储层油气性的判别是石油勘探与开发的核心环节，准确划分油层、水层和干层是制定石油开采方案的重要依据，但由于储层地质赋存条件的复杂性以及划分影响因素的多样性，使得储层油气性的判别与划分极为困难，也降低了石油勘探与开发的效率。本文在前人研究成果的基础上，基于大量现场观测资料，综合理论分析、数学建模和实测验证等方法，对储层油气性的判别及划分进行了系统研究。

首先，基于前人已有的研究成果以及大量的工程实践选取孔隙度、密度、含水饱和度、渗透率、泥质含量作为影响储层油气性识别的主要因素，并从定性和定量两方面对其进行了分析研究。

其次，基于灰狼优化算法与支持向量机的耦合理论对储层油气性进行了判别，得到油层的有效判别率为76.4%，考虑到灰狼优化算法易于陷入局部最优的不足，故采用差分灰狼优化算法对系统参数进行优化，并再次采用差分灰狼进化算法与支持向量机的耦合对储层油气性进行判别，得到油层的有效判别率为87.64%，显著提高了判别精度。

最后，为进一步提高石油的开采率，在已有储层油气性判别结果的基础上，利用属性数学理论建立了油层等级划分模型，通过构造油层各项评价指标的等级划分标准，利用属性综合评价理论对某样本数据属于某个等级进行预测，并选取陕西某油田实测数据对模型进行了验证，所得评价结果与现场实测基本一致，充分证明了该模型的有效性。

综合理论分析、数学建模和实测验证的方法为储层油气性的判定、油层的等级划分提供了理论依据，对确定具有工业开采价值地区、制定石油开采方案进而实现最佳的经济效益具有重要意义。

The identification of oilbearing reservoir is the core link between petroleum exploration and development, and the accurate division of oil, water and dry layers is the basis for making production plans. However, it's difficult to distinguish and classify the oil-bearing reservoir 

due to the complexity of geological and the diversity of influencing factors, it greatly reduces 

the efficiency of oil. Comprehensive theoretical analysis,mathematical model and actual measurement are used to discriminate the oil-bearing reservoir based on previous researchresults and a large number of observation data.  

Firstly, the density, porosity, permeability, water saturation and shale content are selected as the evaluation indexes of reservoir oil and gas properties based on previous research results and a large number of engineering practices, and it has been analyzed and researched from both qualitative and quantitative aspects.

Secondly, the effective discrimination rate of the oil layer is 76.4% based on the coupling theory of grey wolf optimization algorithm and support vector machine, but consider that the grey wolf algorithm is easy to fall into the local optimization, get the system parameters according to differential grey wolf optimization algorithm, and then obtain the rate is 87.64% by the coupling of the differential grey wolf evolution algorithm and the support vector machine, which significantly improves the discrimination accuracy.

Finally, in order to further improve the oil recovery rate, with the identical results, the oil layer classification model was established by attribute mathematical theory, and the classification standards of various evaluation indicators of the oil layer are constructed, and a sample data belongs to which level can be predicted by the attribute comprehensive evaluation theory.The model was validated by selecting the measured data from an oil field in Shanxi, the evaluation results obtained are basically consistent with the field measurements, which fully proves the effectiveness of the model.

Comprehensive theoretical analysis, mathematical modeling, and actual measurement verification methods provide a theoretical basis for the determination of reservoir oil and gas properties and the classification of oil layers, which is important for determining areas with an industrial production value, formulating oil production plans, and realizing the best economic benefits.

[1]邹文, 贺振华, 陈爱萍, 等. 定量交会图技术及其在流体识别中的应用[J]. 石油物探, 2008(1): 45-48.

[2] 韩济全. 用孔隙度与含水饱和度交会图识别储层流体性质[J]. 勘探地球物理进展, 2005(4): 294-296.

[3] 徐德龙, 李涛, 黄宝华, 等. 利用交会图法识别国外油田岩性与流体类型的研究[J]. 地球物理学进展, 2012, 27(3): 1123-1132.

[4] 张颖, 杨丽丽, 陈木银, 等. 阵列感应侵入因子法与声波-电阻率交会图相结合识别镇北地区延安组低阻油层[J]. 西安石油大学学报(自然科学版), 2014, 29(3): 8-14.

[5] 王勃力, 赫文昊, 王小锋, 等. 重叠图法在储层油水层识别中的应用[J]. 应用能源技术, 2016(8): 7-9.

[6] 郭顺, 王震亮, 张小莉, 等. 陕北志丹油田樊川区长6~1低阻油层成因分析与识别方法[J]. 吉林大学学报(地球科学版), 2012, 42(1): 18-24.

[7] 申怡博, 张小莉, 孙佩, 等. 定边东韩长2油层组复杂油水层识别技术[J]. 西北大学学报(自然科学版), 2010, 40(1): 121-125.

[8] 赵长举, 李纪森. 张天渠油田储层测井响应特征及解释模型的改进[J]. 勘探地球物理进展, 2005(04): 297-299+10.

