Geothermics最新文献_第8页

Mineral-mechanical heterogeneity of natural fractures in crystalline rocks: A case study from Utah FORGE 结晶岩天然裂缝的矿物-力学非均质性：以犹他州弗吉格为例

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-19 DOI: 10.1016/j.geothermics.2025.103529

Ettehadi A, Mutume B, Radonjic M

This study presents a multi-scale investigation of mineralogical, structural, and mechanical heterogeneity in a naturally fractured crystalline granite core from Utah FORGE Well 16A (78)-32, retrieved from a depth of 5476.5 ft. Although outside the active injection interval, the sample shares mineralogical and structural features with deeper formations, offering valuable analog insights into fracture behavior in crystalline geothermal reservoirs. X-ray Diffraction (XRD) and Scanning Electron Microscopy with Energy Dispersive Spectroscopy (SEM-EDS) revealed a quartz–feldspar–clay–carbonate assemblage, with quartz and feldspars forming the primary framework and localized enrichment of kaolinite and siderite along fracture interfaces, indicative of hydrothermal alteration. Micro-Computed Tomography (micro-CT) imaging captured a structurally heterogeneous fracture, with mineralized flanks and an unsealed central gap, reflecting episodic sealing and variable fluid–rock interactions. Nanoindentation measurements quantified distinct phase-specific mechanical properties. Quartz (E ≈ 79 GPa, H ≈ 12 GPa) and siderite (E ≈ 154 GPa, H ≈ 7.7–9 GPa) exhibited high stiffness and hardness, supporting mechanical integrity under stress. In contrast, kaolinite (E ≈ 24 GPa, H ≈ 3 GPa) was mechanically weak and ductile. Ankerite, a dominant component of the fracture cement, exhibited moderate stiffness (E ≈ 109 GPa) and hardness (H ≈ 6.6 GPa), contributing to fracture wall reinforcement. Mineral interfaces involving kaolinite demonstrated reduced stiffness and increased plasticity, marking them as potential zones for deformation or reactivation. Based on these micromechanical contrasts, four mineral-mechanical domains were delineated: brittle quartz-rich, semi-brittle feldspar-rich, ductile/reactive clay-carbonate, and mechanically heterogeneous interface zones. This classification provides a framework for translating microscale observations into reservoir-scale behavior predictions. Mechanical domain properties are suited for integration into numerical models of fracture propagation, sealing, and permeability evolution under geothermal conditions. While the study focuses on a shallow core, the findings are transferable to deeper, hotter intervals and contribute to improving fracture stability modeling in high-temperature crystalline systems.

本研究对犹他州FORGE 16A(78)-32井的天然裂缝结晶花岗岩岩心进行了多尺度的矿物学、结构和力学非均质性研究，该岩心深度为5476.5英尺。尽管在活跃注入层段之外，该样品与更深地层具有相同的矿物学和结构特征，为晶体地热储层的裂缝行为提供了有价值的模拟分析。x射线衍射（XRD）和扫描电子显微镜及能谱分析（SEM-EDS）显示石英-长石-粘土-碳酸盐组合，石英和长石构成主要骨架，高岭石和菱铁矿沿裂缝界面局部富集，表明热液蚀变作用。微计算机断层扫描（micro-CT）成像捕获了一条结构上不均匀的裂缝，侧翼矿化，中心空隙未封闭，反映了幕式密封和可变的流体-岩石相互作用。纳米压痕测量量化了不同相的机械性能。石英（E≈79 GPa, H≈12 GPa）和菱铁矿（E≈154 GPa, H≈7.7-9 GPa）具有较高的刚度和硬度，在应力作用下保持机械的完整性。而高岭石（E≈24 GPa, H≈3 GPa）的力学性能较弱，具有延展性。Ankerite是裂缝水泥的主要成分，具有中等的刚度（E≈109 GPa）和硬度（H≈6.6 GPa），有助于裂缝壁的加固。涉及高岭石的矿物界面显示出刚度降低和塑性增加，标志着它们是变形或再激活的潜在区域。基于这些微观力学对比，圈定了4个矿物-力学区域：脆性富石英区、半脆性富长石区、韧性/活性粘土-碳酸盐岩区和机械非均质界面区。这种分类为将微观尺度的观测结果转化为油藏尺度的行为预测提供了一个框架。在地热条件下，力学域特性适合集成到裂缝扩展、密封和渗透率演化的数值模型中。虽然研究主要集中在浅层岩心，但研究结果可应用于更深、更热的层段，并有助于改进高温晶体系统的裂缝稳定性建模。

