Experimental and Artificial Neural Network Investigation on the Thermal Efficiency of Two-Phase Closed Thermosyphon

Abstract

The main purpose of this study is to investigate the thermal efficiency of a Two-Phase Closed Thermosyphon (TPCT). For this purpose, initially, an experimental study was performed, then to predict the other experimental conditions ANN model which has used a wide range of thermal engineering systems was developed. A vertical copper pipe charged with different working fluids as pure water, ethanol, and ethylene glycol were used for TPCT. Impact of the various parameter such as heating power input, inclination angle, cooling water flow rate and working fluid type on the TPCT efficiency are examined. It is found that the increase in the inclination angle increased the TPCT efficiency while the increase in heating power input decreased efficiency. Regression analysis was applied to examine the performance of ANN between estimated and experimental data. The Mean Absolute Percentage Error (MAPE) was found to be less than 1.3 % for the training set and 3.1% for the test data set. The ANN predictions yield R2 in the range of 0.9998 for the training set and 0.9989 for the test data set. The obtained results from the experimental study and ANN were found in good agreement, and it is also concluded that from the study the ANN is a useful tool to estimate such thermal engineering problems. Keywords: Thermal efficiency Heat pipe Two-phase closed thermosyphon ANN