【作者】 喻鹏；

【导师】 唐仲华；

【作者基本信息】 中国地质大学 ， 油气田开发工程， 2014， 博士

【摘要】 高凝油油藏是一类较为特殊的油藏类型,其驱替过程与一般稀油油藏的区别之一就是要考虑高凝油自身的特性。针对该类型油藏的表征、空间预测、三维地质建模等研究方面还比较薄弱,缺乏系统的研究和深入分析。辽河油田静观2区块是一个典型的高凝油油藏,其原油凝固点高于44℃,平均含蜡量38.34%,属世界少有原油类型。本文以静观2区块高凝油油藏为例,基于地质、岩心、钻井、地震、测井、测试和化验分析等大量基础资料,在层序地层学、沉积学、储层地质学、测井地质学、地质统计学、岩石学、石油开发地质学等理论的指导下,开展针对研究区储集层的高分辨率地层对比、沉积体系与沉积相分析、物性及流动单元预测表征和水驱流线数值模拟及热采模拟的综合研究。研究过程同时将地学数据挖掘中的新概念、新方法和新技术同计算机程序应用相结合,全面准确地对静观2区块储集层未取心井段流动单元进行了预测识别,进而通过指示模拟技术刻画表征了研究区流动单元三维精细地质模型,并在流动单元识别、划分的基础上进行渗透率分单元回归解释,同时将地质模型植入地质体,最终完成了三维定量化地质物性模型,形成了一套针对该类型油藏储层三维定量化地质建模的思路和方法,为下一步该种类型油藏油气勘探和储层预测提供了借鉴依据。通过基于流动单元的流线数值模拟,弄清了静观2区块剩余油同注采流线及流动单元的配伍关系。在上述研究基础上,结合静观2块油藏注冷水开发对储层的伤害实验试验资料及注水井地层带温度场数学模型,对区块最佳地面注水温度进行寻优。同时结合油藏测试资料,通过热采物理模拟研究分析原油的高含蜡、高凝固点特性对原油渗流过程的影响,采用数值模拟方法,研究了不同注水温度对油藏开发效果的影响,并对各开发方案进行了预测。论文总共分为8个部分,第一部分主要总结了流动单元的国内外研究现状及发展趋势,同时对目前主流的一些运用在流动单元预测上的地学数据挖掘方法进行了归纳和总结。针对工区储层的认识,总结了工区储集层流动单元的认识上的一些问题。进一步总结归纳了油藏剩余油分布模式、高凝油油藏开发模式以及高凝油油藏剩余油分布规律的研究等一些影响剩余油分布的主要因素以及研究区内剩余油分布研究仍存在的若干问题,同时对文章后续研究内容提出了规划和展望。第二个部分应用Cross的旋回性层序地层学分析方法进行了静观2区块高分辨率地层格架的构建,采用地质-地震-测井联合解释的技术手段,划分、识别和对比了不同旋回的时间地层单元。通过岩性垂向序列、测井地层综合响应和地震反射信息,确定静观2区块沙三段时期存在的5个长期基准面旋回组成的层序地层格架,其中SQ5相当于S31地层,在本区多被剥蚀掉；SQ4由3个中期基准面旋回、8个短期基准面旋回组成；SQ3由3个中期基准面旋回、7个短期基准面旋回组成；SQ2由4个中期基准面旋回、4个短期基准面旋回组成,即在沙河街组沉积时期主要发生了3次明显的水进。针对静观2区块沉积砂体,研究中首先分析了研究区沉积背景、沉积环境、物源供给以及岩石电性特征,在此基础上,重点进行了沉积体系的划分、识别和沉积模式的分析。在资料综合处理,构造、沉积相综合研究以及大量人工制图分析的基础上,建立了静观2区块构造、沉积相三维定量化模型。在构造地质建模中,采用层厚插值面控制窜层难点,最终建立的构造模型保持了小层之间构造特征及构造关系在三维空间上的协调性。沉积微相模型的构建则采用沉积趋势面面控的序贯指示模拟方法来构建,模拟结果较理想。第三个部分依据研究靶区油藏地质情况,依据压汞分析资料建立流动层带指标FZI同排驱压力(Pd)的关系,优选同FZI相关性较高的参数同时参与流动单元划分聚类,得到HU#5、HU#6、HU#7及HU#8四类基本方案。根据Spearman非参数相关系数法对取心井的测井参数同FZI值进行相关分析,优选测井曲线构建测井频率交会区间,结合贝叶斯推论编写软件计算后验概率基库并对储层未取心井段的流动单元进行预测和回判验证,并同成熟的神经网络模式识别技术判别结果作对比,预测正判率较接近,识别效果好,具有一定推广应用价值。流动单元井数据准备完善之后,采用序贯指示模拟对工区储层进行流动单元三维定量化建模,同时分单元分层位拟合变差及概率。最终进一步结合地质理念分动、静态两个方面对工区流动单元预测模型进行验证分析。动态验证过程共总结9口生产井数据,结合井史分析模型合理性；静态验证则主要通过两口井联井剖面上岩性、孔喉半径及流动单元的对应情况来进行分析。第四个部分主要构建静观2区块孔隙度、渗透率解释模型,同时将数学模型植入地质体以建立研究区三维定量化物性模拟模型。孔隙度解释模型的回归采用的是二元回归法,回归方程应用效果较好。以沉积相控制为前提,逐层逐相对孔隙度数据进行分析,采用序贯高斯模拟的方法对研究区孔隙度分布进行定量模拟。渗透率解释模型的建立则摒弃了传统的预测方法,而是根据储层岩心流动单元的分类,运用指数及幂律关系模型对每类储层分别建立渗透率孔隙度关系式,预测结果整体乐观,最终优选拟合优度较好的关系模型参与渗透率的预测计算,并将数学模型植入地质体。分别从平面和垂向上验证渗透率三维模型的精确性,结果表明,渗透率三维模型的总体分布特征同工区沉积相发育规模吻合,抽稀后的重建剖面信息亦证明模型的可靠性。综合分析研究区各种地质资料,运用确定性建模手法构建了储层净毛比模型,为后续数模工作打下基础。第五个部分运用油藏工程的方法对工区储层可采储量及最终采收率进行了预测。从总结归纳四种不同的递减模型出发,对比分析了不同模型参数的意义,同时寻优适合静观2块高凝油油藏产量递减规律的模型。针对研究区实际开发数据对比计算发现,Li-Horne模型模拟结果小于Arps模型指数模拟计算结果,Correa模型计算结果与数模预测值拟合较好。研究结果表明,Arps模型的运用受限,一般以没有重大措施调整为基础；水驱高凝油含水上升过快则可能导致Li-Horne模型计算结果偏低；同时在工区油藏条件的基础上扩充了Correa模型β值的选值范围,即超出-1≤β≤0的值域范围同样适用,且p值越低,递减规律越接近高凝油油藏递减规律,在同类型油藏当中具有一定推广应用价值。第六个部分总结了油藏开发历程及油水两相条件下流线模型的数学模型同时确定了流线的推导过程。最终结合静观2区块地质模型、流体及渗流特征参数、生产数据,运用流线数值模拟器对研究区油水运动规律进行模拟及拟合。通过对油藏开发动态分析和基于流动单元的油藏流线数值模拟,较准确地预测了剩余油分布,同时分析了剩余油同水驱注采流线及流动单元的配伍关系,认为油藏经历5次重大调整至今,水驱流线基本覆盖全区,而剩余油主要富集在注入流线波及较差的流动单元片区(层内层间非均质剩余油)、不同类型流动单元的交触位置、注采井网不完善区域(井间滞留区)以及断层边部地区等。针对静观2区块开发后期剩余油的复杂分布情况,结合区内注水流线及流动单元同剩余油分布的配伍关系及地质资料进行总结归纳,分别从地质因素、油水互驱因素及开发因素这三个方面总结了影响剩余油分布的主要控制原因。第七个部分通过油藏注冷水开发对储层伤害的试验研究、热采物理模拟及热采数值模拟研究,指出了静观2区块部分生产井采收率低下主要是与流体性质有关,生产井和注水井近井地带储层冷伤害严重,油藏流体呈非牛顿流体状态。地层伤害的主要原因是由于注采井井底温度下降导致气体膨胀,从而导致原油中的石蜡结晶、析出沉淀。热物理模拟分别进行了水驱油实验及相渗实验,通过实验发现高凝油油藏水驱油效率明显受原油粘度和实验温度的影响,随着原油粘度的降低和温度的升高,驱替效率逐渐提高；另外驱替效率与岩石本身结构有关,分选差、中值半径小、泥质含量高的岩心,水驱油效率低；随着实验温度的升高,高凝油油藏相渗曲线形态逐渐向右偏移,两相区变宽,等渗点含水饱和度增加,束缚水饱和度升高,残余油饱和度则显著降低：当温度低于析蜡点温度时,相对渗透率曲线随着压力梯度增大向右移动,两相区变宽,增加地层压力梯度可以提高采收率。热物理实验使得工区油藏流体性质认识变得更加清晰,同时实验数据亦为后面热采数值模拟做准备。归纳考虑水井中与油层内传质传热的注水井地层带内温度场的数学模型,通过编程计算出不同注水温度下井筒及井筒外地层的深-温交汇图版,从研究结果可以看出,注入水在到达目的层之前,热损影响所占权重较大。