{"title":"Mineral-mechanical heterogeneity of natural fractures in crystalline rocks: A case study from Utah FORGE","authors":"Ettehadi A, Mutume B, Radonjic M","doi":"10.1016/j.geothermics.2025.103529","DOIUrl":"10.1016/j.geothermics.2025.103529","url":null,"abstract":"<div><div>This study presents a multi-scale investigation of mineralogical, structural, and mechanical heterogeneity in a naturally fractured crystalline granite core from Utah FORGE Well 16A (78)-32, retrieved from a depth of 5476.5 ft. Although outside the active injection interval, the sample shares mineralogical and structural features with deeper formations, offering valuable analog insights into fracture behavior in crystalline geothermal reservoirs. X-ray Diffraction (XRD) and Scanning Electron Microscopy with Energy Dispersive Spectroscopy (SEM-EDS) revealed a quartz–feldspar–clay–carbonate assemblage, with quartz and feldspars forming the primary framework and localized enrichment of kaolinite and siderite along fracture interfaces, indicative of hydrothermal alteration. Micro-Computed Tomography (micro-CT) imaging captured a structurally heterogeneous fracture, with mineralized flanks and an unsealed central gap, reflecting episodic sealing and variable fluid–rock interactions. Nanoindentation measurements quantified distinct phase-specific mechanical properties. Quartz (E ≈ 79 GPa, H ≈ 12 GPa) and siderite (E ≈ 154 GPa, H ≈ 7.7–9 GPa) exhibited high stiffness and hardness, supporting mechanical integrity under stress. In contrast, kaolinite (E ≈ 24 GPa, H ≈ 3 GPa) was mechanically weak and ductile. Ankerite, a dominant component of the fracture cement, exhibited moderate stiffness (E ≈ 109 GPa) and hardness (H ≈ 6.6 GPa), contributing to fracture wall reinforcement. Mineral interfaces involving kaolinite demonstrated reduced stiffness and increased plasticity, marking them as potential zones for deformation or reactivation. Based on these micromechanical contrasts, four mineral-mechanical domains were delineated: brittle quartz-rich, semi-brittle feldspar-rich, ductile/reactive clay-carbonate, and mechanically heterogeneous interface zones. This classification provides a framework for translating microscale observations into reservoir-scale behavior predictions. Mechanical domain properties are suited for integration into numerical models of fracture propagation, sealing, and permeability evolution under geothermal conditions. While the study focuses on a shallow core, the findings are transferable to deeper, hotter intervals and contribute to improving fracture stability modeling in high-temperature crystalline systems.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103529"},"PeriodicalIF":3.9,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145578961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Application of the Monte Carlo method for geothermal resource assessment in low-exploration, high-potential areas: A case study of the Longchuan Basin, Southwest China 蒙特卡罗方法在低勘探高潜力地区地热资源评价中的应用——以龙川盆地为例

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-19 DOI: 10.1016/j.geothermics.2025.103531

Xiaoxue Jiang , Chuanqing Zhu , Fang Xie , Yuanjin Sun , Qian Cao , Dong Sun , Chaohe Fang , Yonghui Huang , Yaodong Xu

The Longchuan Basin, located in the high-temperature geothermal zone of the Mediterranean-Himalayan region, exhibits significant geothermal potential due to intense tectonic activity and a high heat flow background. However, limited exploration and insufficient data have hindered accurate resource assessment. In this study, reliable heat flow values were obtained through systematic drilling temperature measurements and core thermal property analysis, clarifying the spatial distribution of the basin’s temperature field. Using the Monte Carlo volumetric method, key reservoir parameters such as temperature, thickness, and porosity were modeled to quantify the uncertainty of geothermal resource estimates. Results show an average geothermal gradient of 3.87 °C/100 m and surface heat flow ranging from 70 to 90 mW/m². The hydrothermal resource potential is characterized by a P10 of 4.10 × 10¹⁴ KJ, a P50 of 8.87 × 10¹⁴ KJ, and a P90 of 1.59 × 10¹⁵ KJ. For hot dry rock resources, the P10, P50, and P90 values are 1.21 × 10¹⁶ KJ, 1.32 × 10¹⁶ KJ, and 1.43 × 10¹⁶ KJ, respectively. Compared with conventional methods, the Monte Carlo approach effectively captures parameter uncertainty, improving assessment reliability. This study provides a methodological reference for geothermal exploration in under-explored basins with high geothermal potential.

龙川盆地位于地中海-喜马拉雅地区的高温地热带，构造活动强烈，热流背景高，地热潜力突出。然而，有限的勘探和不足的数据阻碍了准确的资源评价。本研究通过系统的钻井测温和岩心热物性分析，获得了可靠的热流值，厘清了盆地温度场的空间分布。利用蒙特卡罗体积法，对温度、厚度和孔隙度等关键储层参数进行建模，量化地热资源估算的不确定性。结果表明，平均地温梯度为3.87°C/100 m，地表热流为70 ~ 90 mW/m²。热液资源潜力的P10值为4.10 × 10¹⁴KJ， P50值为8.87 × 10¹⁴KJ， P90值为1.59 × 10¹⁴KJ。干热岩资源的P10、P50和P90分别为1.21 × 10¹26 KJ、1.32 × 10¹26 KJ和1.43 × 10¹26 KJ。与传统方法相比，蒙特卡罗方法有效地捕捉了参数的不确定性，提高了评估的可靠性。该研究为地热潜力大的未开发盆地的地热勘探提供了方法参考。

{"title":"Application of the Monte Carlo method for geothermal resource assessment in low-exploration, high-potential areas: A case study of the Longchuan Basin, Southwest China","authors":"Xiaoxue Jiang , Chuanqing Zhu , Fang Xie , Yuanjin Sun , Qian Cao , Dong Sun , Chaohe Fang , Yonghui Huang , Yaodong Xu","doi":"10.1016/j.geothermics.2025.103531","DOIUrl":"10.1016/j.geothermics.2025.103531","url":null,"abstract":"<div><div>The Longchuan Basin, located in the high-temperature geothermal zone of the Mediterranean-Himalayan region, exhibits significant geothermal potential due to intense tectonic activity and a high heat flow background. However, limited exploration and insufficient data have hindered accurate resource assessment. In this study, reliable heat flow values were obtained through systematic drilling temperature measurements and core thermal property analysis, clarifying the spatial distribution of the basin’s temperature field. Using the Monte Carlo volumetric method, key reservoir parameters such as temperature, thickness, and porosity were modeled to quantify the uncertainty of geothermal resource estimates. Results show an average geothermal gradient of 3.87 °C/100 m and surface heat flow ranging from 70 to 90 mW/m². The hydrothermal resource potential is characterized by a P10 of 4.10 × 10¹⁴ KJ, a P50 of 8.87 × 10¹⁴ KJ, and a P90 of 1.59 × 10¹⁵ KJ. For hot dry rock resources, the P10, P50, and P90 values are 1.21 × 10¹⁶ KJ, 1.32 × 10¹⁶ KJ, and 1.43 × 10¹⁶ KJ, respectively. Compared with conventional methods, the Monte Carlo approach effectively captures parameter uncertainty, improving assessment reliability. This study provides a methodological reference for geothermal exploration in under-explored basins with high geothermal potential.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103531"},"PeriodicalIF":3.9,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145579086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

