Machine Learning Approach to Predict the Heat Transfer Coefficients Pertaining to a Radiant Cooling System Coupled with Mixed and Forced Convection

Recently, the mixed convection in cooled ceiling–displacement ventilation is popular due to the fact that airborne viruses and energy economy are increasing issues. Artificial neural networks are one of the machine learning tools that widely used as engineering tool. In this study, the heat transfer coefficients for a radiant cooling system coupled with mixed and forced convection have been predicted by machine learning approach. This approach should be noted as an initial experimental investigation using an artificial neural network analysis in the open sources regarding with mixed convection systems in real sized living environments. Experimentally obtained heat transfer coefficients have been used in the development of the feed forward back propagation multi-layer perceptron network model. In order to analyze the effect of the input parameters on the prediction performance, two neural network models with different input parameters have been developed such as various temperatures and velocities/heat transfer rates. By means of feed forward back propagation multi-layer perceptron neural network algorithms, heat transfer coefficients of convection, radiation and total have been predicted using the experimentally acquired dataset including 35 data points belonging to the mixed and forced convection conditions in a well-insulated room. Training, validation and test data sets include 70%, 15% and 15% of the dataset, respectively. Training algorithm has been computed via Levenberg-Marquardt one with 10 neurons in the hidden layer. The findings obtained from the computational model have been evaluated as a result of the comparison by the target data with in the ±5% deviation band for all heat transfer coefficients. The performance factors have been computed and the estimation precision of the numerical models has been examined thoroughly