到达目的层后,由于受油藏温度影响,井筒温度逐渐回升,筒外地层温度在注入水和油藏本身热能的共同作用下,呈现恢复趋势。注入温度80℃图版反映出注水温度在此温度以上,则可保持井筒及筒外地层温度维持在平均析蜡温度(60℃)以上,而温度超过80℃,油层温度增高效果差别没有太大异常,该现象亦在后面热采数模中得到了验证。继续增高注入温度,势必要增加供热成本,故优选80℃作为地面最佳注水温度,即可在兼顾生产成本限制的同时降低储层冷伤害,以达到增产目的。研究区热采数值模拟对比分析了不同注水温度条件下温度对油藏开采效果、井底压力以及油层吸水能力的影响。同时分析了不同方案油藏温度场同剩余油场的配伍关系,常规水驱条件下含油饱和度平面展布图展示了研究区若干注水井附近的小层注水冷带,伤害储层并影响驱油效率。模拟预测结果同样表明,在优势地面注水温度下,能在兼顾经济效益的同时使油藏温度较长时间维持在析蜡温度以上,对提效增产有利。第八个部分则是对整个论文的架构进行全面的总结概括并提出相应建议。本文针对辽河油田静观2区块高凝油油藏严重的非均质性导致油藏在开发过程中存在的各种矛盾,展开流动单元细分和流动单元剩余油分布研究,形成了一套较完整的高凝油油藏储层流动单元划分、对比、预测、表征及应用的配套理论和方法技术；在流动单元分类方法、流动单元空间预测表征手段以及流动单元平面分布的评价方法等方面进行了探索；在剩余油流线数值模拟及注采流线同剩余油配伍关系等方面取得了一定进展,对同类型油藏流动单元的预测表征及剩余油分布研究等具有一定的指导意义。更多还原

【Abstract】 High pour point oil reservior, a kind of special reservoir type, has to be considered its own characteristics, as a significant difference from general thin oil reservoir in its displacement process. While the research about characterization of this reservoir type, spatial prediction, and the3D geological modeling is still relatively weak, there is a lack of systematic studies and deep analyses. Block Jingguan-2of Liaohe Oilfield is a typical high pour point oil reservoir. The freezing point of crude oil is42℃-64℃, and the wax content is37.5%, which is of the world’s rare type. This paper takes Block Jingguan-2high pour point reservoir as an example and launches a series of researches such as the high resolution stratigraphic correlation of reservoir in the study area, sedimentary system and sedimentary phase analysis, physical property and characterization of hydraulic unit prediction, and water flooding streamline numerical simulation and the comprehensive research of thermal simulation, which is based on a large amount of basic data about geology, core, drilling, seismic, logging, testing and laboratory analysis, and also under the guidance of the theory about the sequence stratigraphy, sedimentology, reservoir geology, logging geology, geological statistics, petrology, petroleum development geology. During the study process, new concepts, new methods and new technology application in data mining research, combined with the computer programming, has comprehensively and accurately identified non-coring section hydraulic unit of Block Jingguan-2reservoir. Then through the indicator simulation technique, the hydraulic unit of the study area was characterized by the description of3D fine geologic model, and permeability was regressed by unit which based on hydraulic unit recognition and division. Moreover, geological body was planted into geological model, and finally the3D quantitative geological model properties were completed to form a set of reservoir3D quantitative geological modeling ideas and methods of this type of reservoir oil and gas exploration for the next step and also provided a reference for reservoir prediction. Based on the streamline numerical simulation of hydraulic unit, we make clear the compatibility relation between the remaining oil in Block Jingguan-2and the injection-production flow line and hydraulic unit. Above all, combined with water injection development of the static damage experiment test data and injection wells formation temperature field mathematical model of reservoir in Block Jingguan-2reservoir, this study searched for the best ground water temperature of the block; combined with reservoir test data and analysis the high wax content and high solidifying