Thickness prediction of seismic multi-attributes sand based on association rules and random forests
-
摘要: 地震属性技术是砂体厚度预测的重要手段,由于目前可从地震数据中提取的地震属性种类较多,在利用地震属性技术前,必须优化出对砂体厚度最敏感的地震属性组合,以减少地震属性信息的重复与冗余。为此提出了一种联合关联规则与随机森林回归算法的地震多属性砂体厚度预测方法。随机森林回归算法能够建立地震多属性与砂体厚度之间的非线性关系,并能进行属性选择,但是该算法无法识别地震多种属性中的冗余特征。关联规则能够发现地震属性之间的非线性关联,并能借助卡方检验消除地震属性间的冗余性。分别采用了随机森林回归算法(RFR)、联合关联规则与随机森林回归(AR-RFR)及BP神经网络回归的算法(AR-BP)对滩坝砂岩合成模型和某实际工区进行了砂体厚度预测。对比结果表明,基于关联规则的属性优选得到的属性间相关性低,关联规则与随机森林算法的结合提高了砂体厚度的预测精度。数值实验证明了该方法的有效性。Abstract: Seismic attributes analysis technique is an important tool for sand thickness prediction. Due to the varieties of seismic attributes, the best seismic attributes need to be optimized before the seismic attributes analysis technique is applied to reduce the repeatability and redundancy of the attributes. Therefore, we present an improved random forest regression algorithm combined with associate rules for sand thickness prediction (AR-RFR). Although random forest regression algorithm(RFR) is powerful for the problem characterized for nonlinearity and high dimension in reservoir prediction, it cannot solve attribute reduction problems. The associate rules can find the non-linear relationship among the multi-attributes and can reduce some redundant attributes by means of Chi-squared Test. We apply ordinary RFR, AR-RFR and Neural network regression algorithm combined with associate rules(AR-BP) to a synthetic geological model and a real dataset. The results prove that the selection attributes from associate rules is more efficient than that from random forest. Compared to the drilled wells, AR-RFR has higher precision than RFR and AR-BP. And AR-RFR also can improve the lateral distribution of sand bodies. The method proposed is able to choose efficient seismic attributes and improve prediction of sand thickness.
-
表 1 研究中采用地震属性
Table 1. Seismic attributes used in this study
属性标识 属性名称 属性标识 属性名称 属性标识 属性名称 1 振幅包络 9 导数的瞬时振幅 17 二阶导数的瞬时振幅 2 平均频率 10 主频 18 振幅 3 视极性 11 瞬时频率 19 5/10~15/20滤波器 4 振幅加权余弦相位 12 瞬时相位 20 15/20~25/30滤波器 5 振幅加权频率 13 积分 21 25/30~35/40滤波器 6 振幅加权相位 14 绝对值振幅的包络 22 35/40~45/50滤波器 7 余弦瞬时相位 15 正交道 23 45/50~55/60滤波器 8 振幅导数 16 振幅二阶导数 24 55/60~65/70滤波器 表 2 基于随机森林方法优选的属性之间的相关关系
Table 2. Correlation analysis between various selected seismic attributes from random forest
属性名称 导数 瞬时频率 积分 瞬时相位的余弦 导数 1.000 0 0.348 2 -0.953 8 0.376 5 瞬时频率 0.348 2 1.000 0 -0.356 8 0.293 0 积分 -0.953 8 -0.356 8 1.000 0 -0.317 9 瞬时相位的余弦 0.376 5 0.293 0 -0.317 9 1.000 0 表 3 基于关联规则方法优选的属性之间的相关关系
Table 3. Correlation analysis between various selected seismic attributes from association rules
属性名称 振幅加权频率 振幅加权相位 瞬时振幅的导数 15/20~25/30滤波器 振幅加权频率 1.000 0 0.018 6 -0.316 7 -0.400 9 振幅加权相位 0.018 6 1.000 0 -0.452 5 -0.136 8 瞬时振幅的导数 -0.316 7 -0.452 5 1.000 0 0.328 1 15/20~25/30滤波器 -0.400 9 -0.136 8 0.328 1 1.000 0 -
