Developing the logic for evaluating the indoor noise of a naval vessel using a back-propagation neural network

The noise inside a naval vessel is very important in considering the need for hearing protection, improving the working environment and maintaining good communications for crews living on board a naval vessel. The indoor noise of a ship usually is specified by the Aweighted sound pressure level, but other evaluating parameters are required to reflect human senses more effectively. This paper uses additional psychological responses related to room acoustics, such as NR (noise rating), NC (noise criterion), RC (room criterion), PSIL (preferred speech interference level) and loudness to evaluate the noise inside cabins and offices on a naval vessel. Through comparing these psychological responses to the noise against the survey results describing the crew’s degree of satisfaction with the indoor noise aboard the naval vessel, this paper proposes a new evaluation index using a back-propagation neural network algorithm which can represent the sound quality of the indoor noise in a naval vessel.