煤系含水层及多煤层合采水来源判识方法——以山西保德区块为例  

Identifying methods for coal measure aquifers and multi-coal seam commingled drainage water sources:A case study of the Baode block in Shanxi Province

作　　者：张庆丰 张继坤 李子玲 姚艳斌[2,3] 杜丰丰 孙晓光 郝帅 李军辉 ZHANG Qingfeng;ZHANG Jikun;LI Ziling;YAO Yanbin;DU Fengfeng;SUN Xiaoguang;HAO Shuai;LI Junhui(PetroChina Coalbed Methane Company Limited,Beijing 100028,China;School of Energy Resource,China University of Geosciences(Beijing),Beijing 100o83,China;Coal Reservoir Laboratory of National Engineering Research Center of CBM Development&Utilization,Beijing 100083,China)

机构地区：[1]中石油煤层气有限责任公司,北京100028 [2]中国地质大学(北京)能源学院,北京100083 [3]煤层气开发利用国家工程研究中心煤储层物性实验室,北京100083

出　　处：《中国矿业大学学报》2025年第1期115-126,共12页Journal of China University of Mining & Technology

基　　金：国家自然科学基金杰出青年基金项目(42125205);国家自然科学基金重点项目(42130806)。

摘　　要：煤层气多层合采成功与否的关键在于对各产层产水来源与产气贡献的精确判识.现有排采水来源判识方法的不足,给多煤层合采井的产层组合优化及排采工作制度制定带来较大挑战.为此,本文以保德区块85口煤层气井勘探开发与水化学测试资料为基础,确定了含水层层位,阐述了排采水类型划分依据.应用主成分分析法(PCA)与Bayes判别法,构建了多煤层排采水来源判识模型,形成了“含水层判识、排采水类型划分、排采水来源判识”的多煤层合采水来源判识系列方法.基于27个排采水样的应用,验证了该模型的准确性,解析了合层排采水来源不同情况下的层间干扰强弱.结果表明:4+5号煤层顶板与8+9号煤层顶底板含水层发育稳定、厚度大,对煤层气井生产影响较大.排采水类型划分与排采水来源判识模型均能够进行排采水来源分析.其中排采水类型划分是基于地质工程背景特征,初步定性划分排采水来源,适用于煤层气井全生命周期补给水来源判识;而排采水来源判识模型是结合产出水常规离子特征来进一步动态识别排采水来源,对合层排采层间干扰强弱分析具有一定指示意义.合层开发井排采水来源若为单一煤层水,层间干扰剧烈,产气效果差;若为合采煤层与含水层混合水,层间干扰较强,产气效果一般;若为合采煤层水,层间干扰弱,产气效果较好.该研究为合层排采水来源判识提供了一种可行方法,为多煤层合采层间干扰程度评价、排采设备及排采工作制度调整提供依据.The success of coalbed methane(CBM)drainage in multilayered coal seam gas wells depends on the precise identification of water sources and gas contributions from each productive layer.Current methods for identifying water sources exhibit deficiencies,presenting significant challenges in optimizing layer combinations and designing drainage systems for multi-coalseam gas wells.To address this,this study leveraged exploration and development data from 85 CBM wells in the Baode block,coupled with chemical testing,to identify aquifer strata and establish criteria for classifying drainage water types.Using principal component analysis(PCA)and Bayes discriminant analysis,a model for identifying drainage water sources in multi-coal seams is constructed,yielding methods for aquifer identification,drainage water type classification,and drainage water source identification within multi-coal seam integrated mining.Validation using 27 drainage water samples confirmed the accuracy of the model and assessed the interlayer interference strength under varying drainage water source conditions during integrated mining.The results indicate stable and thick aquifer strata at the top and bottom of coal seams 4+5 and 8+9,significantly impacting well production.Both the classification of drainage water types and the drainage water source identification model are useful for analyzing the origins of drainage water.The classification is based on geological engineering characteristics and provides a preliminary qualitative division of the drainage water sources.This approach is particularly effective for identifying recharge water sources throughout the life cycle of coalbed methane wells.The drainage water source identification model integrates conventional ion characteristics of the produced water to dynamically pinpoint the source of the drainage water.This approach offers insights for analyzing interlayer interference intensity in combined layer drainage.Single-coal-seam water sources in integrated development wells result in severe int

关 键 词：煤层气 排采水来源判识 主成分分析法 Bayes判别法 层间干扰 

分 类 号：P618.11[天文地球—矿床学]