On the use of thermal response tests for deep geothermal exploration in urban areas: A case study made on the Greater Montréal (Canada) 关于在城市地区使用热响应测试进行深层地热勘探：以大蒙特兰萨达尔为例研究（加拿大）

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-15 DOI: 10.1016/j.geothermics.2025.103523

Mafalda M. Miranda , Madeline Chapelet , Félix-Antoine Comeau , Jasmin Raymond , J. Christian Dupuis , Philippe Pasquier , Chrystel Dezayes

Exploring deep geothermal resources can be challenging in dense urban environments and therefore a preliminary screening for suitable target areas using existing data from, for example, the geothermal heat pump industry can help de-risk projects at the early exploration phase. Thermal response tests (TRTs) with initial ground temperature measurements can be used to estimate terrestrial heat flux, a metric to indicate suitable areas for deep geothermal development. The greater the heat flux value, the greater the area suitability. This does not replace deep geothermal exploration drilling for a more accurate assessment of the ground temperature but can help target areas of interest and provide a source of unexploited data. This hypothesis was tested in a study aiming to improve knowledge on deep geothermal resources over the Greater Montréal Area (Canada). TRT datasets were analyzed at 22 locations to estimate the terrestrial heat flux. The results were then compared to heat flux estimates made with bottom-hole temperature data and equilibrium temperature profiles from the 6 deep wells available in the area. Both datasets show a certain degree of correlation. To support these findings, a 400 m deep geothermal exploration well was drilled to provide an accurate equilibrium temperature profile in a zone of interest. The data retrieved from this well validated the results and confirmed that, in this setting, TRT can be used as a de-risking tool for deep geothermal exploration.

在人口密集的城市环境中，勘探深层地热资源可能具有挑战性，因此，利用现有数据对合适的目标区域进行初步筛选，例如，地热热泵行业可以帮助降低项目在早期勘探阶段的风险。具有初始地温测量的热响应测试（trt）可用于估计地热通量，这是一种指示适合进行深层地热开发的地区的度量。热流密度值越大，面积适宜性越大。这并不能取代深层地热勘探钻井，以获得更准确的地温评估，但可以帮助目标区域，并提供未开发的数据来源。这个假设在一项研究中得到了验证，这项研究的目的是提高对大蒙塔姆地区（加拿大）深层地热资源的认识。对22个地点的TRT数据集进行了分析，以估计地面热通量。然后将结果与根据该地区6口深井的井底温度数据和平衡温度剖面估算的热流密度进行比较。两个数据集显示出一定程度的相关性。为了支持这些发现，钻了一口400米深的地热勘探井，在感兴趣的区域提供了准确的平衡温度剖面。从该井中获取的数据验证了结果，并证实在这种情况下，TRT可以作为深层地热勘探的降低风险的工具。

{"title":"On the use of thermal response tests for deep geothermal exploration in urban areas: A case study made on the Greater Montréal (Canada)","authors":"Mafalda M. Miranda , Madeline Chapelet , Félix-Antoine Comeau , Jasmin Raymond , J. Christian Dupuis , Philippe Pasquier , Chrystel Dezayes","doi":"10.1016/j.geothermics.2025.103523","DOIUrl":"10.1016/j.geothermics.2025.103523","url":null,"abstract":"<div><div>Exploring deep geothermal resources can be challenging in dense urban environments and therefore a preliminary screening for suitable target areas using existing data from, for example, the geothermal heat pump industry can help de-risk projects at the early exploration phase. Thermal response tests (TRTs) with initial ground temperature measurements can be used to estimate terrestrial heat flux, a metric to indicate suitable areas for deep geothermal development. The greater the heat flux value, the greater the area suitability. This does not replace deep geothermal exploration drilling for a more accurate assessment of the ground temperature but can help target areas of interest and provide a source of unexploited data. This hypothesis was tested in a study aiming to improve knowledge on deep geothermal resources over the Greater Montréal Area (Canada). TRT datasets were analyzed at 22 locations to estimate the terrestrial heat flux. The results were then compared to heat flux estimates made with bottom-hole temperature data and equilibrium temperature profiles from the 6 deep wells available in the area. Both datasets show a certain degree of correlation. To support these findings, a 400 m deep geothermal exploration well was drilled to provide an accurate equilibrium temperature profile in a zone of interest. The data retrieved from this well validated the results and confirmed that, in this setting, TRT can be used as a de-risking tool for deep geothermal exploration.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103523"},"PeriodicalIF":3.9,"publicationDate":"2025-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Thermal response of heterolithic deposits in flooded coal mines: implication for heat storage potential 透水煤矿异质岩矿床的热响应：对储热潜力的启示

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-14 DOI: 10.1016/j.geothermics.2025.103525

Mylene Receveur , Andres Gonzalez Quiros , Alison Monaghan , Vanessa Starcher , Kyle Walker-Verkuil , David Boon , Jeroen van-Hunen

Heat transfer rates are critical to underground heat storage recovery potential and sustainability of thermal abstraction for heating and cooling buildings. A 17-day heat injection – abstraction experiment into a flooded, disused mine working was conducted at the UK Geoenergy Observatory in Glasgow. Analysis of the thermal response of different lithologies intersected by an injection borehole during and after a heat injection experiment is used to quantify the heat exchange between rock mass and circulating mine water. The monitoring data from Distributed Temperature Sensing (DTS) has been analysed and numerical models using COMSOL Multiphysics were developed to characterise the rates and controls on thermal processes during heat injection and recovery. The results suggest the key control of the borehole construction on the temperature change in the first 10 hour of heat injection. In the long term, the thermal response mainly depends on the thermal conductivity of the lithologies. The radial heat transfer reaches a steady charging rate of 23 W/m² and 16 W/m² in the sandstone and clay intervals, respectively, and a maximum of 14 W/m² and 10 W/m² at the start of recovery. This is accompanied by upward heat diffusion/convection from the mine working. This study demonstrates the ability of DTS to identify lithological heterogeneities at a high resolution, and the importance of considering the overburden structure and lithology for thermal storage applications.