[1] Chopra S, Marfurt K J J G. Seismic attributes: A historical perspective[J]. Geophysics, 2005, 70(5): 3-28. doi: 10.1190/1.2098670 [2] Leiphart D J, Hart B S. Comparison of linear regression and a probabilistic neural network to predict porosity from 3D seismic attributes in Lower Brushy Canyon channeled sandstones, southeast New Mexico[J]. Geophysics, 2001, 66(5): 1349-1358. doi: 10.1190/1.1487080 [3] Hampson D P, Schuelke J S, Quirein J A. Use of multiattribute transforms to predict log properties from seismic data[J]. Geophysics, 2001, 66(1): 220-236. doi: 10.1190/1.1444899 [4] 万里明, 吴均, 卢军凯, 等. 基于Adam-神经网络的致密砂岩脆性评价方法: 以南堡凹陷高北边坡为例[J]. 地质科技通报, 2020, 39(2): 94-103. http://dzkjqb.cug.edu.cn/CN/abstract/abstract9978.shtmlWan L M, Wu J, Lu J K, et al. Brittleness evaluation method of tight sandstone based on Adam-neural network: A case study of a block in Gaobei Slope, Nanpu Sag[J]. Bulletin of Geological Science and Technology, 2020, 39(2): 94-103(in Chinese with English abstract). http://dzkjqb.cug.edu.cn/CN/abstract/abstract9978.shtml [5] Na'imi S R, Shadizadeh S R, Riahi M A, et al. Estimation of reservoir porosity and water saturation based on seismic attributes using support vector regression approach[J]. Journal of Applied Geophysics, 2014, 107: 93-101. doi: 10.1016/j.jappgeo.2014.05.011 [6] 宋建国, 高强山, 李哲. 随机森林回归在地震储层预测中的应用[J]. 石油地球物理勘探, 2016, 51(6): 1202-1211. https://www.cnki.com.cn/Article/CJFDTOTAL-SYDQ201606021.htmSong J G, Gao Q S, Li Z. Application of random forests for regression to seismic reservoir prediction[J]. Oil Geophysical Prospecting, 2016, 51(6): 1202-1211(in Chinese with English abstract). https://www.cnki.com.cn/Article/CJFDTOTAL-SYDQ201606021.htm [7] Ao Y, Li H, Zhu L, et al. Identifying channel sand-body from multiple seismic attributes with an improved random forest algorithm[J]. Journal of Petroleum Science and Engineering, 2019, 173: 781-792. doi: 10.1016/j.petrol.2018.10.048 [8] 闫星宇, 顾汉明, 肖逸飞, 等. XGBoost算法在致密砂岩气储层测井解释中的应用[J]. 石油地球物理勘探, 2019, 54(2): 447-455. https://www.cnki.com.cn/Article/CJFDTOTAL-SYDQ201902024.htmYan X Y, Gu H M, Xiao Y F, et al. XGBoost algorithm applied in the interpretation of tight-sand gas reservoir on well logging data[J]. Oil Geophysical Prospecting, 2019, 54(2): 447-455(in Chinese with English abstract). https://www.cnki.com.cn/Article/CJFDTOTAL-SYDQ201902024.htm [9] 付超, 林年添, 张栋, 等. 多波地震深度学习的油气储层分布预测案例[J]. 地球物理学报, 2018, 61(1): 293-303. https://www.cnki.com.cn/Article/CJFDTOTAL-DQWX201801026.htmFu C, Lin N T, Zhang D, et al. Prediction of reservoirs using multi-component seismic data and the deep learning method[J]. Chinese Journal of Geophysics (in Chinese with English abstract), 2018, 61(1): 293-303(in Chinese with English abstract). https://www.cnki.com.cn/Article/CJFDTOTAL-DQWX201801026.htm [10] 马瑶, 赵江南. 机器学习方法在矿产资源定量预测应用研究进展[J]. 地质科技通报, 2021, 40(1): 132-141. http://dzkjqb.cug.edu.cn/CN/abstract/abstract10099.shtmlMa Y, Zhao J N. Advances in the application of machine learning methods in mineral prospectivity mapping[J]. Bulletin of Geological Science and Technology, 2021, 40(1): 132-141(in Chinese with English abstract). http://dzkjqb.cug.edu.cn/CN/abstract/abstract10099.shtml [11] 郭天颂, 张菊清, 韩煜, 等. 