Research on a comprehensive “sweet spot” evaluation method for shale gas geotechnical engineering based on domain knowledge fusion

ZENG FANHUI; YANG ENHE; ZHENG BINTAO; ZHANG YU; GUO JIANCHUN

doi:10.13809/j.cnki.cn32-1825/te.20260077

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Research on a comprehensive “sweet spot” evaluation method for shale gas geotechnical engineering based on domain knowledge fusion

更新时间：2026-09-23

- Research on a comprehensive “sweet spot” evaluation method for shale gas geotechnical engineering based on domain knowledge fusion
- Petroleum Reservoir Evaluation and Development Pages: 1-11(2026)
- 作者机构：
  
  1.西南石油大学油气藏地质及开发工程全国重点实验室，四川成都 610500
  2.中国石化胜利油田分公司石油工程技术研究院，山东东营 257000
- 作者简介：
- 基金信息：
- DOI：10.13809/j.cnki.cn32-1825/te.20260077
  CLC： TE371
- Received：05 June 2026，
  
  Online First：23 September 2026，
- 稿件说明：
移动端阅览
曾凡辉, 杨恩和, 郑彬涛, 等. 领域知识融合的页岩气地质工程综合“甜点”评价方法研究[J]. 油气藏评价与开发, XXXX, XX(XX): 1-11.

Zeng Fanhui, Yang Enhe, Zheng Bintao, et al. Research on a comprehensive “sweet spot” evaluation method for shale gas geotechnical engineering based on domain knowledge fusion[J]. Petroleum Reservoir Evaluation and Development, XXXX, XX(XX): 1-11.
曾凡辉, 杨恩和, 郑彬涛, 等. 领域知识融合的页岩气地质工程综合“甜点”评价方法研究[J]. 油气藏评价与开发, XXXX, XX(XX): 1-11. DOI： 10.13809/j.cnki.cn32-1825/te.20260077.

Zeng Fanhui, Yang Enhe, Zheng Bintao, et al. Research on a comprehensive “sweet spot” evaluation method for shale gas geotechnical engineering based on domain knowledge fusion[J]. Petroleum Reservoir Evaluation and Development, XXXX, XX(XX): 1-11. DOI： 10.13809/j.cnki.cn32-1825/te.20260077.

摘要

“甜点”准确评价是页岩气高效开发的核心环节。基于物理模型的“甜点”评价方法成本高、精度低，难以满足低成本开发诉求；监督机器学习评价方法需大量标注数据，难以适用于邻井差异大等缺乏标注样本的“甜点”评价；无监督机器学习评价结果只能对“甜点”分类，无法直接判断“甜点”好坏，且未考虑指标权重差异，难以准确反映地质与工程参数的真实贡献。针对以上问题，研究在逼近理想解排序法（TOPSIS）基础上，建立了融合领域知识的页岩气地质工程综合“甜点”半监督评价模型。首先，基于物理规律、专家经验等领域知识确定评价指标、划分指标类型，采用Pearson相关系数法进行验证，通过嵌入半监督机器学习生成虚拟约束标签，构建半监督“甜点”评价数据集。然后，进一步通过TOPSIS方法和欧式距离理论，结合多目标规划算法确定各评价指标权重。最终，建立融合领域知识的综合“甜点”半监督评价模型。微地震事件数、米采气指数与研究模型评价结果的

较最优模型分别提高了5.58%、10.51%。实例应用结果显示：JS1井“甜点”主控因素为总含气量、孔隙度、泊松比，累计权重为41.21%；微地震事件数、米采气指数与评价结果的

均超过0.85。模型能实现页岩气水平井“甜点”准确评价，为地质条件差异大、新井开发等缺少标注数据的开发场景提供了重要参考

与技术支撑。

Abstract

Accurate evaluation of sweet spots is critical for the efficient development of shale gas. Conventional sweet spot evaluation methods based on physical models suffer from high cost and low accuracy

which can hardly meet the demand for low-cost development. Supervised machine learning-based evaluation methods require a large volume of labeled data

making them inapplicable to sweet spot evaluation scenarios with insufficient labeled samples such as significant differences among adjacent wells. Unsupervised machine learning methods can only realize the classification of sweet spots

failing to directly distinguish the quality of sweet spots; moreover

they ignore the weight differences of evaluation indicators

making it difficult to accurately reflect the actual contributions of geological and engineering parameters. To address these issues

