A deep reinforcement learning(DRL)based approach for well-testing interpretation to evaluate reservoir parameters

Parameter inversions in oil/gas reservoirs based on well test interpretations are of great significance in oil/gas industry.Automatic well test interpretations based on artificial intelligence are the most promising to solve the problem of non-unique solution.In this work,a new deep reinforcement learning(DRL)based approach is proposed for automatic curve matching for well test interpretation,by using the double deep Q-network(DDQN).The DDQN algorithms are applied to train agents for automatic parameter tuning in three conventional well-testing models.In addition,to alleviate the dimensional disaster problem of parameter space,an asynchronous parameter adjustment strategy is used to train the agent.Finally,field applications are carried out by using the new DRL approaches.Results show that step number required for the DDQN to complete the curve matching is the least among,when comparing the naive deep Q-network(naive DQN)and deep Q-network(DQN).We also show that DDQN can improve the robustness of curve matching in comparison with supervised machine learning algorithms.Using DDQN algorithm to perform 100 curve matching tests on three traditional well test models,the results show that the mean relative error of the parameters is 7.58%for the homogeneous model,10.66%for the radial composite model,and 12.79%for the dual porosity model.In the actual field application,it is found that a good curve fitting can be obtained with only 30 steps of parameter adjustment.

Interpretable machine learning optimization(InterOpt)for operational parameters:A case study of highly-efficient shale gas development

An algorithm named InterOpt for optimizing operational parameters is proposed based on interpretable machine learning,and is demonstrated via optimization of shale gas development.InterOpt consists of three parts:a neural network is used to construct an emulator of the actual drilling and hydraulic fracturing process in the vector space(i.e.,virtual environment);:the Sharpley value method in inter-pretable machine learning is applied to analyzing the impact of geological and operational parameters in each well(i.e.,single well feature impact analysis):and ensemble randomized maximum likelihood(EnRML)is conducted to optimize the operational parameters to comprehensively improve the efficiency of shale gas development and reduce the average cost.In the experiment,InterOpt provides different drilling and fracturing plans for each well according to its specific geological conditions,and finally achieves an average cost reduction of 9.7%for a case study with 104 wells.

A method to interpret fracture aperture of rock slope using adaptive shape and unmanned aerial vehicle multi-angle nap-of-the-object photogrammetry

The aperture of natural rock fractures significantly affects the deformation and strength properties of rock masses,as well as the hydrodynamic properties of fractured rock masses.The conventional measurement methods are inadequate for collecting data on high-steep rock slopes in complex mountainous regions.This study establishes a high-resolution three-dimensional model of a rock slope using unmanned aerial vehicle(UAV)multi-angle nap-of-the-object photogrammetry to obtain edge feature points of fractures.Fracture opening morphology is characterized using coordinate projection and transformation.Fracture central axis is determined using vertical measuring lines,allowing for the interpretation of aperture of adaptive fracture shape.The feasibility and reliability of the new method are verified at a construction site of a railway in southeast Tibet,China.The study shows that the fracture aperture has a significant interval effect and size effect.The optimal sampling length for fractures is approximately 0.5e1 m,and the optimal aperture interpretation results can be achieved when the measuring line spacing is 1%of the sampling length.Tensile fractures in the study area generally have larger apertures than shear fractures,and their tendency to increase with slope height is also greater than that of shear fractures.The aperture of tensile fractures is generally positively correlated with their trace length,while the correlation between the aperture of shear fractures and their trace length appears to be weak.Fractures of different orientations exhibit certain differences in their distribution of aperture,but generally follow the forms of normal,log-normal,and gamma distributions.This study provides essential data support for rock and slope stability evaluation,which is of significant practical importance.

A Study on Identification of Conglomerate Reservoir Parameters and Oil/Water Beds in Ke82 Well Areas of Junggar Basin

This paper is mainly about the calculation of reservoir parameters and theinterpretation method for identifying oil/water beds in Ke82 well areas of Junggar basin. It isdifficult to determine the reservoir parameters with common logging methods such as core calibrationlog because of the diversity of minerals and rocks and the complexity of pore structures in theconglomerate reservoir of Junggar basin. Optimization logging exploration is a good method todetermine the porosity by establishing the multi-mineral model with logging curves based on theintegration of geological, core and well testing data. Permeability is identified by BP algorithm ofneural network. Hydrocarbon saturation is determined by correlating Archie's and Simandouxformulas. Comparing the exploratory result and core data, we can see that these methods areeffective for conglomerate logging exploration. We processed and explained six wells in the Ke82well areas. And actual interpretation has had very good results, 85 % of which conform to welltesting data. Therefore, this technique will be effective for identifying conglomerate parameters.