传热速率对地下储热回收潜力和采暖制冷建筑的可持续性至关重要。在格拉斯哥的英国地球能源观测站进行了一项为期17天的热注入抽吸实验，该实验是对一个被水淹没的废弃矿井进行的。通过对注热试验期间和试验后注水井相交的不同岩性的热响应分析，量化了岩体与矿井循环水之间的热交换。对分布式温度传感（DTS）的监测数据进行了分析，并使用COMSOL Multiphysics开发了数值模型，以表征热注入和热回收过程中的速率和热过程控制。结果表明，注热前10 h的温度变化是控制钻孔施工的关键。从长期来看，热响应主要取决于岩性的导热性。在砂岩层段和粘土层段，径向换热速率分别为23 W/m2和16 W/m2，在恢复开始时最大换热速率分别为14 W/m2和10 W/m2。这伴随着矿井工作的向上热扩散/对流。这项研究证明了DTS在高分辨率下识别岩性非均质性的能力，以及在储热应用中考虑覆盖层结构和岩性的重要性。

{"title":"Thermal response of heterolithic deposits in flooded coal mines: implication for heat storage potential","authors":"Mylene Receveur , Andres Gonzalez Quiros , Alison Monaghan , Vanessa Starcher , Kyle Walker-Verkuil , David Boon , Jeroen van-Hunen","doi":"10.1016/j.geothermics.2025.103525","DOIUrl":"10.1016/j.geothermics.2025.103525","url":null,"abstract":"<div><div>Heat transfer rates are critical to underground heat storage recovery potential and sustainability of thermal abstraction for heating and cooling buildings. A 17-day heat injection – abstraction experiment into a flooded, disused mine working was conducted at the UK Geoenergy Observatory in Glasgow. Analysis of the thermal response of different lithologies intersected by an injection borehole during and after a heat injection experiment is used to quantify the heat exchange between rock mass and circulating mine water. The monitoring data from Distributed Temperature Sensing (DTS) has been analysed and numerical models using COMSOL Multiphysics were developed to characterise the rates and controls on thermal processes during heat injection and recovery. The results suggest the key control of the borehole construction on the temperature change in the first 10 hour of heat injection. In the long term, the thermal response mainly depends on the thermal conductivity of the lithologies. The radial heat transfer reaches a steady charging rate of 23 W/m<sup>2</sup> and 16 W/m<sup>2</sup> in the sandstone and clay intervals, respectively, and a maximum of 14 W/m<sup>2</sup> and 10 W/m<sup>2</sup> at the start of recovery. This is accompanied by upward heat diffusion/convection from the mine working. This study demonstrates the ability of DTS to identify lithological heterogeneities at a high resolution, and the importance of considering the overburden structure and lithology for thermal storage applications.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103525"},"PeriodicalIF":3.9,"publicationDate":"2025-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predictive analytics and statistical time series imputation of subsurface temperatures for a campus-scale geothermal exchange field 校园地热交换场地下温度的预测分析与统计时间序列反演

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-12 DOI: 10.1016/j.geothermics.2025.103528

Shubham Dutt Attri , Mingxue Jiang , James M. Tinjum , Dante Fratta , Orhun Aydin

This study examines the dynamics of a district-scale geothermal heat exchange (GHX) field via the application of statistical time series models for predictive analysis and imputation of subsurface temperatures measured with fiber-optic distributed temperature sensing (FO-DTS). High-resolution FO-DTS technology was used to monitor subsurface temperatures in a district-scale GHX field (2596 152-m-deep heat exchange wells with a thermal power of 26.6 MW) embedded within multiple geological formations. A data gap from 2019 to 2021, resulting from operational and COVID-19-related disruptions, necessitated the use of statistical modeling techniques for imputation and analysis. The analysis begins with a description and statistical characterization of the data, followed by a detailed examination of the ARIMA model for subsurface temperature modeling. However, because ARIMA was unable to efficiently and accurately capture the complex seasonal patterns in the dataset, the Holt–Winters Exponential Smoothing (HW-ES) method was adopted. HW-ES effectively modeled both trend and seasonality, achieving a Root Mean Square Error (RMSE) of 0.156 in short-term predictions while demonstrating robustness in replication of long-term seasonal oscillations. A hybrid bidirectional approach, combining forward and backward rolling HW-ES models, from the past and the future data, respectively, integrated through weighted averaging, was developed to address the data gap, ensuring a smooth and continuous imputed series. The proposed predictive framework generates probabilistic forecasts with upper and lower bounds, enabling scenario analysis to support long-term planning and operational strategies for ground source heat pump systems. By leveraging operating data, it enhances understanding of subsurface thermal behavior, addresses data challenges, and facilitates more effective field management and geothermal energy optimization. This paper also talks about the gaps and provides future recommendations for more advanced analysis.

