A robust optimization approach of well placement for doublet in heterogeneous geothermal reservoirs using random forest technique and genetic algorithm

Wang, Jiacheng; Zhao, Zhihong; Liu, Guihong; Xu, Haoran

doi:10.1016/j.energy.2022.124427

Cited by 25 publications

(6 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…47 ), so that the value of 190 W/( C) is still a reasonably good estimate for a homogeneous heat transfer coefficient for comparison. Further, it is a known problem in geothermal energy production that parts of the reservoir remain poorly activated and well placement is a crucial parameter to minimize this loss 21 , 22 . To demonstrate the effect of velocity-dependent heat transfer in fractures on reservoir design and well placement, we switch to borehole configuration 2 once the production temperature drops below 110 °C, a reasonable limit for electric power generation from geothermal energy (Fig.…”

Section: Resultsmentioning

confidence: 99%

Velocity-dependent heat transfer controls temperature in fracture networks

Heinze

Pastore

2023

Nat Commun

View full text Add to dashboard Cite

Heat transfer between a fluid and the surrounding rock in the subsurface is a crucial process not only, but most obviously, in geothermal systems. Heat transfer is described by Newton’s law of cooling, relating the heat transferred to a coefficient, the specific surface area, and the temperature difference between rock and fluid. However, parameterizing the heat transfer coefficient in fracture networks poses a major challenge. Here we show that within a fracture network the heat transfer coefficient is strongly heterogeneous but that laboratory single fracture experiments can provide a reasonable estimate in dependence of flow rate. We investigate the distribution of the heat transfer coefficient experimentally as well as numerically and analyze the heat transfer at individual fractures. Our results improve the prediction of temperatures in engineered and natural geothermal systems and allow sustainable management and design of reservoirs considering the role of individual fractures.

show abstract

Section: Resultsmentioning

confidence: 99%

Velocity-dependent heat transfer controls temperature in fracture networks

Heinze

Pastore

2023

Nat Commun

View full text Add to dashboard Cite

show abstract

“…The random forest method is a machine learning algorithm that is flexible and easy to use. As one of the classification and regression algorithms, the random forest algorithm uses a decision tree as the basic decision unit and generates multiple sample sets randomly by repeated sampling of samples [41][42][43]. The computation of random forest is divided into six main steps, and its decision process is shown in Figure 26.…”

Section: Analysis Of Influencing Factor Weights Based On Random Fores...mentioning

confidence: 99%

Investigation of Key Controlling Factors and Applicability Boundaries of Natural Gas Injection for Shale Oil Development: A Case Study of Chang 7 Reservoir, Ordos Basin, China

Xie,

Song,

Cao

et al. 2023

Energies

View full text Add to dashboard Cite

The development of shale oil often encounters a series of challenges, such as insufficient natural energy, rapid declines in production, and low oil recovery. Given its wide availability and low cost, natural gas serves as an ideal injection medium to enhance shale oil recovery. Based on the foundational data from Chang 7 block X in the Ordos Basin and considering the influence of reservoir stress sensitivity, this study uses numerical simulation methods in conjunction with the random forest algorithm to examine gas flooding in shale oil. This research aims to identify the key factors influencing the effectiveness of natural gas flooding for shale oil and to establish a chart for the adaptive natural gas flooding development boundary. The findings suggest that, for reservoirs with permeability less than 0.2 × 10−3 μm2 and oil saturation below 55%, it is advisable to adopt longer fracture lengths during hydraulic fracturing to achieve better development results. After 10 years of gas flooding, continuing the use of this method is not recommended, and adjusting of the development strategy becomes essential. For extended development periods, avoiding long-length fractures is crucial to mitigate gas channeling and ensure the effectiveness of the development process. This study offers theoretical guidance and technical support for developing shale reservoirs with natural gas flooding.

show abstract

“…Geothermal energy is a type of renewable and clean energy, and it widely exists beneath the earth's crust with capacity equivalent to 4 × 10 13 Watts of energy [1,2]. Hence, it has remarkable potential to be an alternative to the fossil fuels.…”

Section: Introductionmentioning

confidence: 99%

“…Existing numerical simulation models help geothermal engineers to predict the thermal front movement and production temperature behavior both in EGS and HAS. Thermo-hydro (TH), thermo-hydro-mechanical (THM), thermo-hydro-mechanical-chemical (THMC) simulation models were developed for geothermal reservoir simulations [13][14][15][16][17][18][19][20][21]. However, TH models may underestimate the production temperature at lower injection rates [19].…”

Section: Introductionmentioning

confidence: 99%