This study addresses the fluid flash evaporation phase change in geothermal production wells. A forced circulation visual experimental platform was designed to investigate flow pattern evolution and differential pressure fluctuation characteristics during flash evaporation, and high-precision flow pattern recognition was achieved via signal decomposition and machine learning. Key steps include: constructing an experimental system with fluid dynamic control, temperature regulation, data acquisition, and a visual pipe section; recording flow patterns (bubble, slug, churn, annular flow) via high-speed photography and analyzing their triggering conditions/morphological features; collecting differential pressure signals (2~3 meters height) and identifying distinct amplitude-frequency-morphology characteristics among flow patterns; applying CEEMD to decompose signals and extract IMF energy spectra; and developing a PSO-LSSVM model using multi-parameters (inlet temperature, velocity, IMF spectra) for high-accuracy recognition. Results provide theoretical support for flash evaporation localization and severity assessment, aiding wellbore optimization and geothermal extraction efficiency improvement.

Understanding thermal energy transfer and fracture evolution in submarine hydrothermal systems is essential for sustainable resource utilization, but simulating these complex multiphase, multi-physics processes is challenging. This study integrates the lattice Boltzmann method with a fully connected neural network to investigate hydrothermal phase separation and its effects on chemical dissolution in carbonate fractures at the pore scale. Specifically, the lattice Boltzmann method simulates gas-liquid phase separation induced by seawater boiling, affecting carbonate fracture dissolution at the pore scale. The fully connected neural network predicts the resulting fracture geometry and dissolution quantities under various physical conditions. Analysis of simulation datasets demonstrates that the fully connected neural network achieves high predictive accuracy, with a total loss of 0.01 and reduces computation time by over 20% compared to traditional methods. The coupled lattice Boltzmann method-fully connected neural network model effectively simulates fractures with sizes ranging from millimeters to centimeters, excelling in handling chemical dissolution, multiphase flows, and multicomponent interactions. This approach offers valuable predictive capabilities for applications such as enhanced geothermal systems and oil reservoir exploitation.

In order to alleviate the environmental crisis and improve energy structure, countries from all over the world have focused on the hot dry rock geothermal resources with great potential and with little pollution. The geothermal heat production from enhanced geothermal system comes with complex multi-field coupling process, and it is of great significance to study the temporal and spatial evolution of geothermal reservoir. In this work, a practical numerical model is established to simulate the heat production process in EGS, and the comparison of thermal-hydraulic (TH), thermal-hydraulic-mechanical (THM) and thermal-hydraulic-mechanical-chemical (THMC) coupling in geothermal reservoir is analyzed. The results show that the stable production stage of the three cases is approximately 5 years; however, compared with TH and THMC coupling, the service-life for THM coupling decreased by 1140 days and 332 days, respectively. The mechanical enhanced effects are offset by the chemical precipitation, and the precipitation from SiO ${}_2$ is much larger than the dissolution of calcite.