基于多元线性回归和BP神经网络的单井能力预测
|
摘要
为提高单井能力预测的精度和可靠性,提出利用地震属性数据,结合多元线性回归方法和BP神经网络方法进行预测。首先提取了研究区目的层的地震属性,然后利用多元线性回归方法和BP神经网络方法建立了单井能力与地质、地震属性之间的函数关系,得出了半定量-定量化的单井产量设计模型,并且验证了模型的预测结果。结果显示:单井能力预测精度总体在80%以上,其中BP神经网络模型预测精度更高,吻合度更好,证明了利用多种地震属性联合预测单井产能是一种卓有成效的方法。
In order to improve the accuracy and reliability of the prediction of single well capability,raised a idea of prediction using seismic attribute data,combined with multiple linear regression method and back propagation neural network method. Firstly,the seismic attributes of the target layer in the study area are extracted,and then the functional relationship between the single well capability and geological and seismic attributes is established by using the multiple linear regression method and the BP neural network method. The semi-quantification-quantitative single well production design model is obtained,and the prediction results of the model are verified. The results show that the prediction accuracy of single well capability is more than 80%,in which the BP neural network model has higher prediction accuracy and better coincidence. It is proved that using multiple seismic attributes to predict single well productivity is an effective method.
In order to improve the accuracy and reliability of the prediction of single well capability,raised a idea of prediction using seismic attribute data,combined with multiple linear regression method and back propagation neural network method. Firstly,the seismic attributes of the target layer in the study area are extracted,and then the functional relationship between the single well capability and geological and seismic attributes is established by using the multiple linear regression method and the BP neural network method. The semi-quantification-quantitative single well production design model is obtained,and the prediction results of the model are verified. The results show that the prediction accuracy of single well capability is more than 80%,in which the BP neural network model has higher prediction accuracy and better coincidence. It is proved that using multiple seismic attributes to predict single well productivity is an effective method.
引文
1康玉柱．中国海相油气田勘探实例之四塔里木盆地塔河油田的发现与勘探[J]．海相油气地质，2005,10(4):31-38Kang Yuzhu. Cases of discovery and exploration of marine fields in China(Part 4):Tahe oilfield in Tarim Basin[J]. Marine Origin Petroleum Geology,2005,10(4):31-38
2 康玉柱．塔里木盆地古生代海相碳酸盐岩储集岩特征[J]．石油实验地质，2007,29(3):217-223Kang Yuzhu. Reservoir rock characteristics of paleozoic marine facies carbonate rock in the Tarim basin[J]. Petroleum Geology&Experiment,2007,29(3):217-223
3 Tian F,Jin Q,Lu X,et al. Multi-layered ordovician paleokarst reservoir detection and spatial delineation:a case study in the Tahe Oilfield,Tarim Basin,Western China[J]. Marine and Petroleum Geology,2016,69:53-73
4 窦之林．碳酸盐岩缝洞型油藏描述与储量计算[J]．石油实验地质，2014,36(1):9-15Dou Zhilin. Description and reserves calculation of fractured-vuggy carbonate reservoirs[J]. Petroleum Geology&Experiment,2014,36(1):9-15
5 陈广坡，潘建国，陶云光，等．碳酸盐岩岩溶型储层综合预测评价技术的应用及效果分析[J]．石油物探，2005,45(1):33-36Chen Guangpo,Pan Jianguo,Tao Yunguang,et al. The application of multiple information for predictive of carbonate karst reservoir and effect analysis[J]. Geophysical Prospect for Petroleum,2005,45(1):33-36
6 胡建国，陈元千，张盛宗，等．预测油气田产量的新模型[J]．石油学报，1995,16(1):79-87Huo Jianguo,Chen Yuanqian,Zhang Zongsheng,et al. A new model to predict production rate of oil and gas fields[J]. Acta Petroleum Sinica,1995,16(1):79-87
