基于支持向量机的川中杂卤石分类识别研究

doi:10.11867/j.issn.1001-8166.2016.10.1041

地球科学进展 ›› 2016, Vol. 31 ›› Issue (10): 1041 -1046. doi: 10.11867/j.issn.1001-8166.2016.10.1041

基于支持向量机的川中杂卤石分类识别研究

陈科贵 ¹(

), 吴刘磊 ^1,,A; ^*(

), 陈愿愿 ², 王刚 ³

1.西南石油大学地球科学与技术学院,四川成都 610500
2.川庆钻探工程有限公司地球物理勘探公司,四川成都 610213
3.中国石油新疆油田分公司勘探开发研究院,新疆克拉玛依 834000

收稿日期:2016-07-18 修回日期:2016-09-15 出版日期:2016-10-20
通讯作者: 吴刘磊 E-mail:chenkegui@21cn.com;1556883301@qq.com
基金资助:
国家自然科学基金项目“四川盆地油钾兼探的地球物理评价方法研究”(编号:41372103);国家重点基础研究发展计划项目“四川三叠纪古特提斯海盆钾分布、评价研究”(编号:2011CB403002)资助

Classification and Recognition of Polyhalite in Chuanzhong Based on Support Vector Machine

Kegui Chen ¹( ), Liulei Wu ^1, ^*( ), Yuanyuan Chen ², Gang Wang ³

1.School of Geoscience and Technology, Southwest Petroleum University, Chengdu 610500, China
2.Geophysical Exploration Company, Chuanqing Drilling Engineering Company Limited, Chengdu 610213, China
3.Research Institute of Exploration and Development, PetroChina Xinjiang Oilfield Company, Karamay 834000,China

Received:2016-07-18 Revised:2016-09-15 Online:2016-10-20 Published:2016-10-20
Contact: Liulei Wu E-mail:chenkegui@21cn.com;1556883301@qq.com
About author:
First author:Chen Kegui(1959-), male, Zigong City, Sichuan Province,Professor.Research areas include petroleum geology, technology of well logging reservoir evaluation and survey research potash in Sichuan.E-mail:chenkegui@21cn.com

*Corresponding author:Wu Liulei(1992-),male, Rugao City,Jiangsu Province, Master student. Research areas include logging interpretation.E-mail:1556883301@qq.com
Supported by:
Project supported by the National Natural Science Foundation of China “The study on geophysical evaluation method of oil and potash in Sichuan Basin” (No.41372103);The State Key Development Program for Basic Research of China “The study on potassium regularity and prediction of marine sediment on the China continental block”(No.2011CB403002)

全文 ( 2 ) 下载PDF ( 599 )

杂卤石是四川盆地主要的固态钾矿物,川中地区大多数杂卤石层不纯,通常伴随石膏层、硬石膏层、盐岩层发育,甚至同层沉积,常规测井解释方法只能粗略地识别杂卤石层。以支持向量机理论和测井解释为基础,测井数据作为输入,构建预测模型,对川中地区下中三叠统杂卤石样本做精细识别,将识别结果与录井资料验证对比,正确率达到90%以上。再以预测模型为基础,结合含杂卤石岩性在测井曲线上的响应情况,构建杂卤石层分类识别模型,识别杂卤石层、石膏质杂卤石层和杂卤石膏岩层,识别正确率达到91.78%,与常规测井解释方法相比具有明显优势。结果表明,将支持向量机运用到找钾矿中具有广阔的前景。

关键词: 杂卤石, 支持向量机, 测井响应, 分类识别

Polyhalite is mainly solid mineral potassium in Sichuan Basin, The most of polyhalite layer in Sichuan region is impurity, and usually accompanied by layers of gypsum, anhydrite, rock salt, and even deposited in the same layer. Conventional logging interpretation method can only roughly identificate polyhalite layers. Based on the theory of Support Vector Machine and logging interpretation methods,this paper creates prediction model with the input of logging curves, and discriminates the polyhalite reservoirs in the lower-middle Triassic strata. Compared with logging data, the accuracy rate of the discrimination results reaches 90%. According to the prediction model, identification model can be established with the curve features of polyhalite to discriminate pure polyhalite reservoirs, gypsiferous polyhalite reservoirs and polyhalite-gypsum reservoirs, the accuracy rate is 91.78%. The study demonstrates that Support Vector Machine is superior to the method of logging interpretation, and it has broad prospects in potash exploration.