point of crude oil simulation on the characteristics of the impact on crude oil seepage process, this study focused on the effects of water temperature on the reservoir development and made the prediction of each development plan by using the numerical simulation method.This paper is divided into8parts. The first part mainly summarized the domestic and international research status and the development trend of the hydraulic unit, and concluded the geological data mining method which was applied for hydraulic unit prediction as a main trend. According to the understanding of the reservoir, this paper summarized some cognitive problems of reservoir hydraulic units; and further concluded the residual oil distribution mode, distribution law and development mode of high pour point oil, which were the main factors that influenced the residual oil distribution; also introduced some existing problems about residual oil distribution research; then put further research content plans and prospects forward.The second part applied the analysis method of Cross the cyclicity of sequence stratigraphy for the high resolution statigraphic framework in Block Jingguan-2and divided, identified and compared different cycle time stratigraphic unit by the means of the geological-seismic-logging joint interpretation technology. Through the vertical sequence, lithology stratigraphy and logging response, and seismic information, this part confirmed that sequence stratigraphic framework was consisted by5long-term base level cycles of Block Jingguan-2in three periods, among them, SQ5was equivalent of S31stratum which was eroding in this block; SQ4was composed by3medium-term base level cycles and8short-term base level cycles; SQ3was composed by3medium-term base level cycles and7short-term base level cycles; SQ2was composed by4medium-term base level cycles and4short-term base level cycles. Those are3mainly obvious water effects during the deposition period of Shahe Street Group.According to the sedimentary sand body of Block Jingguan-2, this study first analyzed the supply of sedimentary background, sedimentary environment, source and electric character of rocks. On this basis, we focused on the analysis of classification, recognition of depositional system and depositional model.In data processing, based on sedimentary and tectonic phase comprehensive research and amount of artificial drawing, this part set a contemplation of three-dimensional quantitative model of structure and sedimentary phases of Block Jing-2. In the tectonic geological modeling, we used the thickness of interpolation surface to control the interzone difficulties, and the final established structure model which was kept structure characteristics and relationships of each layer in the3D space coordination. The sedimentary microphase model was constructed by sequential deposition trend surface control which could indicate the simulation method to simulate an ideal result.The third part, in terms of the reservoir geology of research target and the mercury intrusion analytic data, made a relationship between the index FZI of flow zone and the displacement pressure (Pd), and optimized parameters which had higher correlation with FZI and involved into the classification and clustering at the same time, to obtain4basic scheme:HU#5, HU#6, HU#7and HU#8.According to the Spearman non-parametric correlation coefficient method, this part made the coring well logging