A Study on Oil Well Production Prediction Based on Time Series Dynamic Analysis

doi:10.11885/j.issn.1674-5086.2020.05.22.04

Abstract

Abstract: The effect of oil well production prediction method currently used is not ideal. This study aims at dynamic prediction of oil production, using time series analysis combined with residual error correction. We build an ARIMA (Autoregressive integrated moving average) model with the ability of time series dynamic analysis, to predict the initial value and the real residual oil well production; the residual error was corrected by constructing the (SVM) Support Vector Machine time series prediction model to obtain the predicted value of oil well production combination. The LSTM (Long Short-Term Memory) model is compared with the above methods. The experimental results show that the average relative error rates of the combined prediction model and the LSTM model are 9.81% and 32.44% respectively. The conclusion is that the combined model prediction is more accurate, and provides an effective method for the dynamic prediction of oil well production, which can be used as a fast and real-time auxiliary basis for oil well production planning and has practical value.

Key words: oil well production prediction, ARIMA model, residual error, SVM, LSTM

CLC Number:

TE341

YANG Yang, CHENG Yuefei, QIAO Ying, LIU Jiong. A Study on Oil Well Production Prediction Based on Time Series Dynamic Analysis[J]. Journal of Southwest Petroleum University(Science & Technology Edition), 2020, 42(6): 82-88.

0
/ / Recommend

Add to citation manager EndNote|Ris|BibTeX

URL: http://journal15.magtechjournal.com/Jwk_xnzk/EN/10.11885/j.issn.1674-5086.2020.05.22.04