this study develops a semi-supervised comprehensive evaluation model for shale gas geological and engineering sweet spots integrating domain knowledge based on the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS). Firstly

evaluation indicators are determined and classified based on domain knowledge including physical laws and expert experience

and the Pearson correlation coefficient method is adopted for verification

and virtual constraint labels are generated by embedding semi-supervised machine learning to construct a semi-supervised sweet-spot evaluation dataset; then

through the TOPSIS method and the Euclidean distance theory

the weights of each evaluation indicator are determined by combining multi-objective programming algorithms; finally

a semi-supervised comprehensive sweet-spot evaluation model incorporating domain knowledge is established. Compared with benchmark models

the R² values between the evaluation results of the proposed model and the number of microseismic events as well as the gas production index per meter are improved by 5.58% and 10.51%. Field application results show that the dominant controlling factors of sweet spots in Well JS1 are total gas content

porosity

and Poisson’s ratio

with a cumulative weight of 41.21%; the R² values between the evaluation results and both the number of microseismic events and gas production index per meter exceed 0.85. This model enables accurate sweet-spot evaluation for shale gas horizontal wells and provides important reference and technical support for development scenarios with limited labeled data

such as areas with large geological variations and new well development.

关键词

Keywords

references

聂海宽 , 汪虎 , 何治亮 , 等 . 常压页岩气形成机制、分布规律及勘探前景: 以四川盆地及其周缘五峰组—龙马溪组为例 [J ] . 石油学报 , 2019 , 40 ( 2 ): 131 - 143, 164 .

Nie Haikuan , Wang Hu , He Zhiliang , et al . Formation mechanism, distribution and exploration prospect of normal pressure shale gas reservoir: a case study of Wufeng Formation-Longmaxi Formation in Sichuan Basin and its periphery [J ] . Acta Petrolei Sinica , 2019 , 40 ( 2 ): 131 - 143, 164 .

何希鹏 , 蔡潇 , 高玉巧 , 等 . 页岩气勘探开发实验技术研究进展与发展方向 [J ] . 天然气工业 , 2024 , 44 ( 7 ): 12 - 26 .

He Xipeng , Cai Xiao , Gao Yuqiao , et al . Experimental technologies of shale gas exploration and development: research progress and development direction [J ] . Natural Gas Industry , 2024 , 44 ( 7 ): 12 - 26 .

邹才能 , 丁云宏 , 卢拥军 , 等 . “人工油气藏”理论、技术及实践 [J ] . 石油勘探与开发 , 2017 , 44 ( 1 ): 144 - 154 .

Zou Caineng , Ding Yunhong , Lu Yongjun , et al . Concept, technology and practice of “man-made reservoirs” development [J ] . Petroleum Exploration and Development , 2017 , 44 ( 1 ): 144 - 154 .

柳忠泉 , 曾治平 , 田继军 , 等 . 吉木萨尔凹陷二叠系芦草沟组“甜点”成因与分布预测 [J ] . 岩性油气藏 , 2022 , 34 ( 3 ): 15 - 28 .

Liu Zhongquan , Zeng Zhiping , Tian Jijun , et al . Genesis and distribution prediction of sweet spots of Permian Lucaogou Formation in Jimsar Sag [J ] . Lithologic Reservoirs , 2022 , 34 ( 3 ): 15 - 28 .

罗群 , 张泽元 , 袁珍珠 , 等 . 致密油“甜点”的内涵、评价与优选: 以酒泉盆地青西凹陷白垩系下沟组为例 [J ] . 岩性油气藏 , 2022 , 34 ( 4 ): 1 - 12 .

Luo Qun , Zhang Zeyuan , Yuan Zhenzhu , et al . Connotation, evaluation and optimization of tight oil sweet spots: a case study of Cretaceous Xiagou Formation in Qingxi Sag, Jiuquan Basin [J ] . Lithologic Reservoirs , 2022 , 34 ( 4 ): 1 - 12 .

曾凡辉 , 胡大淦 , 郭建春 , 等 . 基于钻录井资料的页岩气水平井分段多簇差异化压裂参数智能优化 [J ] . 天然气工业 , 2025 , 45 ( 2 ): 84 - 94 .

