Thermal monitoring and optimization of geothermal district heating systems using artificial neural network: A case study

This paper deals with determine the energy and exergy efficiencies and exergy destructions for thermal optimization of a geothermal district heating system by using artificial neural network (ANN) technique. As a comprehensive case study, the Afyonkarahisar geothermal district heating system (AGDHS) in Afyonkarahisar/Turkey is considered and its actual thermal data as of average weekly data are collected in heating seasons during the period 2006–2010 for ANN based monitoring and thermal optimization. The measured data and calculated values are used at the design of Levenberg-Marquardt (LM) based multi-layer perceptron (MLP) in Matlab program. The results of the study are described graphically. The results show that the developed model is found to quickly predict the thermal performance and exergy destructions of the AGDHS with good accuracy. In addition, two main factors play important roles in the thermal optimization: (i) ambient temperature and (ii) flow rates in energy distribution cycle of the AGDHS. Various cases are investigated to determine how to change the energy and exergy efficiencies of the AGDHS for the temperature and flow rate. Finally, a monitoring and performance evaluation of a geothermal district heating system and its components by ANN will reduce the losses and human involvement and make the system more effective and efficient.

► A GDHS is monitored and optimized by a method, combining the ANN with exergy.
► The actual thermal data are collected in heating seasons during the period 2006–2010.
► The measured and calculated values are used at the design of ANN in Matlab.
► ANN is to quickly predict the thermal performance of the AGDHS with good accuracy.
► It is showed how to change its thermal efficiencies for the temperature and flow rate.