Application of a radial basis function neural network to sensor design

摘要

Abstract: One of the important tasks in sensor design is thedevelopment of a model for a sensing phenomena.Artificial neural networks are ideal for such a taskbecause of their capability for representation of themapping functions describing the processes andphenomena which are mathematically difficult or evenintractable. We examined a radial basis function (RBF)neural network for modeling of acoustical properties ofcolloidal TiO$-2$/ slurry. The colloidal slurry is avery complex multiphase medium. The RBF network with aset of local Gaussian functions was trained using thedata from the earlier developed physical model ofTiO$-2$/ slurry. Next the TiO$-2$/ neural model wasused for a prediction of the TiO$-2$/ particle sizedistribution. The resulting prediction accuracies ofthe RBF network were 99.8% for the data used in thetraining process and 88% for the data not used in thetraining. Compared to other available techniques neuralnetworks can offer an effective and time efficientapproach for the modeling of complex materials. !15