基于机器学习的石油多峰模型研究及应用

doi:10.11885/j.issn.1674-5086.2020.05.13.01

摘要/Abstract

摘要： 油田在实际开发过程中，受新区块投产、开发方案调整和"三采"措施等因素的影响，年产量数据会呈现多峰形态。针对经典的Hubbert、HCZ等模型不能直接拟合多峰数据序列的问题，开展了基于机器学习的油田产量多峰预测模型研究。基于Hubbert模型，对多峰数据序列进行分段最小二乘拟合，在拟合误差函数中引入控制分段个数的罚分项，采用动态规划算法，自动求得最优分段的多峰预测模型，该模型运用在实际的油田产量数据上，预测结果达到预期目的。提出了一种通过自动最优分段的线性回归学习来建立油田产量多峰预测模型的方法，在实际应用中具有建模简单、自适应性强的优点。

关键词: 石油产量预测, 机器学习, 动态规划, 多峰预测模型, Hubbert模型

Abstract: In the actual process of oilfield development, affected by the production of new blocks, the adjustment of development plans and the "t three mining" measures, the annual output data will show multi-peak form. As the classical Hubbert, HCZ and other models cannot directly fit the multi-peak data sequence, the multi-peak prediction model of oilfield production based on machine learning is studied. Based on the Hubbert model, the piecewise least squares fitting is performed for multi-peak data sequence, the penalty term controlling the number of segments is introduced into the fitting error function. Using dynamic programming algorithm, the multi-peak Hubbert prediction model for the optimal segment is automatically obtained. The model is applied to the actual oilfield production data, and the prediction results achieve the expected purpose. This paper presents a method to build a multi-peak prediction model of oilfield production through automatic optimal segmental linear regression learning. In practical application, it has the advantages of simple modeling and strong adaptability.

Key words: oil production forecasts, machine learning, dynamic programming, multiple peaks, Hubbert model

中图分类号:

TE311
F224.9

黄诚, 潘雯晋. 基于机器学习的石油多峰模型研究及应用[J]. 西南石油大学学报(自然科学版), 2020, 42(6): 75-81.

HUANG Cheng, PAN Wenjin. Research and Application of Oil Multi-peak Model Based on Machine Learning[J]. Journal of Southwest Petroleum University(Science & Technology Edition), 2020, 42(6): 75-81.

0
/ / 推荐

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://journal15.magtechjournal.com/Jwk_xnzk/CN/10.11885/j.issn.1674-5086.2020.05.13.01

