Active Cancellation of Engine Room Noise using Static and Dynamic Neural Networks

Ship interior noise produced by engine room disturbs the crews and the workers. Also, it causes lots of complications for the crews and the passengers in the ship. Active noise cancellation (ANC) is based on the destructive interference between the primary noise and generated noise from the secondary source. In this research, performance of the static and dynamic neural networks is evaluated in active cancellation of sound noise. For this reason, MLP and RBF are designed and trained as static neural networks. After training, performance of static and dynamic networks in noise attenuation are compared. In order to compare the networks appropriately, training and test samples are similar. Moreover, equal number of layers and neurons are considered for the networks. Noise signals from a SPIB database are used in simulation procedures. The simulation results show that designed neural networks present proper performance in ANC because of using training and validation samples in training process. As it is seen, the trained dynamic network and RBF neural network show better performance in noise attenuation than MLP network and achieve 1 dB noise attenuation more than MLP network.