7 朱亚东，高珉，熊铁，等．预测油气田产量和可采储量的F模型[J]．石油勘探与开发，2000,27(1):54-56Zhu Yadong,Gao Min,Xiong Tie,et al. The F-model for predicting production rate and reserves of oil and gas field[J]. Petroleum Exploration and Development,2000,27(1):54-56
8 周总瑛，张抗，周庆凡，等．油气储量、产量及需求量的常用预测方法[J]．新疆石油地质，2001,22(5):444-447Zhou Zongying,Zhang Kang,Zhou Qingfan,et al. An introduction of common prediction methods for oil and gas reserves[J]. Xinjiang Petroleum Geology,2001,22(5):444-447
9 王俊魁．油气藏产量递减规律的研究与可采储量的预测[J]．新疆石油地质，2007,28(2):189-193Wang Junkui. Production decline rules and recoverable reserve prediction[J]. Xinjiang Petroleum Geology,2007,28(2):189-193
10 李新华，鲁新便，何成江，等．塔河油田碳酸盐岩缝洞储集体地震反射特征量化识别研究与开发应用[J]．石油天然气学报，2012,34(8):78-83,165-166Li Xinhua,Lu Xinbian,He Chengjiang,et al. The research and development application of quantitative recognition of seismic reflection characteristics for fracture-cavity reservoirs in Tahe oilfield[J].Journal of Oil and Gas technology,2012,34(8):78-83,165-166
11 陈智远，孙德彬，罗新生，等．塔中中西部地区鹰山组“串珠状”反射地震地质特征及主控因素分析[J]．石油地质与工程，2014,28(3):41-43,87Chen Zhiyuan,Sun Debin,Luo Xinsheng,et al. Seismic geological characteristics and main controlling factors of“beaded”reflection of Yingshan formation in central and western Tarim Basin[J]. Petroleum Geology and Engineering,2014,28(3):41-43,87
12 郭军参，王振宇，屈海洲，等．塔中岩溶储层地震反射地质特征及其成因机理[J]．重庆科技学院学报(自然科学版)，2012,14(2):24-27Guo Juncan,Wang Zhenyu,Qu Haizhou,et al. Geological characteristics and formation mechanism of karst reservoir seismic reflection in Tazhong area[J]. Journal of Chongqing University of Science and Technology(Natural Sciences Edition),2012,14(2):24-27
13 项云飞，康志宏，郝伟俊，等．基于线性回归与神经网络的储层参数预测复合方法[J]．科学技术与工程，2017,17(31):46-52Xiang Yunfei,Kang Zhihong,Hao Weijun,et al. A composite method of reservoir parameter prediction based on linear regression and neural network[J]. Science Technology and Engineering,2017,17(31):46-52
14 李菊花，刘德华，王宇，等．多元线性回归预测油田的产油量[J]．石油地质与工程，2007,21(5):59-61Li Juhua,Liu Dehua,Wang Yu,et al. Crude output forecast with multiple linear regression equation[J]. Petroleum Geology and Engineering,2007,21(5):59-61
15 王美石，陈祥光．多元线性回归方法在油田产量预测中的应用[J]．油气田地面工程，2004,23(11):25-26Wang Meishi,Chen Xiangguang. Application of multivariate linear regression to output prediction of oilfield[J]. Oil-Gasfield Surface Engineering,2004,23(11):25-26
16 刘欣佳，温庆志，张遂安，等．裂缝性油藏压裂井产能敏感因素评价方法研究[J]．特种油气藏，2017,24(5):170-174Liu Xinjia,Wen Qingzhi,Zhang Suian,et al. Techniques for assessment of productivity sensitive factors in wells with fracturing operations in fractured reservoirs[J]. Special Oil and Gas Reservoirs,2017,24(5):170-174
17 朱红，孔德群，钱旭，等．基于ATD-BP神经网络的页岩气产量预测方法[J]．科学技术与工程，2017,17(31):134-138Zhu Hong,Kong Dequn,Qian Xu,et al. Shale gas production prediction method based on adaptive threshold denoising BP neural network[J]. Science Technology and Engineering,2017,17(31):134-138
18 戚德虎，康继昌．BP神经网络的设计[J]．计算机工程与设计，1998,19(2):48-50Qi Dehu,Kang Jichang. On design of the BP neural network[J].Computer Engineering and Design,1998,19(2):48-50
2 康玉柱．塔里木盆地古生代海相碳酸盐岩储集岩特征[J]．石油实验地质，2007,29(3):217-223Kang Yuzhu. Reservoir rock characteristics of paleozoic marine facies carbonate rock in the Tarim basin[J]. Petroleum Geology&Experiment,2007,29(3):217-223
3 Tian F,Jin Q,Lu X,et al. Multi-layered ordovician paleokarst reservoir detection and spatial delineation:a case study in the Tahe Oilfield,Tarim Basin,Western China[J]. Marine and Petroleum Geology,2016,69:53-73
4 窦之林．碳酸盐岩缝洞型油藏描述与储量计算[J]．石油实验地质，2014,36(1):9-15Dou Zhilin. Description and reserves calculation of fractured-vuggy carbonate reservoirs[J]. Petroleum Geology&Experiment,2014,36(1):9-15