http://journal15.magtechjournal.com/Jwk_xnzk/CN/Y2020/V42/I6/75

参考文献

[1] 邱中建,方辉. 对我国油气资源可持续发展的一些看法[J]. 石油学报, 2005, 26(2):1-5. doi:10.3321/j.issn:0253-2697.2005.02.001 QIU Zhongjian, FANG Hui. Some opinions about sustainable development of petroleum resource in China[J]. Acta Petrolei Sinica, 2005, 26(2):1-5. doi:10.3321/j.issn:0253-2697.2005.02.001
[2] 翟光明. 21世纪中国油气资源远景[J]. 新疆石油地质, 2002, 23(4):271-278. doi:10.3969/j.issn.1001-3873.2002.04.002 ZHAI Guangming. On prospective hydrocarbon resources of China in 21st century[J]. Xinjiang Petroleum Geology, 2002, 23(4):271-278. doi:10.3969/j.issn.1001-3873.2002.04.002
[3] 黎斌林. 全球石油峰值预测及中国应对策略研究[D]. 北京:中国地质大学(北京), 2014. LI Binlin. Global peak oil forecast and Chinese strategies research[D]. Beijing:China University of Geosciences (Beijing), 2014.
[4] HUBBERT M K. Energy from fossil fuels[J]. Science, 1949, 109(2823):103-109.
[5] 冯连勇,赵林,赵庆飞,等. 石油峰值理论及世界石油峰值预测[J]. 石油学报, 2006, 27(5):139-142. doi:10.3321/j.issn:0253-2697.2006.05.030 FENG Lianyong, ZHAO Lin, ZHAO Qingfei, et al. Models of peak oil theory and forecast of world oil peak[J]. Acta Petrolei Sinica, 2006, 27(5):139-142. doi:10.3321/j.issn:0253-2697.2006.05.030
[6] ADAM R B. Testing Hubbert[J]. Energy Policy, 2006, 35(5):3074-3088. doi:10.1016/j.enpol.2006.11.004
[7] 胡建国,张栋杰,陈元千. 油气田产量预测的模型研究[J]. 天然气工业, 1997, 17(5):31-34. HU Jianguo, ZHANG Dongjie, CHEN Yuanqian. A model investigation on forecasting the production of oil and gas field[J]. Natural Gas Industry, 1997, 17(5):31-34.
[8] 陈元千,郝明强. HCZ模型在多峰预测中的应用[J]. 石油学报, 2013, 34(4):747-752. doi:10.7623/syxb201304016 CHEN Yuanqian, HAO Mingqiang. Application of HCZ model to predicting multiple production peaks[J]. Acta Petrolei Sinica, 2013, 34(4):747-752. doi:10.7623/syxb 201304016
[9] 陈元千,郝明强. 多峰预测模型的建立与应用[J]. 新疆石油地质, 2013, 34(3):296-299. CHEN Yuanqian, HAO Mingqiang. Establishement and application of the multi-peak forecasting model[J]. Xinjiang Petroleum Geology, 2013, 34(3):296-299.
[10] 黎斌林. 基于广义翁氏模型与多循环Hubbert模型对全球石油峰值的预测[J]. 资源开发与市场, 2015, 31(4):427-430. doi:10.3969/j.issn.1005-8141.2015.04.011 LI Binlin. Global peak oil prediction by application of multi-cycle Hubbert model[J]. Resource Development & Market, 2015, 31(4):427-430. doi:10.3969/j.issn.1005-8141.2015.04.011
[11] 冯连勇,王建良,赵林. 预测天然气产量的多循环模型的构建及应用[J]. 天然气工业, 2010, 30(7):114-116. doi:10.3787/j.issn.1000-0976.2010.07.030 FENG Lianyong, WANG Jianliang, ZHAO Lin. Construction and application of a multi-cycle model in the prediction of natural gas production[J]. Natural Gas Industry, 2010, 30(7):114-116. doi:10.3787/j.issn.1000-0976.2010.07.030
[12] 陆家亮,赵素平,孙玉平,等. 中国天然气产量峰值研究及建议[J]. 天然气工业, 2018, 38(1):1-9. doi:10.3787/j.issn.1000-0976.2018.01.001 LU Jialiang, ZHAO Suping, SUN Yuping, et al. Natural gas production peaks in China:Research and strategic pro-posals[J]. Natural Gas Industry, 2018, 38(1):1-9. doi:10.3787/j.issn.1000-0976.2018.01.001
[13] 贾爱林. 中国天然气开发技术进展及展望[J]. 天然气工业,2018,38(4):77-86. doi:10.3787/j.issn.1000-0976.2018.04.009 JIA Ailin. Progress and prospects of natural gas development technoloties in China[J]. Natural Gas Industry, 2018, 38(4):77-86. doi:10.3787/j.issn.1000-0976.2018.04.009
[14] BREIMAN L. Bagging predictors[J]. Machine Learning, 1996, 24(2):123-140. doi:10.1007/BF00058655
[15] CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3):273-297. doi:10.1007/BF00994018
[16] YU Jie. Book review:Machine learning for audio, image and video analysis[J]. Journal of Electronic Imaging, 2009, 18(2):029901.
[17] 欧阳永林,田文广,孙斌,等. 中国煤系气成藏特征及勘探对策[J]. 天然气工业, 2018, 38(3):15-23. doi:10.3787/j.issn.1000-0976.2018.03.002 OUYANG Yonglin, TIAN Wenguang, SUN Bin, et al. Characteristics of coal measure gas accumulation and such gas exploration strategies in China[J]. Natural Gas Industry, 2018, 38(3):15-23. doi:10.3787/j.issn.1000-0976.2018.03.002
[18] AL-FATTAH S M, STARTZMAN R A. Analysis of worldwide natural gas production[C]. SPE 57463-MS, 1999. doi:10.2118/57463-MS
[19] AI-JARRI A S, SAUDI A, STARTZMAN R A. Analysis of world crude oil production trends[C]. SPE 37962-MS, 1997. doi:10.2118/37962-MS
[20] BAMNETT V, LEWIS T. Outliers in statistical data[M]. 3rd Edition. New York:John Wiley & Sons, 1994.
[21] 彭珊. 线性回归模型中关于异常点的若干问题的分析[D]. 哈尔滨:东北林业大学, 2014. PENG Shan. Analysis of linear regression model on the outlier problems[D]. Harbin:Northeast Forestry University, 2014.
[22] 杨延娇,门维江. 基于异常点挖掘的聚类算法比较研究[J]. 甘肃联合大学学报(自然科学版), 2008, 22(1):87-90. doi:10.3969/j.issn.1672-691X.2008.01.026 YANG Yanjiao, MEN Weijiang. Clustering algorithms of relatively research based on outlier data mining[J]. Journal of Gansu Lianhe University (Natural Sciences), 2008, 22(1):87-90. doi:10.3969/j.issn.1672-691X.2008.01.026
[23] 王玉梅. 线性回归模型中的异常点分析[D]. 武汉:华中科技大学, 2006. WANG Yumei. Outlier analysis of linear regression model[D]. Wuhan:Huazhong University of Science and Technology, 2006.
[24] 贾小勇,徐传胜,白欣. 最小二乘法的创立及其思想方法[J]. 西北大学学报(自然科学版), 2006, 36(3):507-511. doi:10.3321/j.issn:1000-274X.2006.03.040 JIA Xiaoyong, XU Chuansheng, BAI Xin. The invention and way of thinking on least squares[J]. Journal of Northwest University (Natural Science Edition), 2006, 36(3):507-511. doi:10.3321/j.issn:1000-274X.2006.03.040
[25] 邹乐强. 最小二乘法原理及其简单应用[J]. 科技信息, 2010, 2(23):282-283. doi:10.3969/j.issn.1001-9960.2010.23.875
[26] 黄伏生,赵永胜,刘青年. 油田动态预测的一种新模型[J]. 大庆石油地质与开发, 1987, 6(4):55-62. HUANG Fusheng, ZHAO Yongsheng, LIU Qingnian. A new model for oilfield performance prediction[J]. Petroleum Geology & Oilfield Development in Daqing, 1987, 6(4):55-62.