Intelligent active noise control applied to a laboratory railway coach model

In recent years, due to the need to improve the quality and comfort of railway transportation means, train operators and manufacturers have become increasingly focused on reducing the levels of noise and vibration experienced by passengers. Both passive and active design methods can be used for suppressing noise inside a train coach. The success of this design is strongly linked with the accuracy of the acoustic models in characterizing the sound pressure level distribution inside the train coach. Recent research has emphasized the importance of using nonlinear control techniques in active noise reduction applications. In this paper, both fuzzy and neural modelling paradigms are integrated in the active noise control scheme. Their ability to build reliable nonlinear acoustic model predictions will be tested for the active noise cancellation in a laboratory setup emulating the interior of a train coach. It is shown that the performance of the proposed intelligent control schemes outperforms the classical FIR-based active noise controller.