Rotated General Regression Neural Network

Abstract

A rotated general regression neural network is presented as an enhancement to the general regression neural network. A variable kernel estimate for multivariate densities is considered. A coordinate transformation is adopted which circumvent the difficulty of predicting multimodal distribution with large variance differences between modes which is associated with the general regression neural network. The proposed technique trains the network in a way that the variance differences between modes is kept small and in the same order. Further, the technique reduces the number of indispensable training parameters to two parameters and lowers the load of the computation as well as the time for conditions in which employing separate values of σ is unavoidable. The accuracy of the proposed technique is demonstrated by examining two different cases: the performance map of an axial compressor and the boundary layer profile over a flat plate. The results are compared with those by general regression neural network as well as the corresponding experimental data. Excellent improvement is obtained.