基于微地震约束的多尺度复杂压裂缝网自动反演新方法  

An automatic inversion method for parameter determination of multi-scale complex hydraulic fracture network based on microseismic constraint

作　　者：李俊超 戴城 方思冬 LI Junchao;DAI Cheng;FANG Sidong(Xi'an Shiyou University,Xi'an,Shaanxi 710065,China;Sinopec Exploration&Production Research Institute,Beijing 100083,China)

机构地区：[1]西安石油大学 [2]中国石化石油勘探开发研究院

出　　处：《天然气工业》2023年第12期46-54,共9页Natural Gas Industry

基　　金：国家自然科学基金项目“陆相中低熟页岩油原位转化开发裂缝动态演化机理研究”(编号:52204047)。

摘　　要：水平井多级压裂是实现页岩气藏有效动用的关键技术。然而,大规模体积压裂形成的缝网由不同尺度的压裂主缝和次生裂缝构成,裂缝的分布情况及其参数难以确定。为此,以微地震监测数据作为先验约束,建立考虑微地震湿事件区域与监测灵敏度的多尺度离散裂缝参数模型,再通过嵌入式离散裂缝方法建立正演模型,利用集合卡曼滤波方法自动反演确定多尺度离散裂缝的位置与渗透率等属性参数。研究结果表明:①反演并确定验证模型的多尺度离散裂缝参数,参数均收敛到真实值附近,与前验统计矩相比,参数的不确定性明显下降,产量预测符合率显著提高;②通过定产气量拟合井底流压,反演并确定了实际压裂水平井模型微地震有效区域及缝网参数,拟合阶段与预测阶段井底流压符合率分别提高了75.7%和82.4%;③反演得到的裂缝有效长度是微地震事件点显示裂缝长度的1/2~2/3,表明微地震数据与裂缝参数的不确定性较高;④微地震监测灵敏度的后验值为3.59,说明可能有大量压裂激发的微地震事件没有反映在监测数据中。结论认为,该方法可有效反演并确定微地震约束下的多尺度复杂压裂缝网参数,并可适应具有较高不确定性的微地震数据和缝网模型。Multiple-stage horizontal well fracturing is a key technology for the effective exploitation of shale gas reservoirs.However,the fracture network formed by large-scale volumetric fracturing consists of primary and secondary fractures of varying scales,making it extremely challenging to determine the distribution and parameters of these fractures.Therefore,taking microseismic monitoring data as prior constraints,this paper establishes a multi-scale discrete fracture parameter model that takes into account the microseismic event zones and monitoring sensitivity,and constructs a forward model by means of the embedded discrete fracture modeling.In addition,the positions and attribute parameters(e.g.permeability)of multi-scale discrete fractures are automatically inverted and determined using the ensemble Kalman filtering method.And the following research results are obtained.First,the multi-scale discrete fracture parameters of the validation model are inverted and determined to converge near the true values.Compared with the prior statistical moments,the uncertainty of parameters is decreased greatly,leading to the significant improvement of production prediction accuracy.Second,the effective microseismic areas and fracture network parameters in the model of actual fractured horizontal well are inverted and determined by fitting the bottom hole flowing pressure at a constant gas production rate.The coincidence rate of bottom hole flowing pressure rises by 75.7%during the fitting stage and 82.4%during the prediction stage.Third,the inverted effective fracture length is half to two thirds of the fracture length shown by the microseismic event point,which indicates a relatively high degree of uncertainty between microseismic data and fracture parameters.Fourth,the posterior value of microseismic monitoring sensitivity is 3.59,suggesting that a significant number of fracturing induced microseismic events may not be reflected in the monitoring data.In conclusion,this method can effectively invert and determine the parame

关 键 词：页岩气藏 压裂水平井 多尺度裂缝 微地震数据 嵌入式离散裂缝模型 集合卡曼滤波 历史拟合 参数反演 

分 类 号：TE377[石油与天然气工程—油气田开发工程]