Designing and optimizing the control schemes of geothermal energy systems is a challenging and time-consuming work due to the vast parameter space and computationally intensive simulations. Canonical evolutionary optimization approaches are laborious, slow to converge, and may not provide optimal well-control scheme for geothermal energy systems. To tackle these issues, this work reports a machine learning-guided real-time flow control optimization for enhanced geothermal systems. This approach fully leverages existing data throughout the optimization phase by creating multi-fidelity surrogate models, which comprise coarse and fine models. The coarse model strategically selects a subset of variables to develop a low-fidelity representation, while the fine model utilizes all available variables to construct a high-fidelity surrogate. Knowledge transfer from coarse surrogate can guide the fine surrogate search into a promising subspace. Active learning technique is further leveraged to improve the accuracy of surrogate by iteratively querying the most informative data points. To evaluate effectiveness of the proposed approach, benchmark function suites and two fractured geothermal energy systems are employed in comparison with conventional evolutionary algorithms and advanced surrogate-assisted methods. The results illustrate the capability of the workflow to enhance the efficiency and effectiveness of real-time decision making. This workflow paves a new path for complex and computationally intensive design optimization problems. Document Type: Original article Cited as: Chen, G., Jiao, J. J., Wang, Z., Dai, Q. Multi-fidelity machine learning with knowledge transfer enhances geothermal energy system design and optimization. Advances in Geo-Energy Research, 2025, 16(3): 244-259. https://doi.org/10.46690/ager.2025.06.05
The fracturing technique is widely used in many fields. Fracture has a greater impact on the movement of fluids in formations. Knowing information about a fracture is key to judging its effect, but detailed information about complex fracture networks is difficult to obtain. In this paper, we propose a new method to describe the shape of a complex fracture network. This method is based on microseismic results and uses the L-system to establish a method for characterizing a complex fracture network. The method also combines with decomposition to construct a new method called the multiobjective fracture network inversion algorithm based on decomposition (MOFNIAD). The coverage of microseismic monitoring results and the degree of fitting of production data are the two objective functions of the inversion fracture network. The multiobjective fracture network inversion algorithm can be optimized to obtain multiple optimal solutions that meet different target weights. Therefore, this paper established a multischeme decision method that approached the ideal solution, sorting technology and AHP to provide theoretical guidance for finding a more ideal fracture network. According to the error of microseismic monitoring results, we established two cases of fracture to verify the proposed method. Judging from the results of the examples, the fracture network finally obtained was similar to actual fractures.