parameters associated with FZI values for correlation analysis and selected2kinds of logging curve to build log frequency range. Combined with Bayesian inference written software, the reservoir non-hydraulic unit coring wells was well predicted and verified. Then the result compared with the recognition technology and discriminant results of mature neural network pattern, and presented the closed prediction hit ratio, which had certain application values.After the hydraulic unit well data was prepared, this study simulated the hydraulic unit3D quantitative modeling of reservoir by using sequential indicator and matched the deviation and probability for every unit layer. At last combined with geological method, we did the validation analysis of the outward hydraulic unit model in the two aspects of dynamic and static geological concepts. In the dynamic verification process, this paper summarized9production wells data combined with lateral analysis model of rationality; static verification was analyzed through2wells and well profiles on the lithology, pore throat radius and the corresponding conditions of hydraulic units.The fourth part mainly built the porosity, permeability interpretation model of Block Jingguan-2, and at the same time presented the mathematical model which implanted geologic body to establish3D simulation model of quantitative compound in the study area. Porosity interpretation model was regressed by the binary regression method, and the effect of the regression equation application was good. On the premise of sedimentary phase control, the paper analyzed the porosity data of every layer and every phase, and adopted the method of sequential Gaussian simulation to simulate the porosity distribution quantitatively in the study area.Permeability interpretation model abandoned the traditional prediction methods, but the index and power law relationship model was applied to establish the permeability-porosity relation for each type of reservoir by using the reservoir core hydraulic unit classification. The result was overall optimistic and showed that the best matching relationship model was participated in the calculation with the prediction of permeability and the geological body was planted into the mathematical model. This part verified the permeability precision of the3D model respectively from the plane and vertical. The results showed that the overall distribution characteristics of the permeability3D model was identical to the development scale of sedimentary phase and the information of the reconstructed profiles also proved the reliability of the model.In the fifth part, the method of reservoir engineering was applied to predict the reservoir recoverable reserves and the ultimate recovery. To begin with the summary of4kinds of different regressive models, this paper contrasted the meaning of different model parameters, and searched for the best model for Block Jingguan-2to express the declining production of high pour point oil reservoir.The data calculation and comparison of the actual development in the study area presented that the simulation results of Li-Home model was less than that of the Arps index model and the calculation results of Correa model matched the predicted values of mathematical model well. The results showed that the limited use of Arps model was not generally based on the adjustment of significant measures; too fast rise of the water content in the high pour