X区基岩物性参数解释模型建立

基岩的岩性复杂,在平面上与纵向上差别很大。在基岩内部发育基质孔隙、溶孔和裂缝等,且油层、气层和水层的响应规律复杂,对于计算基岩的物性参数具有挑战性。针对研究区概况,采用常规测井资料、核磁测井资料、成像测井资料对研究区的物性参数进行计算,建立适用于研究区的物性参数解释模型。在建立孔隙度解释模型的过程中,根据常规测井资料以及核磁测井资料来建立孔隙度解释模型,采用核磁资料计算总孔隙度以及有效孔隙度,用常规测井资料计算中子、密度和声波孔隙度,运用中子和密度孔隙度采用气层校正公式计算有效孔隙度;对于渗透率解释模型,采用核磁测井资料KSDR以及KTIM公式来计算,用RQI和FZI来定量计算渗透率。结果表明,采用气层校正公式计算孔隙度的效果较好,采用RQI和FZI计算渗透率结果较好。

融合可解释机器学习的成品汽油调和配方质量预测评价与致因分析

受成品汽油调和配方需“先验”评价与修正的驱动,本研究将轻量级梯度提升树(LightGBM)与可解释机器学习(SHAP)方法相结合,兼顾复杂模型精度高与后验SHAP可解释性强的各自优势,提出了一种调和配方质量预测评价及致因分析方法。该方法先引用改进遗传算法(IGA)优化LightGBM的超参数,建立了可同时预测成品汽油性能和环保指标的模型,并结合汽油国ⅥA标准与企业生产实际制定了配方质量评价标准,实现配方“先验”评价;再基于SHAP的全局和局部致因分析,对缺陷配方给出了易于操作的单变量定性修正建议。实验结果表明:相比于传统BP网络和随机森林(RF)、以及采用随机搜索和GA优化参数的LightGBM等模型,IGA_LightGBM模型可得到更全面和精准的预测指标,SHAP致因分析可给出契合实际的修正建议。该方法是智能算法代替人工的有益探索。

多源数据驱动的轧机振动预测及可解释性分析

为研究轧制过程动态工艺参数对轧机振动的影响规律,改善现有研究中机理模型精度较低且数据模型缺乏可解释性的问题,采用极端梯度提升(Extreme Gradient Boosting,XGBoost)算法建立基于多源数据的轧机振动预测模型,并使用SHapley Additive exPlanations(SHAP)框架对预测模型进行解释。通过与其他预测模型相比,XGBoost预测模型可以利用工艺参数实现对轧机运行状态的高精度预测。基于SHAP框架解释的结果表明,出入口厚度、轧制力、轧制速度对轧机振动影响较大,后张力对轧机振动影响较小。研究为提高轧机设备与工艺参数的匹配度,实现将工业数据应用于轧机振动预测和分析提供理论基础。

基于改进时序网络的钻进参数可解释实时预测

实时准确预测钻进参数变化趋势对现场钻井作业具有重要参考价值。针对智能模型在现场作业应用中面临的钻进参数可获取性限制,提出了一种基于注意力时域卷积网络(AT-TCN)的钻进参数超前预测方法。该方法不仅考虑了录井曲线随深度变化的趋势和自相关性,同时嵌入高拓展性的注意力机制模块,使模型更好地捕捉钻进参数的动态变化。利用现场钻井数据集测试,评估了模型在预测4种关键钻进参数(扭矩、立管压力、钻井液当量密度和机械钻速)方面的有效性和准确性。研究结果表明:AT-TCN预测当量密度的准确率最高达到99%,且在模型精度和计算效率上,均优于其他4种深度学习模型,能够有效捕捉钻进参数的变化趋势。AT-TCN还提供模型的双重可解释性,可从时序和特征维度方面反映输入序列对预测结果的影响。研究结果有望为钻井作业的安全性、高效性作出重要贡献,具有较强的落地应用价值。

基于无人机遥感的云南松林冠参数提取研究

以滇中地区典型天然云南松纯林为对象,结合无人机遥感影像获取和外业标准地调查,利用传统目视解译、多尺度分割和分水岭分割方法进行单木冠幅和郁闭度2个林冠参数的提取研究。结果表明,以地面实测数据为参考,3种单木分割方法的冠幅提取精度分别达91.48%、87.33%、84.04%;以目视解译结果为参考,多尺度分割方法提取郁闭度的精度达90.24%,且决定系数(R2)达0.8845;分水岭分割方法提取郁闭度的精度达87.40%,R2达0.7437。研究结果表明,基于无人机遥感影像的多尺度分割方法能很好地提取云南松纯林的林冠参数信息,且提取精度满足森林资源调查的要求,可有效提高森林资源调查的效率。

Assessment of electrostatic discharge sensitivity of nitrogen-rich heterocyclic energetic compounds and their salts as high energy-density dangerous compounds:A study of structural variables