[9] 邓宇. 大情字井葡萄花油层复杂流体测井综合解释[D]. 大庆：东北石油大学,2018.

[10] 冯璐珈. 储层物性参数测井解释模型的建立[D]. 杭州：浙江大学,2006.

[11] 白洋, 谭茂金, 肖承文, 等. 致密砂岩气藏动态分类委员会机器测井流体识别方法[J]. 地球物理学报, 2021, 64(5): 1745-1758.

[12] 应芳芳, 李朋. Fisher判别分析法在油页岩识别中的应用[J]. 长江大学学报(自科版), 2014, 11(14): 16-20+4.

[13] 罗德江. 基于核Fisher判别的碎屑岩储层流体识别[J]. 地球物理学进展, 2013, 28(4): 1919-1924.

[14] 卢志远, 秦军, 陈雪昆, 等. 模糊综合评判法在昌吉油田吉7井区梧桐沟组油层分类中的应用[J]. 内蒙古石油化工, 2015, 41(11): 137-140.

[15] 王晔, 王向公, 沈华, 等. 基于模糊综合评判的储层流体识别技术[J]. 山东理工大学学报 (自然科学版), 2015, 29(3): 56-59.

[16] 陈清华, 程祥, 王晶, 等. 人工神经网络在闵北断块低阻油层识别上的应用[J]. 海洋地质前沿, 2015, 31(11): 52-57.

[17] 王晓飞, 陈鸣, 申小虎, 等. 神经网络交会图在低阻油层流体识别中的应用[J]. 石化技术, 2019, 26(11): 122-123.

[18] Mercer J. Xvi. function of positive and negative type, and their connection the theory of integral equations[J]. Philosophical transactions of the royal society of London. Series A, containing papers of a mathematical or physical character, 1909, 209(441-458): 415-446.

[19] Ma Aizerman. Theoretical foundations of the potential function method in pattern recognition learning[J]. Automation and remote control, 1964, 25: 821-837.

[20] Buser B.A training algorithm for optimal margin classifier[C]. The 5th Annual ACM workshop on computational learning theory, 1992, 1992: 144-152.

[21] Corina Cortes, Vladimir Vapnik. Support-vector networks[J]. Machine learning, 1995, 20(3): 273-297.

[22] Thorsten Joachims. Text categorization with support vector machines: learning with manvrelevant features[C]. European conference on machine learning springer, berlin, heidelberg, 1998: 137-142.

[23] 朱立云, 刘媛华. 支持向量机在意大利葡萄酒种类识别中的应用[J]. 软件导刊, 2020, 19(11): 29-32.

[24] 胡谦锋, 陈沁磊. 基于支持向量机的图书分类管理研究[J]. 计算机与数字工程, 2021, 49(1): 202-207.

[25] 袁颖, 于少将, 王晨晖, 等. 基于网格搜索法优化支持向量机的围岩稳定性分类模型[J]. 地质与勘探, 2019, 55(2): 608-613.

[26] 刘小生, 章治邦. 基于改进网格搜索法的SVM参数优化[J]. 江西理工大学学报, 2019, 40(1): 5-9.

[27] 纪洁, 胡汉, 高远, 等. 基于遗传算法优化参数的支持向量机风电功率预测[J]. 电子测试, 2020(21): 32-35.

[28] 陈果. 基于遗传算法的支持向量机分类器模型参数优化[J]. 机械科学与技术, 2007(3): 347-350.

[29] 刘昆. 粒子群算法优化支持向量机的网络流量混沌预测[J]. 现代电子技术, 2019, 42(2): 120-123.

[30] 赵倩, 陈芳芳, 甘露. 基于改进粒子群算法优化支持向量机的风电功率预测[J]. 电气技术, 2020, 21(12): 12-16.

[31] 吴璇, 薛峰, 余敏. 基于蚁群算法的支持向量机室内蓝牙标定定位[J]. 江西师范大学学报(自然科学版), 2020, 44(2): 148-152.

[32] 高雷阜, 张秀丽, 王飞. 改进蚁群算法在SVM参数优化研究中的应用[J]. 计算机工程与应用, 2015, 51(13): 139-144.

[33] Mirjalili, Seyedali, Mirjalili, et al. Grey wolf optimizer[J]. Advances in engineering software, 2014, 69: 46-61.

[34] Mostafa, Eman, Abdel-Nasser, et al. Application of mutation operators to grey wolf optimizer for solving emission-economic dispatch problem[C]. 2018 international conference on innovative trends in computer engineering, 2018: 278-282.

[35] Jiang Tianhua, Zhang Chao. Application of grey wolf optimization for solving combinatorial problems: job shop and flexible job shop scheduling cases[J]. IEEE, 2018, 6: 26231-26240.

[36] Yogapriva J,Nithva B. Based grey wolf optimization for effective medical image retrieval system[C]. 2018 International conference on communication and signal processing, 2018: 1046-1050.