http://journal15.magtechjournal.com/Jwk_xnzk/EN/Y2020/V42/I6/82

References

[1] 缪长生,张晨阳,李振华,等. 塔里木油田产量预测方法探讨[J]. 中国管理信息化,2016,19(22):84-85. doi:10.3969/j.issn.1673-0194.2016.22.055 MIAO Changsheng, ZHANG Chenyang, LI Zhenhua, et al. A study on the production prediction method of Tarim Oilfield[J]. China Management Informationization, 2016, 19(22):84-85. doi:10.3969/j.issn.1673-0194.2016.22.055
[2] 陈元千. 对翁氏预测模型的推导及应用[J]. 天然气工业, 1996, 16(2):22-26. CHEN Yuanqian. Derivation and application of Weng's prediction model[J]. Natural Gas Industry, 1996, 16(2):22-26.
[3] 郭建军. 气田水平井开发技术政策研究——以克拉美丽气田为例[D]. 北京:中国地质大学(北京), 2012. GUO Jianjun. The horizontal well technology policy research in gas field:Taking the Kelameili Gas Field as an example[D]. Beijing:China University of Geosciences (Beijing), 2012.
[4] 赵艳艳,袁向春,康志江. 缝洞型碳酸盐岩油藏油井产量及压力变化模型[J]. 石油与天然气地质, 2010, 31(1):54-56, 62. ZHAO Yanyan, YUAN Xiangchun, KANG Zhijiang. Flow rate and pressure variation modeling for production wells in fractured-vuggy carbonates reservoirs[J]. Oil & Gas Geology, 2010, 31(1):54-56, 62.
[5] 徐波,薛中天,韩继勇. 用灰色系统模型预测油井产量[J]. 石油钻采工艺, 2001, 23(3):48-49. doi:10.13639/j.odpt.2001.03.019 XU Bo, XUE Zhongtian, HAN Jiyong. Oil production prediction by grey system model[J]. Oil Drilling & Production Technology, 2001, 23(3):48-49. doi:10.13639/j.odpt. 2001.03.019
[6] 鲁建中. 油田年产油量与含水率预测方法[J]. 大庆石油地质与开发, 2007, 26(4):62-65. LU Jianzhong. Yearly oil production and water cut forecasting method[J]. Petroleum Geology & Oilfield Development in Daqing, 2007, 26(4):62-65.
[7] 王春梅. 基于神经网络的数据挖掘算法研究[J]. 现代电子技术, 2017, 40(11):111-114. doi:10.16652/j.issn.1004-373x.2017.11.029 WANG Chunmei. Research on data mining algorithm based on neural network[J]. Modern Electronics Technology, 2017, 40(11):111-114. doi:10.16652/j.issn.1004-373x.2017.11.029
[8] 陶砾,杨朔,杨威. 深度学习的模型搭建及过拟合问题的研究[J]. 计算机时代, 2018(2):14-17, 21. doi:10.16644/j.cnki.cn33-1094/tp.2018.02.004 TAO Li, YANG Shuo, YANG Wei. Research on model building and over-fitting of deep learning[J]. Computer Era, 2018(2):14-17, 21. doi:10.16644/j.cnki.cn33-1094/tp.2018.02.004
[9] 李春生,谭民浠,张可佳. 基于改进型BP神经网络的油井产量预测研究[J]. 科学技术与工程,2011,11(31):7766-7769. LI Chunsheng, TAN Minxi, ZHANG Kejia. Research on single well production prediction based on improved BP neural networks[J]. Science Technology and Engineering, 2011, 11(31):7766-7769.
[10] 谷建伟,周梅,李志涛,等. 基于数据挖掘的长短期记忆网络模型油井产量预测方法[J]. 特种油气藏,2019,26(2):77-81. doi:10.3969/j.issn.1006-6535.2019.02.013 GU Jianwei, ZHOU Mei, LI Zhitao, et al. Oil well production forecast with long-shourt term memory network model based on data mining[J]. Special Oil and Gas Reservoirs, 2019, 26(2):77-81. doi:10.3969/j.issn.1006-6535.2019.02.013
[11] 田祥雨,刘立龙,陈军,等. 基于ARIMA与HoltWinters组合模型的电离层TEC预报[J]. 测绘科学技术学报, 2018, 35(1):44-48. TIAN Xiangyu, LIU Lilong, CHEN Jun, et al. Ionospheric TEC prediction based on the combination model of ARIMA and holt-winters[J]. Journal of Geomatics Science and Technology, 2018, 35(1):44-48.
[12] KIRSHNER H, MAGGIO S, UNSER M. A sampling theory approach for continuous ARMA identification[J]. IEEE Transactions on Signal Processing, 2011, 59(10):4620-4634. doi:10.1109/TSP.2011.2161983
[13] 刘军,柴洪洲,常宜峰,等. 改进的修正预测法预报电离层[J]. 测绘科学技术学报, 2011, 28(1):19-22. LIU Jun, CHAI Hongzhou, CHANG Yifeng, et al. Predicting ionospheric using improved correcting prediction method[J]. Journal of Geomatics Science and Technology, 2011, 28(1):19-22.
[14] FRAUSTO-SOLÍS J, CHI-CHIM M, SHEREMETOV L. Forecasting oil production time series with a populationbased simulated annealing method[J]. Arabian Journal for Science and Engineering, 2015, 40(4):1081-1096.
[15] VAPNIK V. Statistical learning theory[M]. New York:DBLP, 1998.
[16] 丁宏飞. 基于智能优化算法的支持向量机回归及其应用[D]. 成都:西南交通大学, 2011. DING Hongfei. Support vector machine regression based on intelligent optimization algorithm and its application[D]. Chengdu:Southwest Jiaotong University, 2011.
[17] 苏筱倩,安俊琳,张玉欣,等. 支持向量机回归在臭氧预报中的应用[J]. 环境科学, 2019, 40(4):1697-1704. doi:10.13227/j.hjkx.201809134 SU Xiaoqian, AN Junlin, ZHANG Yuxin, et al. Application of support vector machine regression in ozone forecasting[J]. Environmental Science, 2019, 40(4):1697-1704. doi:10.13227/j.hjkx.201809134
[18] 杨敏,丁剑,王炜. 基于ARIMA-SVM模型的快速公交停站时间组合预测方法[J]. 东南大学学报(自然科学版),2016,46(3):651-656. doi:10.3969/j.issn.1001-0505.2016.03.033 YANG Min, DING Jian, WANG Wei. Hybrid dwell time prediction method for bus rapid transit based on ARIMASVM model[J]. Journal of Southeast University (Natural Science Edition), 2016, 46(3):651-656. doi:10.3969/j.issn.1001-0505.2016.03.033
[19] 宋国君,国潇丹,杨啸,等. 沈阳市PM_2.5 浓度ARIMASVM组合预测研究[J]. 中国环境科学, 2018, 38(11):4031-4039. doi:10.19674/j.cnki.issn1000-6923.2018.0445 SONG Guojun, GUO Xiaodan, YANG Xiao, et al. ARIMA-SVM combination prediction of PM_2.5 concentration in Shenyang[J]. China Environmental Sciences, 2018, 38(11):4031-4039. doi:10.19674/j.cnki.issn1000-6923.2018.0445
[20] 郑荣,颜七笙. 基于ARIMA与SVM的国际铀资源价格预测[J]. 计算机工程与应用, 2016, 52(1):146-150. doi:10.3778/j.issn.1002-8331.1401-0234 ZHENG Rong, YAN Qisheng. Uranium resource price forecasting based on ARIMA and SVM model[J]. Computer Engineering and Application, 2016, 52(1):146-150. doi:10.3778/j.issn.1002-8331.1401-0234
[21] 漆莉,李革,李勤. ARIMA模型在流行性感冒预测中的应用[J]. 第三军医大学学报, 2007, 29(3):267-269. doi:10.3321/j.issn:1000-5404.2007.03.028 QI Li, LI Ge, LI Qin. Application of ARIMA model on predictive incidence of influenza[J]. Acta Academiae Medicinae Militrris Tertiae, 2007, 29(3):267-269. doi:10.3321/j.issn:1000-5404.2007.03.028
[22] DAVID P. jModelTest:Phylogenetic model averaging[J]. Molecular Biology and Evolution, 2008, 25(7):1253-1256. doi:10.1093/molbev/msn083
[23] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780. doi:10.1162/neco.1997.9.8.1735