Energetic and economic evaluations of geothermal district heating systems by using ANN

This paper proposes an artificial neural network (ANN) technique as a new approach to evaluate the energy input, losses, output, efficiency, and economic optimization of a geothermal district heating system (GDHS). By using ANN, an energetic analysis is evaluated on the Afyon geothermal district heating system (AGDHS) located in the city of Afyonkarahisar, Turkey. Promising results are obtained about the economic evaluation of that system. This has been used to determine if the existing system is operating at its optimal level, and will provide information about the optimal design and profitable operation of the system. The results of the study show that the ANN model used for the prediction of the energy performance of the AGDHS has good statistical performance values: a correlation coefficient of 0.9983 with minimum RMS and MAPE values. The total cost for the AGDHS is profitable when the PWF is higher than 7.9. However, the PWF of the AGDHS was found to be 1.43 for the given values. As a result, while installing a GDHS, one should take into account the influences of the PWF, ambient temperature and flow rate on the total costs of the system in any location where it is to be established.

► Each energetic and economic evaluation of the AGDHS is investigated by using ANN. ► Actual thermal data are collected for the heating seasons in the period 2006–2010. ► ANN is to quickly predict the behavior of the physical AGDHS with good accuracy. ► Information about the best design and most profitable oper/ating of the system is provided. ► Influences of the PWF, ambient temperature and flow rate on the costs are shown.