Nitrogen-rich heterocyclic energetic compounds(NRHECs)and their salts have witnessed widespread synthesis in recent years.The substantial energy-density content within these compounds can lead to potentially dangerous explosive reactions when subjected to external stimuli such as electrical discharge.Therefore,developing a reliable model for predicting their electrostatic discharge sensitivity(ESD)becomes imperative.This study proposes a novel and straightforward model based on the presence of specific groups(-NH_(2) or-NH-,-N=N^(+)-O^(-)and-NNO_(2),-ONO_(2) or-NO_(2))under certain conditions to assess the ESD of NRHECs and their salts,employing interpretable structural parameters.Utilizing a comprehensive dataset comprising 54 ESD measurements of NRHECs and their salts,divided into 49/5 training/test sets,the model achieves promising results.The Root Mean Square Error(RMSE),Mean Absolute Error(MAE),and Maximum Error for the training set are reported as 0.16 J,0.12 J,and 0.5 J,respectively.Notably,the ratios RMSE(training)/RMSE(test),MAE(training)/MAE(test),and Max Error(training)/Max Error(test)are all greater than 1.0,indicating the robust predictive capabilities of the model.The presented model demonstrates its efficacy in providing a reliable assessment of ESD for the targeted NRHECs and their salts,without the need for intricate computer codes or expert involvement.

Fracture network types revealed by well test curves for shale reservoirs in the Sichuan Basin,China

Pressure buildup testing can be used to analyze fracture network characteristics and conduct quantitative interpretation of relevant parameters for shale gas wells,thus providing bases for assessing the well productivity and formulating proper development strategies.This study establishes a new well test interpretation model for fractured horizontal wells based on seepage mechanisms of shale reservoirs and proposes a method for identifying fracturing patterns based on the characteristic slopes of pressure buildup curves and curve combination patterns.The pressure buildup curve patterns are identified to represent three types of shale reservoirs in the Sichuan Basin,namely the moderately deep shale reservoirs with high pressure,deep shale reservoirs with ultra-high pressure,and moderately deep shale reservoirs with normal pressure.Based on this,the relationship between the typical pressure buildup curve patterns and the fracture network types are put forward.Fracturing effects of three types of shale gas reservoir are compared and analyzed.The results show that typical flow patterns of shale reservoirs include bilinear flow in primary and secondary fractures,linear flow in secondary fractures,bilinear flow in secondary fractures and matrix,and linear flow in matrix.The fracture network characteristics can be determined using the characteristic slopes of pressure buildup curves and curve combinations.The linear flow in early secondary fractures is increasingly distinct with an increase in primary fracture conductivity.Moreover,the bilinear flow in secondary fractures and matrix and the subsequent linear flow in the matrix occur as the propping and density of secondary fractures increase.The increase in the burial depth,in-situ stress,and stress difference corresponds to a decrease in the propping of primary fractures that expand along different directions in the shale gas wells in the Sichuan Basin.Four pressure buildup curve patterns exist in the Sichuan Basin and its periphery.The pattern of pressure buildup curves of shale reservoirs in the Yongchuan area can be described as 1/2/→1/4,indicating limited stimulated reservoir volume,poorly propped secondary fractures,and the forming of primary fractures that extend only to certain directions.The pressure buildup curves of shale reservoirs in the main block of the Fuling area show a pattern of 1/4/→1/2 or 1/2,indicating greater stimulated reservoir volume,well propped secondary fractures,and the forming of complex fracture networks.The pattern of pressure buildup curves of shale reservoirs in the Pingqiao area is 1/2/→1/4→/1/2,indicating a fracturing effect somewhere between that of the Fuling and Yongchuan areas.For reservoirs with normal pressure,it is difficult to determine fracture network characteristics from pressure buildup curves due to insufficient formation energy and limited liquid drainage.

GB/T37635-2019国家标准制定研究与解读

文中对GB/T37635-2019《纺织品弹性织带耐疲劳外观变化试验方法》国家标准进行解读。重点介绍标准的适用范围、试验原理;对试样尺寸、拉伸速度、拉伸比例、疲劳拉伸次数和外观检查间隔周期等试验参数进行研究,确定试验参数;并从试样重复性和试验室比对两方面对试验方法进行验证。结果表明:该试验方法具有良好的重现性;该标准对于规范弹性织带的质量、统一上下游企业对弹性织带的评价方法和标准具有重要意义。

W-S模型中B参数的确定方法

从泥质砂岩储层特点出发 ,以W S模型为理论依据 ,详细研究了W S模型中平衡阳离子电化学当量电导B的确定方法。在 15块岩心电阻率测量数据的基础上 ,研究出了B参数随溶液矿化度和温度变化的经验关系式。此项成果已应用于饱和度数字处理工作 ,目前已经处理了 60余口井 ,应用效果较好。