本研究通过应用统计时间序列模型对光纤分布式温度传感（FO-DTS）测量的地下温度进行预测分析和计算，研究了区域尺度地热交换（GHX）场的动力学。高分辨率FO-DTS技术用于监测区域尺度的GHX油田（2596口152米深热交换井，热功率26.6 MW）的地下温度，这些热交换井嵌入多个地质构造中。由于运营和covid -19相关中断，2019年至2021年期间存在数据缺口，因此有必要使用统计建模技术进行估算和分析。分析从数据的描述和统计特征开始，然后详细检查ARIMA模型的地下温度建模。然而，由于ARIMA无法高效准确地捕获数据集中复杂的季节模式，因此采用了Holt-Winters指数平滑（HW-ES）方法。HW-ES有效地模拟了趋势和季节性，在短期预测中实现了0.156的均方根误差（RMSE），同时在长期季节性振荡的复制中显示出鲁棒性。提出了一种混合双向方法，将过去和未来数据的前向和后向滚动HW-ES模型相结合，通过加权平均进行整合，以解决数据差距问题，确保平稳连续的输入序列。提出的预测框架生成具有上限和下限的概率预测，使情景分析能够支持地源热泵系统的长期规划和运营策略。通过利用作业数据，该系统增强了对地下热行为的理解，解决了数据挑战，促进了更有效的现场管理和地热能优化。本文还讨论了差距，并为将来进行更深入的分析提供了建议。

{"title":"Predictive analytics and statistical time series imputation of subsurface temperatures for a campus-scale geothermal exchange field","authors":"Shubham Dutt Attri , Mingxue Jiang , James M. Tinjum , Dante Fratta , Orhun Aydin","doi":"10.1016/j.geothermics.2025.103528","DOIUrl":"10.1016/j.geothermics.2025.103528","url":null,"abstract":"<div><div>This study examines the dynamics of a district-scale geothermal heat exchange (GHX) field via the application of statistical time series models for predictive analysis and imputation of subsurface temperatures measured with fiber-optic distributed temperature sensing (FO-DTS). High-resolution FO-DTS technology was used to monitor subsurface temperatures in a district-scale GHX field (2596 152-m-deep heat exchange wells with a thermal power of 26.6 MW) embedded within multiple geological formations. A data gap from 2019 to 2021, resulting from operational and COVID-19-related disruptions, necessitated the use of statistical modeling techniques for imputation and analysis. The analysis begins with a description and statistical characterization of the data, followed by a detailed examination of the ARIMA model for subsurface temperature modeling. However, because ARIMA was unable to efficiently and accurately capture the complex seasonal patterns in the dataset, the Holt–Winters Exponential Smoothing (HW-ES) method was adopted. HW-ES effectively modeled both trend and seasonality, achieving a Root Mean Square Error (RMSE) of 0.156 in short-term predictions while demonstrating robustness in replication of long-term seasonal oscillations. A hybrid bidirectional approach, combining forward and backward rolling HW-ES models, from the past and the future data, respectively, integrated through weighted averaging, was developed to address the data gap, ensuring a smooth and continuous imputed series. The proposed predictive framework generates probabilistic forecasts with upper and lower bounds, enabling scenario analysis to support long-term planning and operational strategies for ground source heat pump systems. By leveraging operating data, it enhances understanding of subsurface thermal behavior, addresses data challenges, and facilitates more effective field management and geothermal energy optimization. This paper also talks about the gaps and provides future recommendations for more advanced analysis.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103528"},"PeriodicalIF":3.9,"publicationDate":"2025-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multiple-level fault controlled hydrothermal system in Huangshadong, Southeast China: Insights from coupled geophysical and geochemical investigations 黄沙洞多层断裂控制的热液系统：地球物理和地球化学耦合调查的启示

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-12 DOI: 10.1016/j.geothermics.2025.103526

Lu Liu , Xianan Qin , Zheng’an Wei , Min Zhang , Honglei Shi , Xiting Long , Wenjing Lin , Mingxiang Qiu , Guiling Wang , Shihua Qi , Fei Liu

The Huangshadong geothermal field exemplifies fault-controlled hydrothermal systems, where structural features dictate heat convergence and fluid pathways. Integrated geophysical, hydrogeochemical, and drilling data reveal a multiple-level fault system governs fluid dynamics, deep-seated faults (F1) channel ascending thermal fluids, while secondary faults (F2, F3) regulate shallow dispersal. Temperature fields show localized highs at faults intersections. A shallow aquifer with low-resistivity zones enabling thermal fluid diffusion and mixing with cold groundwater, overlying high-resistivity concealed granite. F4 acts as a barrier, evidenced by sharp thermal and geochemical gradients. The deep thermal water is primarily from meteoric water and mixes with shallow cold water (81–86 %) in the fault zone. Combined with the actual reservoir pressure, the reservoir temperature is estimated to be approximately 167.9–187.4°C and the circulation depths are 5.2–5.9 km. Sr isotope and ion analysis indicate that evaporites and silicates in granites are primarily involved in the water-rock intereaction. Noble gas isotopes confirm a crustal heat source, with mantle contributions of <6 %.

黄沙洞地热田是断裂控制的热液系统，其构造特征决定了热汇聚和流体通道。综合地球物理、水文地球化学和钻井数据表明，一个多级断裂系统控制着流体动力学，深部断裂（F1）引导上行热流体，而次级断裂（F2、F3）调节浅层扩散。温度场在断层交叉处显示局部高。具有低电阻率带的浅层含水层，使热流体扩散并与冷地下水混合，覆盖在高电阻率隐伏花岗岩上。F4是一个屏障，有明显的热梯度和地球化学梯度。深层热水主要来自大气降水，在断裂带与浅层冷水混合（81% ~ 86%）。结合储层实际压力，估计储层温度约为167.9 ~ 187.4℃，循环深度为5.2 ~ 5.9 km。Sr同位素和离子分析表明，花岗岩中的蒸发岩和硅酸盐主要参与了水岩相互作用。稀有气体同位素证实了地壳热源，地幔贡献了6%。