Key words: Ployhalite, Support Vector Machine, Logging response, Classified discrimination.

中图分类号:

P619.21+1

表1 川中地区杂卤石分布层位及厚度

Table 1 Polyhalite isopch map of in the middle of Sichuan Basin

方位	构造	井号	累计厚度/m
川中	广安构造	广3井	7
		广100井	20
		广参2井	19
	龙女寺构造	女深2井	1
		女111井	24
		女112井	6.5
	华蓥山构造	华西1井	3
	华蓥山构造	华西2井	6.5
	罗渡溪构造	罗7井	3
	罗渡溪构造	罗11井	5

表1 川中地区杂卤石分布层位及厚度

Table 1 Polyhalite isopch map of in the middle of Sichuan Basin

方位	构造	井号	累计厚度/m
川中	广安构造	广3井	7
		广100井	20
		广参2井	19
	龙女寺构造	女深2井	1
		女111井	24
		女112井	6.5
	华蓥山构造	华西1井	3
	华蓥山构造	华西2井	6.5
	罗渡溪构造	罗7井	3
	罗渡溪构造	罗11井	5

图1 杂卤石测井响应特征图

Fig.1 Logging interpretaition of polyhalite

图2 支持向量机结构图

Fig.2 The structure of SVM

表2 采用不同核函数对比

Table 2 Comparison of different kernel functions

核函数	准确率/%
多项式内积函数	85.36
RBF核函数	90.31
Sigmoid核函数	52.02

表2 采用不同核函数对比

Table 2 Comparison of different kernel functions

核函数	准确率/%
多项式内积函数	85.36
RBF核函数	90.31
Sigmoid核函数	52.02

图3 模型整体流程

Fig.3 The overall process of model

表3 学习样本示例

Table 3 Examples of study samples

样品号	测井参数值
样品号	AC/(μs/ft)	DEN/(g/cm³)	GR/API	CNL/%	RLLD/(Ω·m)
1	85.39	2.75	158.36	10.3	9 987.3
2	74.32	2.72	145.62	12.78	9 635.6
3	86.78	2.78	160.23	13.01	9 856.3
4	90.62	2.76	132.36	13.11	9 632.2
5	95.32	2.65	152.78	10.44	9 846.61
6	79.25	2.9	146.37	17.28	9 986.7

表3 学习样本示例

Table 3 Examples of study samples

样品号	测井参数值
样品号	AC/(μs/ft)	DEN/(g/cm³)	GR/API	CNL/%	RLLD/(Ω·m)
1	85.39	2.75	158.36	10.3	9 987.3
2	74.32	2.72	145.62	12.78	9 635.6
3	86.78	2.78	160.23	13.01	9 856.3
4	90.62	2.76	132.36	13.11	9 632.2
5	95.32	2.65	152.78	10.44	9 846.61
6	79.25	2.9	146.37	17.28	9 986.7

图4 广3井杂卤石层识别结果对比图

Fig.4 Correlation of polyhalite discrimination in Guang 3 well

表4 杂卤石层识别正确率表

Table 4 Accuracy of polyhalite discrimination

井名	测试样本	预测结果	正确率/%
女112井	23	21	91.3
广100井	33	30	90.9
女110井	16	15	93.76
罗7井	10	9	90

表4 杂卤石层识别正确率表

Table 4 Accuracy of polyhalite discrimination

井名	测试样本	预测结果	正确率/%
女112井	23	21	91.3
广100井	33	30	90.9
女110井	16	15	93.76
罗7井	10	9	90

表5 分类识别杂卤石层结果对比表

Table 5 Corralation of classification results of polyhalite

井段/m	录井分析	测井解释	支持向量机判断
2 792.8~2 795	石膏质杂卤石	杂卤石	石膏质杂卤石
2 987~2 992.5	杂卤石夹薄层石膏	杂卤石	杂卤石夹薄层石膏
2 994~3 002	石膏质杂卤石	杂卤石	膏质杂卤石
3 007.5~3 009	杂卤石	杂卤石	杂卤石
3 015~3 016	杂卤石	杂卤石	杂卤石膏岩
3 027~3 030	杂卤石膏岩	杂卤石	杂卤石膏岩

