Using wavelet basis function neural network for nonlinear dynamic modelling of manifold pressure

In this paper, an approach for nonlinear dynamic modelling of the manifold pressure in automotive engines is proposed. In this approach, a neural network with the hidden layer that consists of wavelets is used for the construction of a nonlinear mapping. In order to model the dynamic characteristics of the manifold pressure, the auto-regressive connection is introduced into the wavelet basis function neural network. Finally, the experimental results of the modelling of the manifold pressure are illustrated.