Compressor performance prediction using a novel feed-forward neural network based on Gaussian kernel function

摘要: In this article, a novel artificial neural network integrating feed-forward back-propagation with Gaussian kernel function is proposed for the prediction of compressor performance map. To demonstrate potential capability approach typical interpolated and extrapolated predictions, other two classical data-driven modeling methods including support vector machine are compared. An assessment performed discussed on sensitivity different models to number training samples (48 samples, 32 18 samples). All results indicate that in article has superior existing machine, especially extrapolation small samples. Furthermore, study can be utilized refining performance-based improved simulation ...