Development on the sound index for the evaluation of sound quality of a refrigerator using artificial neural network

초록

The noise radiated from refrigerator is not high sound pressure. However the background noise in a home is very quite. Thus, the noise of the refrigerator is very much annoyance in a home. Especially, in the night, people feel that this noise is very loudness. Until now, in order to evaluate the noise of the refrigerator the A-weighted sound pressure has been used but it does not gives whole information about the sound quality of the refrigerator. In this paper, sound index is developed for the evaluation based on the human sensibility. It is a new objective measurement method for the noise of the refrigerator. In order to develop this new method the artificial neural network has been used. Sound metrics are used for the input of the neural network. For the training of the neural network, the subjective evaluations for the sounds of many refrigerators have been used. In order to consider the directivity of the noise from refrigerator, two sound index are developed depend on the measurement direction.