应用测井资料评价四川盆地南部页岩气储层

页岩气储层的测井解释在国内刚刚起步,与常规储层相比有着不同的测井响应特征。因此,应用ΔlogR方法(孔隙度和电阻率曲线重叠法)识别页岩气时应考虑到重叠基线的选取、岩性的变化等的影响;用电成像测井资料对不同地质特征的拾取、裂缝的有效性评价、地应力的分析是关键技术。经过对Wx井页岩气储层部分物性参数的初步处理,有针对性地选取测井项目,探索出了对页岩气的测井解释模式,这对今后页岩气储层测井解释评价工作的开展具有重要的参考意义。

基于知识库的解释模型智能优选测井数据处理方法

在复杂储层评价中,针对不同性质特点的目标层段自动选取最佳校正图版和解释方程,这对测井曲线进行智能化处理是非常需要但同时又是难于解决的一个问题。依据测井曲线特征可将测井处理解释知识分为基础信息、规则和策略,并结合解释参数、解释模型和解释图版等测井资料,构建了能够通过数据挖掘中的关联规则和项目集加权增量算法实现自动更新的测井处理解释知识库,基于该知识库构建了一个测井处理解释子系统,能够初步实现复杂储层测井处理解释从知识获取到参数和模型智能优选的全过程。通过在国家重大科技专项一体化平台CIFLog上的集成运行,证明了该处理流程及算法是科学有效的。

注水前后储层参数解释模型的建立

建立准确的储层参数测井解释模型,是进行储层非均质性表征的前提和基础。在综合研究胡状集油田沙三段注水前后岩心、测试、地质等资料的基础上,分析该区储层注水后容易变化的主客观因素,建立储层物性解释模型,使储层的非均质研究由静态向动态方向发展,为精细研究高含水期剩余油分布规律奠定基础。

自然资源要素智能解译研究进展与方向

自然资源要素解译是自然资源调查监测工作中最基础、投入工作量最大、技术难度最高的工作,存在可解译类别少、要素边界不准不全、类别属性可靠性不高等诸多挑战。本文从自然资源调查监测应用场景出发,首先从要素解译到场景要素耦合的解译范式、数据驱动到知识驱动的解译方法、人工目视到人机协同的解译手段3个方面阐述了自然资源要素自动解译的研究进展;然后探讨了6个重点研究方向及其研究内容,包括场景要素耦合解译、知识驱动语义理解、人机协同智能解译、内外一体与三维环境解译、关键参数精准计算与定量反演,以及高可信质量控制与真实性验证;最后进行了总结与展望,以期为自然资源要素智能解译研究提供思路,推动解决困扰遥感影像自动解译方法在自然资源调查监测业务应用中的瓶颈问题。

AVO多参数反演在气层检测和储层定量研究中的应用

利用Ａｋｉ和Ｒｉｃｈａｒｄｓ的近似方程进行多种ＡＶＯ岩性参数的反演，可获得零偏移距纵波剖面、横波剖面、纵波和横波相对速度剖面、气层指示剖面和截距与斜率乘积剖面，用于地质解释。为了获取定量的岩石物理参数，将反演得到纵波零偏移距剖面和经重采样后的反射横波剖面进行合成声波测井处理，得到纵、横波速度剖面。利用纵、横波速度与孔隙度和泥质含量的数学模型及已得到的纵、横波速度，定量计算储层的孔隙度和泥质含量，并得到相应的孔隙度和泥质含量剖面。利用不同地区的两个研究实例，按Ｓ．Ｒ．Ｒｕｔｈｅｒｆｏｒｄ对气藏的定义对其进行分类，并利用ＡＶＯ反演结果说明其不同的ＡＶＯ特征。并系统地说明了上述剖面在气藏检测和储层定性→定量解释中的应用。

基于随钻录井资料确定页岩气储层参数

为了达到随钻快速识别与评价页岩气的录井显示目的,基于页岩气储层同时具有致密砂岩、致密碳酸盐岩油气层的特征,运用岩屑伽马录井、地层元素录井及钻时资料识别页岩气储层岩性的方法与标准,提出了运用气测及综合录井资料确定地层孔隙度与渗透率、地层压力与地层破裂压力、地层含气饱和度与地层含气量、气层产能预测的方法,丰富了录井页岩气解释评价方法与手段。对中国石油化工股份有限公司首口页岩气勘探试验井———J111井录井资料进行了分析和探讨。结果表明:上述方法为低孔隙度低渗透率复杂孔隙结构的页岩气层的精细解释提供了参考,有较强的实用性,满足了页岩气录井现场快速解释评价的需要。