Analysis of influencing factors of the production performance of an enhanced geothermal system (EGS) with numerical simulation and artificial neural network (ANN)

Abstract Understanding the sensitivity to different reservoir parameters can help optimize the use of a designated reservoir. Four key parameters, namely, fracture permeability, well spacing, injection temperature, and injection rate, are considered in this study. The effects of various factors on the thermal performance of the 4300-4700 m granodiorite reservoir in the Zhacang geothermal field in Guide Basin, Qinghai Province, China is analyzed via numerical simulation and artificial neural network (ANN). The ANN models are designed to develop an effective system in less time. The training and test data of the ANN models are the results of numerical simulation. The prediction accuracy is measured by the coefficient of determination and the root mean squared error. Results demonstrate that the use of ANN for predicting the production temperature has high prediction accuracy. Finally, the effects of various factors on the total heat extraction are further analyzed. The results show that the injection rate exerts the largest influence on total heat extraction, followed by the injection temperature and well spacing, and fracture permeability is the least relevant. Increasing the injection flow rate, lowering the injection temperature, increasing the distance between the injection and the production well, and reducing the fracture permeability can improve heat production within certain ranges. In this study, the combination of an injection temperature of 30°C, injection flow rate of 60 kg/s, fracture permeability of 1 × 10 −12 and well spacing of 600 m was chosen as the best scheme for the heat production.The accumulative total energy produced in 30 years period is 4.08 × 10 16 J based on the simulation results, which can save 1.7 × 10 9 kg of the coal.