Zeng Fanhui , Hu Dagan , Guo Jianchun , et al . Intelligent optimization of staged multi-cluster differentiated hydraulic fracturing parameters of shale gas horizontal wells based on drilling and mud logging data [J ] . Natural Gas Industry , 2025 , 45 ( 2 ): 84 - 94 .

Yang H H , Li F W , Wang W , et al . A novel approach for identifying sweet spots in tight reservoir fracturing engineering based on physical-data dual drive [J ] . Journal of Applied Geophysics , 2025 , 238 : 105735 .

郭建春 , 吴涛 , 曾凡辉 , 等 . 深层页岩气储层高效改造关键技术与矿场实践 [J ] . 天然气工业 , 2025 , 45 ( 12 ): 125 - 138 .

Guo Jianchun , Wu Tao , Zeng Fanhui , et al . Key technologies and field practices of full-field support and transformation for deep shale gas reservoirs [J ] . Natural Gas Industry , 2025 , 45 ( 12 ): 125 - 138 .

张琳琳 , 王孔杰 , 赖枫鹏 , 等 . 鄂尔多斯盆地西缘奥陶系乌拉力克组海相页岩气储层“甜点”分类评价 [J ] . 石油实验地质 , 2024 , 46 ( 1 ): 191 - 201 .

Zhang Linlin , Wang Kongjie , Lai Fengpeng , et al . Classification and evaluation of sweet spots of marine shale gas reservoir in Ordovician Wulalike Formation on the westen margin of Ordos Basin [J ] . Petroleum Geology and Experiment , 2024 , 46 ( 1 ): 191 - 201 .

Zhang S , Liu H M , Liu Y L , et al . Main controls and geological sweet spot types in Paleogene shale oil rich areas of the Jiyang Depression, Bohai Bay basin, China [J ] . Marine and Petroleum Geology , 2020 , 111 : 576 - 587 .

郭军 , 苑坤 , 郭经纬 , 等 . 桂中—南盘江地区黔水地1井打屋坝组页岩储层含气性及可压性评价 [J ] . 中国地质调查 , 2023 , 10 ( 6 ): 25 - 30 .

Guo Jun , Yuan Kun , Guo Jingwei , et al . Gas bearing and compressibility evaluation of shale reservoir in Dawuba Formation of Qianshuidi 1 Well in Guizhong-Nanpanjiang area [J ] . Geological Survey of China , 2023 , 10 ( 6 ): 25 - 30 .

Sun N , Gu Y , Zhang X L , et al . Fractureability evaluation of shale gas reservoirs [J ] . Energy Sources, Part A: Recovery, Utilization, and Environmental Effects , 2022 , 44 ( 1 ): 2595 - 2603 .

Zeng F H , Gong G B , Zhang Y , et al . Fracability evaluation of shale reservoirs considering rock brittleness, fracture toughness, and hydraulic fracturing-induced effects [J ] . Geoenergy Science and Engineering , 2023 , 229 : 212069 .

聂云丽 , 高国忠 . 基于随机森林的页岩气“甜点”分类方法 [J ] . 油气藏评价与开发 , 2023 , 13 ( 3 ): 358 - 367 .

Nie Yunli , Gao Guozhong . Classification of shale gas “sweet spot” based on Random Forest machine learning [J ] . Reservoir Evaluation and Development , 2023 , 13 ( 3 ): 358 - 367 .

冷钇江 , 唐洪明 , 王滢 , 等 . 基于层次分析法的页岩储层地质甜点优化评价 [J ] . 西南石油大学学报(自然科学版) , 2025 , 47 ( 5 ): 62 - 74 .

Leng Yijiang , Tang Hongming , Wang Ying , et al . Geology sweet spot of shale gas optimization evaluation based on analytic hierarchy process [J ] . Journal of Southwest Petroleum University(Science & Technology Edition) , 2025 , 47 ( 5 ): 62 - 74 .

Liu Y Z , Zeng J H , Qiao J C , et al . An advanced prediction model of shale oil production profile based on source-reservoir assemblages and artificial neural networks [J ] . Applied Energy , 2023 , 333 : 120604 .

郭建春 , 张宇 , 曾凡辉 , 等 . 非常规油气储层智能压裂技术研究进展与展望 [J ] . 天然气工业 , 2024 , 44 ( 9 ): 13 - 26 .