5 陈广坡，潘建国，陶云光，等．碳酸盐岩岩溶型储层综合预测评价技术的应用及效果分析[J]．石油物探，2005,45(1):33-36Chen Guangpo,Pan Jianguo,Tao Yunguang,et al. The application of multiple information for predictive of carbonate karst reservoir and effect analysis[J]. Geophysical Prospect for Petroleum,2005,45(1):33-36
6 胡建国，陈元千，张盛宗，等．预测油气田产量的新模型[J]．石油学报，1995,16(1):79-87Huo Jianguo,Chen Yuanqian,Zhang Zongsheng,et al. A new model to predict production rate of oil and gas fields[J]. Acta Petroleum Sinica,1995,16(1):79-87
7 朱亚东，高珉，熊铁，等．预测油气田产量和可采储量的F模型[J]．石油勘探与开发，2000,27(1):54-56Zhu Yadong,Gao Min,Xiong Tie,et al. The F-model for predicting production rate and reserves of oil and gas field[J]. Petroleum Exploration and Development,2000,27(1):54-56
8 周总瑛，张抗，周庆凡，等．油气储量、产量及需求量的常用预测方法[J]．新疆石油地质，2001,22(5):444-447Zhou Zongying,Zhang Kang,Zhou Qingfan,et al. An introduction of common prediction methods for oil and gas reserves[J]. Xinjiang Petroleum Geology,2001,22(5):444-447
9 王俊魁．油气藏产量递减规律的研究与可采储量的预测[J]．新疆石油地质，2007,28(2):189-193Wang Junkui. Production decline rules and recoverable reserve prediction[J]. Xinjiang Petroleum Geology,2007,28(2):189-193
10 李新华，鲁新便，何成江，等．塔河油田碳酸盐岩缝洞储集体地震反射特征量化识别研究与开发应用[J]．石油天然气学报，2012,34(8):78-83,165-166Li Xinhua,Lu Xinbian,He Chengjiang,et al. The research and development application of quantitative recognition of seismic reflection characteristics for fracture-cavity reservoirs in Tahe oilfield[J].Journal of Oil and Gas technology,2012,34(8):78-83,165-166
11 陈智远，孙德彬，罗新生，等．塔中中西部地区鹰山组“串珠状”反射地震地质特征及主控因素分析[J]．石油地质与工程，2014,28(3):41-43,87Chen Zhiyuan,Sun Debin,Luo Xinsheng,et al. Seismic geological characteristics and main controlling factors of“beaded”reflection of Yingshan formation in central and western Tarim Basin[J]. Petroleum Geology and Engineering,2014,28(3):41-43,87
12 郭军参，王振宇，屈海洲，等．塔中岩溶储层地震反射地质特征及其成因机理[J]．重庆科技学院学报(自然科学版)，2012,14(2):24-27Guo Juncan,Wang Zhenyu,Qu Haizhou,et al. Geological characteristics and formation mechanism of karst reservoir seismic reflection in Tazhong area[J]. Journal of Chongqing University of Science and Technology(Natural Sciences Edition),2012,14(2):24-27
13 项云飞，康志宏，郝伟俊，等．基于线性回归与神经网络的储层参数预测复合方法[J]．科学技术与工程，2017,17(31):46-52Xiang Yunfei,Kang Zhihong,Hao Weijun,et al. A composite method of reservoir parameter prediction based on linear regression and neural network[J]. Science Technology and Engineering,2017,17(31):46-52
14 李菊花，刘德华，王宇，等．多元线性回归预测油田的产油量[J]．石油地质与工程，2007,21(5):59-61Li Juhua,Liu Dehua,Wang Yu,et al. Crude output forecast with multiple linear regression equation[J]. Petroleum Geology and Engineering,2007,21(5):59-61
15 王美石，陈祥光．多元线性回归方法在油田产量预测中的应用[J]．油气田地面工程，2004,23(11):25-26Wang Meishi,Chen Xiangguang. Application of multivariate linear regression to output prediction of oilfield[J]. Oil-Gasfield Surface Engineering,2004,23(11):25-26
16 刘欣佳，温庆志，张遂安，等．裂缝性油藏压裂井产能敏感因素评价方法研究[J]．特种油气藏，2017,24(5):170-174Liu Xinjia,Wen Qingzhi,Zhang Suian,et al. Techniques for assessment of productivity sensitive factors in wells with fracturing operations in fractured reservoirs[J]. Special Oil and Gas Reservoirs,2017,24(5):170-174
17 朱红，孔德群，钱旭，等．基于ATD-BP神经网络的页岩气产量预测方法[J]．科学技术与工程，2017,17(31):134-138Zhu Hong,Kong Dequn,Qian Xu,et al. Shale gas production prediction method based on adaptive threshold denoising BP neural network[J]. Science Technology and Engineering,2017,17(31):134-138
18 戚德虎，康继昌．BP神经网络的设计[J]．计算机工程与设计，1998,19(2):48-50Qi Dehu,Kang Jichang. On design of the BP neural network[J].Computer Engineering and Design,1998,19(2):48-50