致密砂岩气藏裂缝测井评价方法研究

发布时间：2018-05-10 14:48

本文选题：致密砂岩 + 测井　；参考：《长江大学》2015年硕士论文

【摘要】：随着油气的不断开采,目前高孔高渗的储层愈来愈少,而低孔低渗的储层成为了研究的重点；随着资源的缺乏以及技术水平相应提高,页岩气、砂岩致密气也成为了如今探测开采的焦点。在这两类气藏开采的过程中,裂缝的评估与预测是重中之重。对于裂缝的预测评估而言,现在一般是通过成像测井来判断裂缝发育程度,而成像测井因为成本原因,不能广泛大量的在油田里运用,因此开始寻求通过常规测井的各种手段与方法来判断裂缝发育程度。目前大多数的常规测井手段基本上都是利用裂缝在各种曲线上的响应特征诸如电阻率,声波等来识别裂缝,都是利用测井曲线数值上的差异来进行识别。但是单一的运用常规测井资料来预测裂缝发育与否仍然具有一定的局限性,而单一的依靠成像测井资料来识别又不现实,因此需要更加实际有效的方法。本文以SH油田的区块两个层段作为研究区,研究区内平均孔隙度为7.51%,平均渗透率为0.20mD,属于致密砂岩气藏。首先以区块内的薄片资料分析研究目的层的岩性,分别利用交会图法、模糊数学法以及神经网络法识别岩性。交会图法与神经网络法识别符合率均达到80%以上,运用交会图法模版能准确快捷的识别出区域岩性,将区域内的砂岩细化分为岩屑砂岩、岩屑石英砂岩以及石英砂岩三种岩性。在不同岩性的基础上,分别通过两个不同的角度来预测裂缝发育程度。利用电成像测井资料来刻度常规测井资料上裂缝的响应特征,总结出该区块裂缝测井响应特征；再利用对应的测井曲线计算出裂缝参数,通过R/S分形法计算对应测井曲线的分维数值,将其与成像测井资料相对照得到裂缝发育不同程度的分维数值范围以及对应裂缝参数范围；利用5口电成像测井资料的61个层位、岩心分析资料及常规测井资料利用R/S分形法对储层裂缝进行发育程度的划分。得到相应的图版。研究区内发育裂缝时,岩屑石英砂岩分维数大于1.15,石英砂岩与岩屑砂岩分维数大于1.1,实际处理11口井19个层位,平均符合率达到73.6%；利用偶极声波成像测井资料建立不同岩性的纵横波时差模型,通过纵横波时差模型计算相应的岩石力学参数,再将岩石力学参数的值与裂缝参数结合计算出不同裂缝发育程度的岩石力学参数值的范围标准,充分利用3口偶极声波井的56个层位的资料、岩心分析资料及对应常规测井资料建立不同岩性的岩石力学参数模版,而在岩石力学参数里面,杨氏模量来预测裂缝发育在该区块最为适用,研究区内存在裂缝时,岩屑石英砂岩的杨氏模量小于31,岩屑砂岩的杨氏模量小于36；实际处理5口井15个层位,平均符合率达到66.7%,能够初步满足实际需求。
[Abstract]:With the continuous exploitation of oil and gas, there are fewer and fewer reservoirs with high porosity and high permeability, and the reservoirs with low porosity and low permeability become the focus of research. Tight sandstone gas has also become the focus of exploration and mining. In the process of producing these two types of gas reservoirs, the evaluation and prediction of fractures is the most important. For the prediction and evaluation of fractures, it is generally used now to judge the degree of fracture development through imaging logging, and imaging logging cannot be widely used in oil fields due to cost reasons. Therefore, we began to look for various means and methods of conventional logging to judge the degree of fracture development. At present, most conventional logging methods use the response characteristics of fractures on various curves, such as resistivity and sound waves, to identify fractures, and use the difference of logging curves to identify fractures. However, the single use of conventional logging data to predict fracture development or not still has some limitations, but it is not realistic to rely solely on imaging logging data to identify fractures, so a more practical and effective method is needed. In this paper, two layers in the block of SH oilfield are taken as the study area. The average porosity and permeability in the study area are 7.51 and 0.20mDrespectively, which belong to the tight sandstone gas reservoir. Firstly, the lithology of the target layer is analyzed with the slice data of the block, and the lithology is identified by cross plot method, fuzzy mathematics method and neural network method respectively. The coincidence rate of cross plot method and neural network method is more than 80%. The lithology of the area can be identified accurately and quickly by using the cross plot template. The sandstone in the area can be divided into three types: lithic sandstone, lithic quartz sandstone and quartz sandstone. On the basis of different lithology, the degree of fracture development is predicted from two different angles. The response characteristics of fractures on conventional logging data are calibrated by using electrical imaging logging data, and the logging response characteristics of fractures in this block are summarized, and the fracture parameters are calculated by using the corresponding logging curves. The fractal dimension of the corresponding logging curve is calculated by the R / S fractal method, and compared with the imaging logging data, the fractal dimension value range and the corresponding fracture parameter range of different degrees of fracture development are obtained, and the 61 layers of 5 electrical imaging logging data are used. Core analysis data and conventional logging data are used to divide the development degree of reservoir fractures by using R / S fractal method. Get the corresponding plates. In the development of fractures, the fractal dimension of lithic quartz sandstone is greater than 1.15, the fractal dimension of quartz sandstone and lithic sandstone is greater than 1.1, and the 19 layers of 11 wells are actually treated. The average coincidence rate is 73.6.The longitudinal and shear wave moveout models with different lithology are established by using dipole acoustic imaging data, and the corresponding rock mechanics parameters are calculated by the P-S wave moveout model. Then combining the values of rock mechanics parameters with fracture parameters to calculate the range standard of rock mechanics parameters with different fracture development degrees, the data of 56 layers of 3 dipolar acoustic wells are fully utilized. The core analysis data and the corresponding conventional logging data are used to establish the lithologic parameters template of different lithology. Among the rock mechanics parameters, Young's modulus is the most suitable for predicting fracture development in this block. When there are fractures in the study area, The Young's modulus of lithic quartz sandstone is less than 31, the Young's modulus of lithic sandstone is less than 36, and the average coincidence rate of treating 5 wells and 15 layers is 66.7, which can meet the actual demand.
【学位授予单位】：长江大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P618.13

【参考文献】