苏里格致密砂岩渗流分析及压裂产能预测方法研究

发布时间：2018-08-23 19:11

【摘要】：致密砂岩气藏目前是最具勘探开发意义的非常规天然气领域。由于致密砂岩孔隙度低、渗透率低、含气饱和度低，渗流机理复杂，建立高精度的储层孔渗饱解释模型困难；非均质性强，低阻气层高低阻水层并存，致使致密砂岩含气性评价困难；生产中压裂措施对地层的影响，给致密砂岩气层的储层压后产能预测带来一系列挑战。开展致密砂岩储层的渗流特征分析、建立高精度解释模型、正确评价含气性、准确预测储层产能，分析研究区的产气、产水分布有助于指导致密砂岩油气田的合理开发。本文以苏里格地区二叠系的盒8段致密砂岩气层为目的层，围绕储层评价以及压裂产能预测展开研究。从储层的储集性分析展开，研究苏里格地区致密砂岩的储集特征、岩石学特征，分析矿物含量对产能的影响，分析含气测井特征。从宏观角度分析孔隙度、渗透率、含气饱和度、气水相对渗透率对致密砂岩渗流的影响。在经验统计的基础上，开展神经网络法建立孔渗饱的解释模型。使用Elman_Adaboost强预测器进行孔隙度的计算，使用SVR进行渗透率和饱和度的计算。从常规测井曲线的图版法出发，建立含气性评价的定性评价指标，利用数学手段即小波分析、GRNN网络曲线重构法对常规测井曲线进行分析建立含气性定量评价指标。最后综合分析压裂产能的影响因素及并预测压裂产能。从压裂产能的理论出发，分析压裂产能的影响因素。重点讨论测井参数和压裂施工参数对压裂产能的影响，并使用R型主成分分析法进行影响因素的降维分析，选出主要的影响因素。基于主成分分析选出的主因子，利用GRNN网络建立单层产能的预测模型。对合试层段产能劈分方法的研究，在单层产能预测模型的基础上建立GRNN单点产能预测模型，利用建立的GRNN单点产能预测模型对苏里格地区致密砂岩气藏进行井中预测，并分析苏里格地区盒8段的产气量、产水量进行预测分析，分析苏里格盒8段的产气、产水的分布特征。
[Abstract]:Tight sandstone gas reservoir is currently the most important unconventional natural gas field with exploration and development significance. Due to the low porosity, low permeability, low gas saturation and complex seepage mechanism of tight sandstone, it is difficult to establish a high precision pore and permeability saturation interpretation model of reservoir. It is difficult to evaluate the gas-bearing property of tight sandstone and the influence of fracturing measures on formation in production brings a series of challenges to the prediction of reservoir productivity after pressure in tight sandstone gas reservoir. The analysis of percolation characteristics of tight sandstone reservoir, the establishment of high precision interpretation model, the correct evaluation of gas-bearing property, the accurate prediction of reservoir productivity, the analysis of gas production and distribution of water production in the study area are helpful to guide the rational development of tight sandstone oil and gas field. This paper focuses on reservoir evaluation and fracturing productivity prediction based on tight sandstone gas reservoir of Bo8 member of Permian in Sulige area. In this paper, the reservoir characteristics and petrological characteristics of tight sandstone in Sulige area are studied. The effects of mineral content on productivity and gas logging characteristics are analyzed. The effects of porosity, permeability, gas saturation and relative permeability of gas and water on the permeability of tight sandstone are analyzed from the macroscopic point of view. On the basis of empirical statistics, a neural network method was developed to establish an interpretation model of pore and osmotic saturation. Elman_Adaboost strong predictor is used to calculate porosity and SVR is used to calculate permeability and saturation. Based on the chart method of conventional logging curve, the qualitative evaluation index of gas-bearing property is established, and the quantitative evaluation index of gas-bearing property is established by means of mathematical means, I. E. wavelet analysis GRNN network curve reconstruction method. Finally, the influencing factors of fracturing productivity and the prediction of fracturing productivity are analyzed synthetically. Based on the theory of fracturing productivity, the influencing factors of fracturing productivity are analyzed. The effects of logging parameters and fracturing operation parameters on fracturing productivity are discussed emphatically. The main influencing factors are selected by using R-type principal component analysis method to reduce the dimension of the factors. Based on the principal factor selected by principal component analysis (PCA), a single layer productivity prediction model is established by using GRNN network. Based on the single-layer productivity prediction model, the GRNN single-point productivity prediction model is established, and the GRNN single-point productivity prediction model is used to predict the tight sandstone gas reservoir in Sulige area. The gas production and water production of section 8 in Sulige area were analyzed, and the distribution characteristics of gas production and water production in section 8 of Sulige box were analyzed.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE312

【参考文献】