Guo Jianchun , Zhang Yu , Zeng Fanhui , et al . Research progress and prospects of intelligent fracturing technology for unconventional reservoirs [J ] . Natural Gas Industry , 2024 , 44 ( 9 ): 13 - 26 .

Zeng F H , Wu T , Guo J C , et al . Unsupervised learning based on prior knowledge enhancement for the classification and evaluation of geological-engineering sweet spots [J ] . SPE Journal , 2026 , 31 ( 1 ): 95 - 115 .

Li J X , Zhu X Y , Wang H R , et al . Stacked co-training for semi-supervised multi-label learning [J ] . Information Sciences , 2024 , 677 : 120906 .

Yang Y , Liu X C . A robust semi-supervised learning approach via mixture of label information [J ] . Pattern Recognition Letters , 2015 , 68 : 15 - 21 .

Qin Y , Ding S F , Wang L J , et al . Research progress on semi-supervised clustering [J ] . Cognitive Computation , 2019 , 11 ( 5 ): 599 - 612 .

Xu R , Yan T , Sun S H , et al . Intelligent evaluation method for identifying favorable shale oil areas based on improved stacked sparse autoencoder [J ] . Oil Shale , 2025 , 42 ( 1 ): 79 - 114 .

Fan H J , Zhao X Q , Liang X , et al . Semi-supervised learning–based petrophysical facies division and “sweet spot” identification of low-permeability sandstone reservoir [J ] . Frontiers in Earth Science , 2022 , 9 : 805342 .

陈豪杰 , 董青卫 , 刘锐楷 , 等 . 融合先验知识与引导策略搜索的机器人轴孔装配方法 [J ] . 计算机应用研究 , 2025 , 42 ( 4 ): 1018 - 1024 .

Chen Haojie , Dong Qingwei , Liu Ruikai , et al . Robotic pin-hole assembly method integrating prior knowledge and guided policy search [J ] . Application Research of Computers , 2025 , 42 ( 4 ): 1018 - 1024 .

Ma S Q , Liu Z , Song Y Q , et al . A domain knowledge-based semi-supervised pancreas segmentation approach [C ] // Neural information processing . Singapore : Springer , 2024 : 69 - 82 .

梁玉凯 , 宋荣彩 , 郑华安 , 等 . 基于熵权法-TOPSIS-灰色关联法的储层综合评价: 以莺歌海盆地黄流组为例 [J ] . 成都理工大学学报(自然科学版) , 2024 , 51 ( 1 ): 91 - 101 .

Liang Yukai , Song Rongcai , Zheng Hua’an , et al . Comprehensive evaluation of reservoirs based on the entropy weight method, TOPSIS, and the gray correlation method: taking the Huangliu Formation in Yinggehai Basin as an example [J ] . Journal of Chengdu University of Technology (Science & Technology Edition) , 2024 , 51 ( 1 ): 91 - 101 .

陈军涛 , 李文昕 , 薛峰 , 等 . 主客观组合赋权的TOPSIS-RSR注浆效果综合评价方法 [J ] . 煤炭科学技术 , 2023 , 51 ( S2 ): 191 - 199 .

Chen Juntao , Li Wenxin , Xue Feng , et al . Comprehensive evaluation of TOPSIS-RSR grouting effect based on subjective and objective combined weights [J ] . Coal Science and Technology , 2023 , 51 ( S2 ): 191 - 199 .

冷钇江 , 唐洪明 , 王滢 , 等 . 基于层次分析法的页岩储层地质“甜点”优化评价 [J ] . 西南石油大学学报(自然科学版) , 2025 , 47 ( 5 ): 62 - 74 .

Leng Yijiang , Tang Hongming , Wang Ying , et al . Geology sweet spot of shale gas optimization evaluation based on analytic hierarchy process [J ] . Journal of Southwest Petroleum University (Science & Technology Edition) , 2025 , 47 ( 5 ): 62 - 74 .

申有义 , 王凯峰 , 唐书恒 , 等 . 沁水盆地榆社—武乡区块二叠系煤系页岩储层地质建模及“甜点”预测 [J ] . 岩性油气藏 , 2024 , 36 ( 4 ): 98 - 108 .

Shen Youyi , Wang Kaifeng , Tang Shuheng , et al . Geological modeling and “sweet spot” prediction of Permian coal measures shale reservoirs in Yushe-Wuxiang block, Qinshui Basin [J ] . Lithologic Reservoirs , 2024 , 36 ( 4 ): 98 - 108 .