{"title":"Multiple-level fault controlled hydrothermal system in Huangshadong, Southeast China: Insights from coupled geophysical and geochemical investigations","authors":"Lu Liu , Xianan Qin , Zheng’an Wei , Min Zhang , Honglei Shi , Xiting Long , Wenjing Lin , Mingxiang Qiu , Guiling Wang , Shihua Qi , Fei Liu","doi":"10.1016/j.geothermics.2025.103526","DOIUrl":"10.1016/j.geothermics.2025.103526","url":null,"abstract":"<div><div>The Huangshadong geothermal field exemplifies fault-controlled hydrothermal systems, where structural features dictate heat convergence and fluid pathways. Integrated geophysical, hydrogeochemical, and drilling data reveal a multiple-level fault system governs fluid dynamics, deep-seated faults (F1) channel ascending thermal fluids, while secondary faults (F2, F3) regulate shallow dispersal. Temperature fields show localized highs at faults intersections. A shallow aquifer with low-resistivity zones enabling thermal fluid diffusion and mixing with cold groundwater, overlying high-resistivity concealed granite. F4 acts as a barrier, evidenced by sharp thermal and geochemical gradients. The deep thermal water is primarily from meteoric water and mixes with shallow cold water (81–86 %) in the fault zone. Combined with the actual reservoir pressure, the reservoir temperature is estimated to be approximately 167.9–187.4°C and the circulation depths are 5.2–5.9 km. Sr isotope and ion analysis indicate that evaporites and silicates in granites are primarily involved in the water-rock intereaction. Noble gas isotopes confirm a crustal heat source, with mantle contributions of <6 %.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103526"},"PeriodicalIF":3.9,"publicationDate":"2025-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Experimental study on the evolution of pore structure in carbonate geothermal reservoirs during matrix acidification under alternating temperature loads 交变温度作用下基质酸化过程中碳酸盐地热储层孔隙结构演化的实验研究

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-11 DOI: 10.1016/j.geothermics.2025.103527

Chunlei Wei , Daobing Wang , Jinsheng Sun , Xianbin Huang , Jintang Wang , Fan Liu , Chang Liu

Due to the strong heterogeneity commonly observed in carbonate geothermal reservoirs, conventional matrix acidizing techniques often yield suboptimal stimulation results. To address this issue, this study proposes an alternating thermal loading synergistic acidizing technique to enhance the matrix acidizing performance of carbonate reservoirs. To elucidate the improvement mechanism under this technique, spontaneous imbibition and conductivity monitoring experiments were conducted to quantitatively characterize the evolution of pore structure and fluid migration behavior in rock samples. The results indicate that alternating thermal loading significantly enhances the acidizing effect, increasing the unit-area imbibition mass of the core by 16 %, while porosity and permeability are improved by 3.7 and 2.3 times, respectively. Conductivity monitoring of the imbibition solution further reveals that thermal cycling markedly accelerates the diffusion rate of mineral ions from the rock matrix into the solution. This study confirms the synergistic enhancement mechanism between cyclic thermal stress and acid dissolution, providing a new approach and technical pathway for the efficient stimulation of highly heterogeneous geothermal reservoirs.

由于碳酸盐岩地热储层普遍存在很强的非均质性，传统的基质酸化技术往往不能获得最佳的增产效果。为了解决这一问题，本研究提出了一种交替热负荷协同酸化技术，以提高碳酸盐岩储层基质的酸化性能。为了阐明该技术的改善机理，开展了自吸和导电性监测实验，定量表征了岩石样品中孔隙结构的演化和流体运移行为。结果表明：交替热负荷显著提高了酸化效果，岩心单位面积吸胀质量提高了16%，孔隙度和渗透率分别提高了3.7倍和2.3倍。对渗吸溶液的电导率监测进一步表明，热循环显著加快了矿物离子从岩石基质向溶液的扩散速度。该研究证实了循环热应力与酸溶作用的协同增强机制，为高非均质地热储层的高效改造提供了新的思路和技术途径。

{"title":"Experimental study on the evolution of pore structure in carbonate geothermal reservoirs during matrix acidification under alternating temperature loads","authors":"Chunlei Wei , Daobing Wang , Jinsheng Sun , Xianbin Huang , Jintang Wang , Fan Liu , Chang Liu","doi":"10.1016/j.geothermics.2025.103527","DOIUrl":"10.1016/j.geothermics.2025.103527","url":null,"abstract":"<div><div>Due to the strong heterogeneity commonly observed in carbonate geothermal reservoirs, conventional matrix acidizing techniques often yield suboptimal stimulation results. To address this issue, this study proposes an alternating thermal loading synergistic acidizing technique to enhance the matrix acidizing performance of carbonate reservoirs. To elucidate the improvement mechanism under this technique, spontaneous imbibition and conductivity monitoring experiments were conducted to quantitatively characterize the evolution of pore structure and fluid migration behavior in rock samples. The results indicate that alternating thermal loading significantly enhances the acidizing effect, increasing the unit-area imbibition mass of the core by 16 %, while porosity and permeability are improved by 3.7 and 2.3 times, respectively. Conductivity monitoring of the imbibition solution further reveals that thermal cycling markedly accelerates the diffusion rate of mineral ions from the rock matrix into the solution. This study confirms the synergistic enhancement mechanism between cyclic thermal stress and acid dissolution, providing a new approach and technical pathway for the efficient stimulation of highly heterogeneous geothermal reservoirs.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103527"},"PeriodicalIF":3.9,"publicationDate":"2025-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reconstructing regional subsurface thermal structure of the Okavango rift zone, northern Botswana, and its low enthalpy geothermal potential 博茨瓦纳北部奥卡万戈裂谷带区域地下热结构重建及其低焓地热潜力

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-11 DOI: 10.1016/j.geothermics.2025.103519