基于粒子群优化支持向量机的延长县滑坡易发性评价[J]. 地质科技情报, 2019, 38(3): 236-243. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKQ201903025.htmGuo T S, Zhang J Q, Han Y, et al. Evaluation of landslide susceptibility in Yanchang County based on particle swarm optimization-based support vector machine[J]. Geological Science and Technology Information, 2019, 38(3): 236-243(in Chinese with English abstract). https://www.cnki.com.cn/Article/CJFDTOTAL-DZKQ201903025.htm [12] 印兴耀, 周静毅. 地震属性优化方法综述[J]. 石油地球物理勘探, 2005, 40(4): 482-489. doi: 10.3321/j.issn:1000-7210.2005.04.027Yin X Y, Zhou J Y. Summary of optimum methods of seismic attributes[J]. Oil Geophysical Prospecting, 2005, 40(4): 482-489(in Chinese with English abstract). doi: 10.3321/j.issn:1000-7210.2005.04.027 [13] 王晓阳, 桂志先, 高刚, 等. K-L变换地震属性优化及其在储层预测中的应用[J]. 石油天然气学报, 2008, 30(3): 96-98. doi: 10.3969/j.issn.1000-9752.2008.03.022Wang X Y, Gui Z X, Gao G, et, al. Seismic attribute optimization and its application in reservoir prediction by using K-L transform[J]. Journal of Oil and Gas Technology, 2008, 30(3): 96-98. doi: 10.3969/j.issn.1000-9752.2008.03.022 [14] Singh Y. Lithofacies detection through simultaneous inversion and principal component attributes[J]. The Leading Edge, 2007, 26(12): 1568-1575. doi: 10.1190/1.2821944 [15] Roden R, Smith T, Sacrey D. Geologic pattern recognition from seismic attributes: Principal component analysis and self-organizing maps[J]. Interpretation, 2015, 3(4): 59-83. https://library.seg.org/journal/inteio [16] 倪艳. Isomap算法在地震属性参数降维中的应用[D]. 成都: 成都理工大学, 2007.Ni Y. Nonlinear dimensionality reduction of Isomap in the analysis of seismic attribute parameter data[D]. Chengdu: Chengdu University of Technology, 2007(in Chinese with English abstract). [17] Liu X F, Zheng X D, Xu G C, et al. Locally linear embedding-based seismic attribute extraction and applications[J]. Applied Geophysics, 2010, 7(4): 365-375. doi: 10.1007/s11770-010-0260-2 [18] 宋维琪, 刘江华, 王小马, 等. 预测油气的地震属性优化组合、灰关联分析技术[J]. 石油勘探与开发, 2002, 29(5): 34-36. doi: 10.3321/j.issn:1000-0747.2002.05.011Song W Q, Liu J H, Wang X M, et al. Applying optimum combination of sciesmic attribute and gray correlotion analysis technology to the prediction of oil and gas reservoirs[J]. Petroleum Exploration and Development, 2002, 29(5): 34-36(in Chinese with English abstract). doi: 10.3321/j.issn:1000-0747.2002.05.011 [19] 何琰, 彭文, 殷军. 利用地震属性预测渗透率[J]. 石油学报, 2001, 22(6): 34-36. https://www.cnki.com.cn/Article/CJFDTOTAL-SYXB200106006.htmHe Y, Peng W, Yin J. Permeability prediction by Seismic Attribute data[J]. Acta Petrolei Sinica, 2001, 22(6): 34-36(in Chinese with English abstract). https://www.cnki.com.cn/Article/CJFDTOTAL-SYXB200106006.htm [20] 谢东, 王永刚, 乐友喜, 等. 地震属性分析技术在子寅油田开发中的应用[J]. 