表5 分类识别杂卤石层结果对比表

Table 5 Corralation of classification results of polyhalite

井段/m	录井分析	测井解释	支持向量机判断
2 792.8~2 795	石膏质杂卤石	杂卤石	石膏质杂卤石
2 987~2 992.5	杂卤石夹薄层石膏	杂卤石	杂卤石夹薄层石膏
2 994~3 002	石膏质杂卤石	杂卤石	膏质杂卤石
3 007.5~3 009	杂卤石	杂卤石	杂卤石
3 015~3 016	杂卤石	杂卤石	杂卤石膏岩
3 027~3 030	杂卤石膏岩	杂卤石	杂卤石膏岩

[1]	Huang Xuanzhen.Polyhalite—An ideal potash[C]∥Non-metallic Mineral Materials, Environmental Protection, Ecology, Health and Symposium.Chinese Ceramic Society Professional Committee of Non-metallic Minerals and National Building Materials Industry Committee on Science and Technology Education,2003:7.
	[黄宣镇. 杂卤石——一种理想的钾肥[C]∥非金属矿物材料与环保、生态、健康研讨会论文集.中国硅酸盐学会非金属矿专业委员会、国家建材工业科技教育委员会,2003:7.]
[2]	Chen Kegui, Li Chunmei, Li Li, et al.Geophysical logging criteria and discriminant model for the potassium-rich strata and their application to Sichuan Basin: A case study of Guang’an area of central Sichuan[J]. Acta Geoscientica Sinica,2013,31(5):623-630.
	[陈科贵,李春梅,李利,等. 四川盆地含钾地层的地球物理测井标志、判别模型与应用——以川中广安地区为例[J]. 地球学报,2013,34(5): 623-630.]
[3]	Chen Kegui, Li Chunmei, Zhang Jie,et al.Calculating the apparent content of K2O with logging data[J].Journal of Oil and Gas Technology,2012,34(3):96-100,167.
	[陈科贵,李春梅,张杰,等. 运用测井资料进行K2O视含量计算方法研究[J]. 石油天然气学报,2012,34(3):96-100,167.]
[4]	Pu Jie, Wu Xia, Li Hang.Reseach of the solid potash salts logging method in Sichuan Basin[J].Computer Applications of Petroleum,2013,(1):31-34.
	[蒲杰,吴霞,李航. 四川盆地固态钾盐测井方法研究[J]. 石油工业计算机应用,2013,(1):31-34.]
[5]	Chen Kegui,Li Li,Li Chunmei, et al.Exolution of the potash-rich areas in evaporation baisn during the epigenetic stage with continental block being active[J].Advances in Earth Science,2014,29(4):515-522.
	[陈科贵,李利,李春梅,等. 活动陆块背景下蒸发盆地后生阶段富钾区的演变[J]. 地球科学进展,2014,29(4):515-522.]
[6]	Song Yanjie, Zhang Jianfeng, Yan Weilin, et al.A new identification method for complex lithology with support vector machine[J]. Journal of Daqing Petroleum Institute,2007,31(5):18-20.
	[宋延杰,张剑风,闫伟林,等. 基于支持向量机的复杂岩性测井识别方法[J]. 大庆石油学院学报,2007,31(5):18-20.]
[7]	Liang Ya’na.Support Vector Machine in Logging Curve Hierarchical Application[D].Chengdu:Chengdu University of Technology,2012.
	[梁亚纳. 支持向量机在测井曲线分层中的应用[D].成都:成都理工大学,2012.]
[8]	Chen Kegui, Chen Xu, Zhang Jiahao.Combined methods of permeability logging evaluate in glutenite reservoirs—A case study of Badaowan Formation in Karamay Oilfield[J]. Advances in Earth Science,2015,30(7):773-779.
	[陈科贵,陈旭,张家浩. 复合渗透率测井评价方法在砂砾岩稠油油藏的应用——以克拉玛依油田某区八道湾组为例[J]. 地球科学进展,2015,30(7):773-779.]
[9]	Chen Kegui, Li Li, Wang Gang,et al.Analysis of logging response characteristics and potassium-forming conditions of early and middle Triassic Strata in Nanchong Basin[J].Mineral Deposits,2014,33(5):1 069-1 080.