[37] Jitkongchuen, Duangjai, Phaidang, Pongsak. Grey wolf algorithm with borda count for feature selection in classification[C]. 2018 3rd international conference on control and robotics engineering, 2018: 238-242.

[38] Chen Huiling, Cui Xiaojun, WeiYan, et al. An improved grey wolf optimization strategy enhanced SVM and its application in predeciting the second major[J]. Mathematical prob- lems in engineering, 2017: 12.

[39] Fahad M, Rehman Z-U, Khan S, et al. Grey wolf optimization based clustering algorithm for vehicular ad-hocnetwork[J]. Computer & electircal engineering, 2018, 70: 853-870.

[40] Ibrahim Rehab Ali, Abd Elaziz Mohamed, Lu Sonfeng. Chaotic opposition based grey-wolf optimization algorithm based on differential evolution and disruption operator for global optimization[C]. Expert systems with applications, 2018, 108(1-27).

[41] Carlos Galvan-Tejada, Laura Zanella-Calzada, Hamurabi Gamboa-Rosales. et al. Depression episodes detection in unipolar and bipolar patients:A methodology with feature extraction and feature selection with genetic algorithms using activity motion singal as information source [C]. Mobile information systems, 2019, 12.

[42] Shubham Gupta, Kusum Deep.A novel random walk grey wolf optimizer[C]. Swarm and evolutionary computation, 2019, 44: 101-112.

[43] Shubham Gupta, Kusum Deep.A hybrid self-adaptive sine cosine algorithm with opposition based learning[C]. Expert systems with applications, 2019, 119: 201 -230.

[44] 燕志星, 王海瑞, 吕维宗. 基于GWO-SVM算法的滚动轴承故障诊断[J]. 化工自动化及仪表, 2019, 46(10): 806-810.

[45] 熊魁, 岳长喜, 刘冬梅, 等. 基于IGWO算法优化的SVM模拟电路故障诊断[J]. 微电子学与计算机, 2019, 36(1): 16-21.

[46] 魏燕明, 甘旭升, 刘飞, 等. 基于灰狼优化算法的军用运输机航路规划方法[J]. 火力与指挥控制, 2021, 46(01): 155-161.

[47] 陈金红, 程刚. 灰狼优化算法投影寻踪模型在云南省水量分配中的应用[J]. 三峡大学学报(自然科学版), 2016, 38(5): 29-35.

[48] 徐松金, 龙文. 基于随机收敛因子和差分变异的改进灰狼优化算法[J]. 科学技术与工程, 2018, 18(23): 252-256.

[49] Luo K.Enhanced grey wolf optimizer with a model for dynamically estimating the location of the prey[J]. Applied soft computing, 2019, 77: 225-235.

[50] Gaidhane P J,Nigam M J.Ahybird grey wolf optimizer and artificial bee colony algorithm for enhancing the performance of complex systens[J]. Journal of computational science, 2018, 27: 284-302.

[51] Majeed M, Patri S. A hybrid if WOA and mGWO algorithms for global optimization and analog ciruit design automation[J]. The international journal for computation and mathematics in electrical and electronic engineering, 2018.

[52] 秦寿康. 综合评价原理与应用[M]. 北京: 电子工业出版社, 2003: 215-224.

[53] 程乾生. 属性数学-属性测度和属性统计[J]. 数学的实践与认识, 1998, 22 (2): 97-107．

[54] 程乾生. 属性识别理论模型及其应用[J]. 北京大学学报, 1997, 33(1): 12-20．

[55] 杜志刚, 张小东, 王晓东. 基于属性数学理论的隧道瓦斯突出危险性评价[J]. 地下空间与工程学报, 2019, 15(06): 1866-1873.

[56] 文畅平, 白银涌, 孙政, 陈宗辉. 基于属性数学理论的排土场滑坡风险预测及分级[J]. 自然灾害学报, 2016, 25(06): 158-166.

[57] Storn R, Price K.Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of global optimization, 1997, 11(4): 341- 359.

[58] Yan Zhang, Qiu Junbo, Zhang Yonggang, et al. The adoption of a support vector machine optimized by GWO to the prediction of soil liquefaction[J]. Environment- al earth sciences, 2021, 80(9): 1-9.

[59] Su Xiyuan, Cao Changqing, Zeng Xiaodong, et al. Application of DBN and GWO-SVM in analog circuit fault diagnosis[J]. Scientific reports, 2021, 11(1): 7969-7969.

[60] Li Lingling, Liu Jiaqi, Zhao Weibing, et a1. Fault diagnosis of high speed brushle-ss permanent magnet DC motor based on support vector machine optimized by mo-dified grey wolf optimization algorithm[J]. Symmetry, 2021, 13(2).

[61] Zhu Aijun, Xu Chuanpei, Li Zhi, et a1. Hybridizing grey wolf optimization with differential evolution for global optimization and test scheduling for 3D stacked SO-C[J]. Journal of systems engineering and electronics, 2015, 26(2): 317-328．

[62] 苏金明.MATLAB工具箱应用[M]. 北京：电子工业出版社, 2004.