Izzul Qudsi , Daffa Arrofi , Luis Sebastian Gallegos , Ahmed Salem

Previous studies identified the potential of the geothermal area within the Okavango rift zone in Northern Botswana due to the southward extension of The Eastern African Rift System (EARS). Although surface heat flow has been estimated in the area, including radiogenic heat production in the upper crust improves the accuracy of the calculations. The presence of moderate surface heat flow suggests that low-enthalpy geothermal resources could be used for direct-use applications such as heating, agriculture, and other sustainable energy solutions. This study evaluates geothermal potential in the Okavango Rift Zone by estimating the depth of the required temperature threshold using a constrained heat flow model. Regional thermal structures were modeled using constrains including Curie Point Depth, Moho depth, thermal conductivity of the layers, and radiogenic heat production, were constructed to achieve the goals. We iteratively adjusted the surface heat flow to determine the best fit of the model by comparing the temperature at CPD and Moho with reference values and predictions from the cooling plate model. The model proposed in this study minimized the surface heat flow estimation error, matching wellbore values within ± 2 mW/m². Based on our results, Maun City possibly has the potential for direct use or binary geothermal applications with a depth of <1.4 km and 2.3 km, respectively. Integrating radiogenic heat production into the geotherm model, improves the surface heat flow estimation and provides a robust methodology applicable to other geothermal fields with similar tectonic systems.

先前的研究发现，由于东非裂谷系统（EARS）的向南延伸，博茨瓦纳北部奥卡万戈裂谷带的地热区域潜力巨大。虽然已经估计了该地区的地表热流，但包括上地壳的放射性成因热生产提高了计算的准确性。地表热流适中的存在表明，低焓地热资源可用于供热、农业和其他可持续能源解决方案等直接利用应用。本研究利用约束热流模型估算了所需温度阈值的深度，从而评估了奥卡万戈裂谷带的地热潜力。利用居里点深度、莫霍深度、地层热导率和放射成因产热等约束条件对区域热结构进行建模，以实现目标。通过将CPD和Moho温度与参考值和冷却板模型的预测值进行比较，反复调整地表热流以确定模型的最佳拟合。本研究提出的模型最大限度地减少了表面热流估计误差，与井眼值匹配在±2 mW/m2以内。根据我们的研究结果，Maun City可能具有直接利用或二元地热应用的潜力，深度分别为1.4 km和2.3 km。将放射性成因产热整合到地热模型中，可以改善地表热流估算，并为具有类似构造体系的其他地热田提供一种可靠的方法。

{"title":"Reconstructing regional subsurface thermal structure of the Okavango rift zone, northern Botswana, and its low enthalpy geothermal potential","authors":"Izzul Qudsi , Daffa Arrofi , Luis Sebastian Gallegos , Ahmed Salem","doi":"10.1016/j.geothermics.2025.103519","DOIUrl":"10.1016/j.geothermics.2025.103519","url":null,"abstract":"<div><div>Previous studies identified the potential of the geothermal area within the Okavango rift zone in Northern Botswana due to the southward extension of The Eastern African Rift System (EARS). Although surface heat flow has been estimated in the area, including radiogenic heat production in the upper crust improves the accuracy of the calculations. The presence of moderate surface heat flow suggests that low-enthalpy geothermal resources could be used for direct-use applications such as heating, agriculture, and other sustainable energy solutions. This study evaluates geothermal potential in the Okavango Rift Zone by estimating the depth of the required temperature threshold using a constrained heat flow model. Regional thermal structures were modeled using constrains including Curie Point Depth, Moho depth, thermal conductivity of the layers, and radiogenic heat production, were constructed to achieve the goals. We iteratively adjusted the surface heat flow to determine the best fit of the model by comparing the temperature at CPD and Moho with reference values and predictions from the cooling plate model. The model proposed in this study minimized the surface heat flow estimation error, matching wellbore values within ± 2 mW/m<sup>2</sup>. Based on our results, Maun City possibly has the potential for direct use or binary geothermal applications with a depth of <1.4 km and 2.3 km, respectively. Integrating radiogenic heat production into the geotherm model, improves the surface heat flow estimation and provides a robust methodology applicable to other geothermal fields with similar tectonic systems.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103519"},"PeriodicalIF":3.9,"publicationDate":"2025-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High-resolution mapping of shallow geothermal resources in Spain through multivariate spatial analysis 通过多元空间分析西班牙浅层地热资源的高分辨率制图

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-08 DOI: 10.1016/j.geothermics.2025.103524

Sergio Alejandro Camargo Vargas, Cristina Sáez Blázquez, Daniel Herranz Herranz, Miguel Ángel Maté-González

Identifying areas with potential for low-enthalpy geothermal energy exploitation is a key step toward sustainable and decentralized energy planning, particularly in countries like Spain, where seasonal high temperatures and marked thermal contrasts influence subsurface heat dynamics. This study presents a nationwide assessment of shallow geothermal potential using a multivariate spatial analysis approach. The methodology integrates key geological and thermal variables, including subsurface temperature gradients at different levels, lithological composition and other influential properties such as thermal conductivity, permeability, and volumetric heat capacity. These parameters were processed through normalization, thematic weighting, map algebra, and spatial classification methods within a GIS-based framework. As a result, geothermal favorability indices were generated, allowing for the identification and mapping of areas with varying levels of potential for low and very low-enthalpy applications. The resulting geothermal potential map was validated through comparison with existing geological information and previously published studies. The outcomes confirm the reliability and applicability of the multivariate approach as a preliminary tool for identifying shallow geothermal resources. This contributes not only to a better understanding of the geothermal landscape in Spain but also to supporting informed decision-making in regional energy planning and sustainable land management strategies.

