Model-based dimensionless neural networks for fin-and-tube condenser performance evaluation

Fin-and-tube condenser performance modeling with neural network and response surface methodology

“…What's more, the percentage of training data and the number of neurons have a combined influence on NN accuracy and over-fitting probability. As pointed out by Yang et al (2014), at certain percentage, the more the hidden neurons, the better the training accuracy and the higher the over-fitting risk. On the other hand, at certain number of hidden neurons, more training data will mitigate the over-fitting risk.…”

“…On the other hand, at certain number of hidden neurons, more training data will mitigate the over-fitting risk. It can be noticed that a trial-and-error process was applied to get the ''optimal'' NN structure in most NN modeling (Xie et al, 2007;Yang et al, 2014;Zhao et al, 2010;Zhao and Zhang, 2010). This process proved to be effective but inefficient, since it's quite time-consuming and has to be performed all over again in different problems.…”

“…Yang (2008) and Mohanraj et al (2012) reviewed NN applications in thermal engineering and in air-conditioning, refrigeration and heat pump systems, respectively. As for the heat exchanger analysis, there have emerged abundant researches for the past decades (Diaz et al, 1999;Islamoglu, 2003;Pacheco-Vega et al, 2001b;Peng and Ling, 2008;Wang et al, 2006;Xie et al, 2007;Yang et al, 2014;Zhao and Zhang, 2010). In all related researches, good agreement using neural network was always claimed.…”