Neuro-adaptive active control schemes for noise cancellation and vibration suppression are developed and presented in this chapter. Multilayered perceptron and radial basis function neural networks are considered in both the modelling and control contexts. A feedforward ANC structure is considered for optimum cancellation of broadband noise in a three-dimensional propagation medium. Online adaptation and training mechanisms allowing the neural network architecture to characterise the optimal controller within the ANC system are developed. The neuro-adaptive ANC algorithms thus developed are implemented within a free-field noise environment and simulation results verifying their performances in the cancellation of broadband noise are presented and discussed.

Inspec keywords: adaptive control; vibration control; noise abatement; radial basis function networks; neurocontrollers; learning (artificial intelligence); multilayer perceptrons

Other keywords: radial basis function neural network; free field noise environment; multilayered perceptron; optimum cancellation; noise cancellation; three dimensional propagation medium; vibration control; neural network architecture; neuroadaptive ANC algorithm; vibration suppression; broadband noise cancellation; online adaptation; broadband noise control; ANC system; feedforward ANC structure; optimal controller; neuroadaptive active control scheme; active control; training mechanism

Subjects: Neurocontrol