确定具有低焓地热能源开发潜力的地区是实现可持续和分散能源规划的关键步骤，特别是在西班牙等国家，季节性高温和明显的热差异影响地下热动力学。本文采用多元空间分析方法对全国浅层地热潜力进行了评价。该方法整合了关键的地质和热变量，包括不同层次的地下温度梯度、岩性组成和其他影响性质，如导热系数、渗透率和体积热容。在基于gis的框架内，通过归一化、专题加权、地图代数和空间分类等方法对这些参数进行处理。因此，生成了地热有利指数，从而可以识别和绘制具有不同低焓和极低焓应用潜力的区域。通过与现有地质信息和先前发表的研究结果进行比较，验证了所得的地热潜力图。结果证实了多元方法作为识别浅层地热资源的初步工具的可靠性和适用性。这不仅有助于更好地了解西班牙的地热景观，而且有助于在区域能源规划和可持续土地管理战略方面做出明智的决策。

{"title":"High-resolution mapping of shallow geothermal resources in Spain through multivariate spatial analysis","authors":"Sergio Alejandro Camargo Vargas, Cristina Sáez Blázquez, Daniel Herranz Herranz, Miguel Ángel Maté-González","doi":"10.1016/j.geothermics.2025.103524","DOIUrl":"10.1016/j.geothermics.2025.103524","url":null,"abstract":"<div><div>Identifying areas with potential for low-enthalpy geothermal energy exploitation is a key step toward sustainable and decentralized energy planning, particularly in countries like Spain, where seasonal high temperatures and marked thermal contrasts influence subsurface heat dynamics. This study presents a nationwide assessment of shallow geothermal potential using a multivariate spatial analysis approach. The methodology integrates key geological and thermal variables, including subsurface temperature gradients at different levels, lithological composition and other influential properties such as thermal conductivity, permeability, and volumetric heat capacity. These parameters were processed through normalization, thematic weighting, map algebra, and spatial classification methods within a GIS-based framework. As a result, geothermal favorability indices were generated, allowing for the identification and mapping of areas with varying levels of potential for low and very low-enthalpy applications. The resulting geothermal potential map was validated through comparison with existing geological information and previously published studies. The outcomes confirm the reliability and applicability of the multivariate approach as a preliminary tool for identifying shallow geothermal resources. This contributes not only to a better understanding of the geothermal landscape in Spain but also to supporting informed decision-making in regional energy planning and sustainable land management strategies.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103524"},"PeriodicalIF":3.9,"publicationDate":"2025-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145474468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A machine learning hybrid model for heat transfer per meter prediction of energy pipe pile 能源管桩每米换热预测的机器学习混合模型

IF 3.9 2区工程技术 Q3 ENERGY & FUELS

Geothermics

Pub Date : 2025-11-06 DOI: 10.1016/j.geothermics.2025.103521

Siming Dong , Chenglong Wang , Hanlong Liu , Abdelmalek Bouazza , Xuanming Ding , Gangqiang Kong

This study proposes a hybrid framework that combines interpretable machine learning, multi-physical field simulation, and Bayesian optimization to address the challenges posed by diverse structural and geological conditions, as well as the lack of interpretability in predicting the heat transfer performance of energy pipe piles. A three-dimensional transient heat transfer numerical model was used to generate a comprehensive dataset. This dataset includes variations in key heat transfer parameters (pipe pile structure, fluid parameters, and material thermal properties). A backpropagation neural network (BPNN) was used for prediction, with the impact of various hyperparameters analyzed. Model performance was significantly improved through Bayesian optimization, which automatically identified the most suitable hyperparameter combinations. To improve interpretability, the SHAP (Shapley additive explanations) method was used to quantify the contribution of each input feature to the model’s output. Among all features, the inlet water temperature had the highest influence, accounting for 23.4 % of the prediction. SHAP also provided insights into how each feature affects the model’s decision-making process.

本研究提出了一个结合可解释机器学习、多物理场模拟和贝叶斯优化的混合框架，以解决不同结构和地质条件带来的挑战，以及预测能源管桩传热性能缺乏可解释性的问题。采用三维瞬态传热数值模型生成了一个综合数据集。该数据集包括关键传热参数（管桩结构、流体参数和材料热性能）的变化。采用反向传播神经网络（BPNN）进行预测，分析了各种超参数的影响。通过贝叶斯优化，自动识别出最合适的超参数组合，显著提高了模型性能。为了提高可解释性，使用Shapley加性解释（Shapley additive explanation）方法来量化每个输入特征对模型输出的贡献。其中，进口水温对预测的影响最大，占预测的23.4%。SHAP还提供了对每个特征如何影响模型决策过程的见解。

{"title":"A machine learning hybrid model for heat transfer per meter prediction of energy pipe pile","authors":"Siming Dong , Chenglong Wang , Hanlong Liu , Abdelmalek Bouazza , Xuanming Ding , Gangqiang Kong","doi":"10.1016/j.geothermics.2025.103521","DOIUrl":"10.1016/j.geothermics.2025.103521","url":null,"abstract":"<div><div>This study proposes a hybrid framework that combines interpretable machine learning, multi-physical field simulation, and Bayesian optimization to address the challenges posed by diverse structural and geological conditions, as well as the lack of interpretability in predicting the heat transfer performance of energy pipe piles. A three-dimensional transient heat transfer numerical model was used to generate a comprehensive dataset. This dataset includes variations in key heat transfer parameters (pipe pile structure, fluid parameters, and material thermal properties). A backpropagation neural network (BPNN) was used for prediction, with the impact of various hyperparameters analyzed. Model performance was significantly improved through Bayesian optimization, which automatically identified the most suitable hyperparameter combinations. To improve interpretability, the SHAP (Shapley additive explanations) method was used to quantify the contribution of each input feature to the model’s output. Among all features, the inlet water temperature had the highest influence, accounting for 23.4 % of the prediction. SHAP also provided insights into how each feature affects the model’s decision-making process.</div></div>","PeriodicalId":55095,"journal":{"name":"Geothermics","volume":"134 ","pages":"Article 103521"},"PeriodicalIF":3.9,"publicationDate":"2025-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145474469","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0