	[陈科贵,李利,王刚,等. 四川盆地南充盐盆下、中三叠统测井响应特征及成钾条件分析[J]. 矿床地质,2014,33(5):1 069-1 080.]
[10]	Huang Xi.The Enrichment Regularity of Triassic Potassium-rich Brines of the Salt-bearing Sichuan Basin[D].Beijing:China University of Geosciences(Beijing),2013.
	[黄熙. 四川盆地三叠纪盐盆富钾卤水富集规律[D].北京:中国地质大学(北京),2013.]
[11]	Huang Xuanzhen.The first polyhalite deposit of China[J].Yunnan Geology,1996,15(1):52-61.
	[黄宣镇. 中国首例杂卤石矿床[J]. 云南地质,1996,15(1):52-61.]
[12]	Liu Dejun, Ran Qunying, Wang Bin.Application of support vector machine in logging analysis of Qijia sagin Daqing Oilfiel[J].Geophysical Prospecting for Petroleum,2007, 46(2):156-161.
	[刘得军,冉群英,王斌. 支持向量机在大庆齐家凹陷测井解释中的应用[J]. 石油物探,2007,46(2):156-161.]
[13]	Li Xinhu.Lithology identification methods contrast based on support vector machines at different well logging parameter[J]. Coalgeology & Exploration,2007,35(3):72-76,80.
	[李新虎. 基于不同测井曲线参数集的支持向量机岩性识别对比[J]. 煤田地质与勘探,2007,35(3):72-76,80.]
[14]	Zhang Ying, Pan Baozhi.The application of SVM and FMI to the lithologic identification of volcanic rocks[J]. Geophysical and Geochemical Exploration,2011,35(5):634-638,642.
	[张莹,潘保芝. 支持向量机与微电阻率成像测井识别火山岩岩性[J]. 物探与化探,2011,35(5):634-638,642.]
[15]	Yexin Liu. Support Vector Machine (SVM)-based geostatistical inversion: An integrated solution to honor well-log data and seismic attributes[C]∥CSPG CSEG CWLS Convention.Alberta,Canada,2009: 309-312.
[16]	Wang Guochang, Carr T R, Ju Yiwen, et al.Identifying organic-rich Marcellus Shale lithofacies by support vector machine classifier in the Appalachian Basin[J]. Computers & Geosciences, March,2014, 64: 52-60,doi:10.1016/j.caqeo.2013.12.002.
[17]	Hassan S M, Soliman O S, Mahmoud A S.Optimized data input for the support vector machine classifier using ASTER data. Case study: Wadi Atalla area, Eastern Desert, Egypt[J]. Carpathian Journal of Earth and Environmental Sciences,2015,10(1): 15-26.
[18]	Yu Daiguo, Sun Jianmeng,Wang Huanzeng,et al.A new method for logging lithology identification—SVM[J]. Petroleum Geology & Oilfield Development,2005,24(5):93-95,110.
	[于代国,孙建孟,王焕增,等. 测井识别岩性新方法——支持向量机方法[J]. 大庆石油地质与开发,2005,24(5):93-95,110.]
[19]	Shu Shiyu.SVM Predictive Optimizationalgorithm and Application Based Onensemble Learning[D].Shanghai:Donghua University,2015.
	[束诗雨. 基于集成学习的支持向量机预测优化算法及其应用[D].上海:东华大学,2015.]

[1]	杨福强,陈科贵,黄长兵,陈愿愿,李进,马小林. PSO-LIBSVM在钾盐矿层识别中的应用研究[J]. 地球科学进展, 2019, 34(7): 757-764.
[2]	陈科贵, 李进, 黄长兵, 陈愿愿, 王刚, 刘阳. BP神经网络在富钾卤水中的应用研究[J]. 地球科学进展, 2018, 33(6): 614-622.
[3]	张睿,马建文. 支持向量机在遥感数据分类中的应用新进展[J]. 地球科学进展, 2009, 24(5): 555-562.

阅读次数

全文

摘要