point oil through water flooding might lead to the lower calculation results of Li-Home model; the β value range of Correa model was expanded with the condition of the work area reservoir, that was the same reliability beyond the range-1<(3<0, and as β value went lower, the declining law was more closed to the law of high pour point reservoir.The sixth part summarized the mathematical model of hydraulic unit under the condition of the reservoir development course and oil-water two phase. Then combined with the geological model of Block Jingguan-2, fluid and percolation characteristic parameters and production data, this part used the streamline numerical simulator to simulate and match the oil-water movement law.Through dynamic analysis of reservoir development and streamline numerical simulation based on hydraulic units of reservoir, this paper accurately predicted the residual oil distribution, at the same time analyzed the compatibility relation between residual oil and water driven injection-production flow line and mobile cell. The reservoir was thought to go through five major adjustments till now. And the water flooding streamline basically covered the entire area, while the residual oil enriched mainly in injection streamline waves and the poor hydraulic unit area (anisotropic residual oil in inner layers and interior layers), contact position of different hydraulic units, imperfect spacing area of injection-production well (interwell retention area) and the fault edge region.Combined with the compatibility relation between water injection flow and hydraulic unit and residual oil distribution, and geological data, this part summarized the complex distribution of residual oil after developing the Block Jingguan-2, respectively from these three aspects:the geological factors, factors of oil and water alternating flooding and development factors, to summarize the main control factors of residual oil distribution.After doing some researches about the reservoir damaged by injection development, thermal physical simulation and thermal numerical simulation research, the seventh part pointed out that the low recovery of some production wells in Block Jingguan-2was mainly related to fluid properties, and production and injection wells near the wellbore reservoir were damaged seriously, and showed that the fluid had non-Newtonian fluid properties. The reason of the stratum damage was because the bottom temperature of the injection-production well dropped, then leading the gas expansion, paraffin wax crystallization and precipitation. The water flooding experiments and phase permeability experiments were carried out respectively during the thermal physical simulation, and the result showed that the water displacement efficiency of high pour point oil reservoir was affected by viscosity of crude oil and experimental temperature; with the loss of the crude oil viscosity and temperature rise, displacement efficiency gradually improved; displacement efficiency was related to rock structure itself, that with small poor sorting, median radius and high shale content of the core, the water displacement efficiency was low; with the increase of experimental temperature, relative permeability curve shape of high pour-point oil reservoir gradually offset to the right, and two phase zone got width, such as water saturation of isotonic points increased and irreducible water saturation increased, while residual oil