陈峥嵘 , 刘伟 , 朱学申 , 等 . 深部煤层气地质-工程一体化“甜点”区预测 [J ] . 煤田地质与勘探 , 2025 , 53 ( 6 ): 191 - 200 .

Chen Zhengrong , Liu Wei , Zhu Xueshen , et al . Prediction of geological and engineering integrated sweet spots of deep coalbed methane [J ] . Coal Geology & Exploration , 2025 , 53 ( 6 ): 191 - 200 .

Cheng B J , Xu T J , Luo S Y , et al . Method and practice of deep favorable shale reservoirs prediction based on machine learning [J ] . Petroleum Exploration and Development , 2022 , 49 ( 5 ): 1056 - 1068 .

孙伟峰 , 刘凯 , 张德志 , 等 . 结合钻井工况与Bi-GRU的溢流与井漏监测方法 [J ] . 石油钻探技术 , 2023 , 51 ( 3 ): 37 - 44 .

Sun Weifeng , Liu Kai , Zhang Dezhi , et al . A kick and lost circulation monitoring method combining Bi-GRU and drilling conditions [J ] . Petroleum Drilling Techniques , 2023 , 51 ( 3 ): 37 - 44 .

Zhang X L , Xu Z S . Extension of topsis to Multiple Criteria Decision Making with Pythagorean Fuzzy Sets: Extension of topsis to multiple criteria decision making [J ] . International Journal of Intelligent Systems , 2014 , 29 ( 12 ): 1061 - 1078 .

杨三军 , 荆亚茹 , 姜润发 . 矿山救援队应急救援能力评价指标体系构建 [J ] . 中国安全科学学报 , 2021 , 31 ( 8 ): 180 - 188 .

Yang Sanjun , Jing Yaru , Jiang Runfa . Construction of evaluation index system for emergency rescue capabilities of mine rescue teams [J ] . China Safety Science Journal , 2021 , 31 ( 8 ): 180 - 188 .

Chen P Y . Effects of normalization on the entropy-based TOPSIS method [J ] . Expert Systems with Applications , 2019 , 136 : 33 - 41 .

Liu Y , Xie G Q , Zheng H L , et al . Optimization of fracturing sweet spot in deep carbonate reservoirs by combining TOPSIS and AHP algorithm [J ] . Processes , 2025 , 13 ( 9 ): 2777 .

史璨 , 林伯韬 , 谢勃勃 , 等 . 基于双“甜点”的页岩储层可压性评价方法 [J ] . 深圳大学学报(理工版) , 2024 , 41 ( 2 ): 183 - 191 .

Shi Can , Lin Botao) , Xie Bobo , et al . Evaluation method of shale reservoir fracability based on double sweet spots [J ] . Journal of Shenzhen University (Science and Engineering) , 2024 , 41 ( 2 ): 183 - 191 .

Deng Y X , Wang W D , Su Y L , et al . An unsupervised machine learning based double sweet spots classification and evaluation method for tight reservoirs [J ] . Journal of Energy Resources Technology , 2023 , 145 ( 7 ): 072602 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Developmental characteristics of natural fractures in shale oil and gas reservoirs and their controls on reservoir formation and accumulation

Study on microscopic characteristics of shale under thermal effects and their application: a case study of Longmaxi Formation in southern Sichuan Basin

Geology-engineering integrated study of bottom hole assembly selection for shale oil and gas horizontal wells in Junggar Basin

Production capacity prediction model of shale gas fractured horizontal wells considering rock creep

Progress, challenges, and suggestions for Lower Cambrian shale gas exploration and development in Sichuan Basin and its periphery

Related Author

YUAN DENGFU

WANG BOYANG

LIU BO

LI SHICHAO

XIAO FEI

QAMAR YASIN

HUANG YIMING

YANG JIANGUO

Related Institution

School of Earth Sciences, Northeast Petroleum University

Institute of Unconventional Oil & Gas, Northeast Petroleum University

Shenyang Center of China Geological Survey (Northeast Geological Science and Technology Innovation Center)

Sichuan Geological Survey and Research Institute

Shale Gas Geological Evaluation and Efficient Development Key Laboratory of Sichuan Province

AI问答

⁰