS) algorithm is used to adaptively update control filter parameters. A low-frequency noise in the range up to 500 Hz is considered.