saturation was significantly reduced; and when the temperature went below the temperature of wax precipitation point, the relative permeability curve moved right as the pressure gradient increased, two phase zone got widened, and the recovery efficiency improved. Thermal physics experiments made the understanding of reservoir fluid properties in the work area clearer, at the same time the experimental data was well prepared for thermal recovery numerical simulation.Under the consideration of the temperature field mathematical model of wells and formation which could transfer the heat and mass in injection wells between oil layers and through different injection temperature programming calculation of the layer deep-intersection chart, the result showed that before the injected water reached the goal layer, the influence of heat loss weight large. After getting to the destination layer, the wellbore temperature, affected by the temperature of the reservoir, rose gradually, and the stratum temperature outside the wellbore presented recovery trend under the common action of the injected water and reservoir itself. The chat of injection temperature80℃reflected that above this temperature, the layers could keep the stratum temperature at the average wax temperature (60℃), while beyond80℃, it was not that different from the temperature increasing effect of the reservoir, which was also verified in the thermal recovery mathematical model. The heating costs must be increased if we continued to increase injection temperature, therefore80℃was selected to be the best temperature for ground water injection, which both took the production cost into account and reduced the reservoir cold damage to achieve the purpose to increase the production. Thermal recovery model analyzed the reservoir development effect of different injection water temperature, the influence of bottom pressure and reservoir water absorption capacity, and analyzed the compatibility relations between different types of reservoir temperature field and residual oil field. The plane distribution of oil saturation layout reflected that the small layer water injection, under the condition of conventional water flooding, damaged the reservoir by affecting oil displacement efficiency. Simulation results also showed that under the water injection temperature of dominant ground, considering both economic benefits and the longer soaking time of reservoir temperature, it was favorable to enhance the efficiency and increase the production.The eighth part conducted a comprehensive summary for the whole paper framework.While the high pour point oil reservoir of Block Jingguan-2in Liaohe Oilfield existed serious heterogeneity which led to some problems during the development process of the reservoir, this paper conducted the residual oil distribution researches of hydraulic unit subdivision and hydraulic units, and formed a integrated set of theory and method which could apply for the hydraulic unit division, correlation, prediction, characterization and application of high pour point oil reservoir; it also explored the multiple classification methods of hydraulic unit space forecast and evaluation method of plane distribution; it made some progress in residual oil streamline simulation and the compatibility relation between injection-production flow line and residual oil, which had important guiding significance in the same type of reservoir exploration.更多还原