石油物探, 2003, 42(1): 72-76. doi: 10.3969/j.issn.1000-1441.2003.01.016Xie D, Wang Y G, Yue Y X, et al. Application of seismic attribute analysis technology in the production of Ziyin Oilfield[J]. Geophysical Prospecting for Petroleum, 2003, 42(1): 72-76(in Chinese with English abstract). doi: 10.3969/j.issn.1000-1441.2003.01.016 [21] Dorrington K P, Link C A. Genetic-algorithm/neural-network approach to seismic attribute selection for well-log prediction[J]. Seg Technical Program Expanded Abstracts, 2004, 21(1): 212-221. http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=SEGEAB000021000001001654000001&idtype=cvips&prog=normal [22] 黄捍东, 赵迪, 张如伟, 等. 基于灰关联和遗传算法的地震属性优化方法[J]. 石油地球物理勘探, 2010, 45(3): 381-383. https://www.cnki.com.cn/Article/CJFDTOTAL-SYDQ201003010.htmHuang H D, Zhao D, Zhang R W, et al. Seismic attribute optimization method based on grey relational analysis and genetic algorithm[J]. Oil Geophysical Prospecting, 2010, 45(3): 381-383(in Chinese with English abstract). https://www.cnki.com.cn/Article/CJFDTOTAL-SYDQ201003010.htm [23] 武建华, 宋擒豹, 沈均毅, 等. 基于关联规则的特征选择算法[J]. 模式识别与人工智能, 2009, 22(2): 256-262. doi: 10.3969/j.issn.1003-6059.2009.02.013Wu J H, Song Q B, Shen J Y, et al. Feature selection algorithm based on association rules[J]. Pattern Recognition and Artificial Intelligence, 2009, 22(2): 256-262(in Chinese with English abstract). doi: 10.3969/j.issn.1003-6059.2009.02.013 [24] Chenxi H, Xin H, Yu F, et al. Sample imbalance disease classification model based on association rule feature selection[J]. Pattern Recognition Letters, 2020, 133. https://www.sciencedirect.com/science/article/pii/S0167865520300945 [25] 刘小珊, 罗文强, 李飞翱, 等. 基于关联规则的滑坡演化阶段判识指标[J]. 地质科技情报, 2014, 33(2): 160-164. doi: 10.3969/j.issn.1009-6248.2014.02.018Liu X S, Luo W Q, Li F A, et al. Identification index of landslide evolution stage based on association rule[J]. Geological Science and Technology Information, 2014, 33(2): 160-164(in Chinese with English abstract). doi: 10.3969/j.issn.1009-6248.2014.02.018 [26] 袁照威. 基于机器学习与多信息融合的致密砂岩储层井震解释方法研究[D]. 北京: 中国地质大学(北京), 2017.Yuan Z W. Interpretation methods of tight sandstone reservoir with seismic data and well logs based on machine learning method and multi-information fusion[D]. Beijing: China University of Geosciences (Beijing), 2017. [27] Agrawal R. Mining association rules between sets of items in large databases[C]//Anon. Acm Sigmod International Conference on Management of Data. [S. l. ]: [s. n. ], 1993. [28] Agrwal R. Fast algorithms for mining association rules in large databases[C]//Anon. Proceedings of the international conference on very large data bases. [S. l. ]: [s. n. ], 1994. [29] Han J, Pei J, Yin Y, et al. Mining frequent patterns without candidate generation: A frequent-pattern tree approach[J]. Data Mining & Knowledge Discovery, 2004, 8(1): 53-87. http://doi.ieeecomputersociety.org/resolve?ref_id=doi:10.1023/B:DAMI.0000005258.31418.83&rfr_id=trans/td/2008/07/ttd2008070865.htm [30] Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32. doi: 10.1023/A:1010933404324