梁成钢,李菊花,陈依伟,等.基于朴素贝叶斯算法评估页岩油藏产能[J].深圳大学学报理工版,2023,40(1):66-73.[doi:10.3724/SP.J.1249.2023.01066] LIANG Chenggang,LI Juhua,CHEN Yiwei,et al.Productivity evaluation based on naive Bayesian algorithm in shale reservoir[J].Journal of Shenzhen University Science and Engineering,2023,40(1):66-73.[doi:10.3724/SP.J.1249.2023.01066]
Productivity evaluation based on naive Bayesian algorithm in shale reservoir
LIANG Chenggang1,LI Juhua2,CHEN Yiwei1,QIN Shunli2,ZHANG Jinfeng1,and HU Ke1
1.Jiqing Oilfield Operation Area of Xinjiang Oilfield Company, CNPC, Karamay 831700, Xinjiang Uygur Autonomous Region, P. R. China;2.School of Petroleum Engineering, Yangtze University, Wuhan 430100, Hubei Province, P. R. China
Shale oil, as an unconventional oil and gas resource with huge reserves, has become an important replacement resource and great significance to develop. Aiming at the problems of rapid decline oil production in the depletion development of shale oil reservoirs by "depleted horizontal well+volume fracturing", and the lower recovery factor predicted by the existing productivity evaluation methods in shale reservoir, we construct a productivity evaluation method for fractured horizontal wells based on naive Bayes algorithm. Taking the Jimsar shale reservoir in Xinjiang oilfield as the target reservoir, we establish the multi-classification Naive Bayes prior probability model and the conditional probability model containing five attributes of geological parameters and engineering parameters by taking the three-year cumulative production as the classification evaluation index. Then we obtain the posterior-probability model of the four types of production capacity based on Bayesian theory to achieve the capacity assessment of fractured horizontal wells in shale oil reservoirs. The results show that the proposed shale reservoir capacity classification method based on the Naive Bayes model is applicable. The accuracy of productivity prediction is 94% for Class Ⅰ wells, 71% for Class Ⅱ wells, 87% for Class Ⅲ wells, and 92% for Class Ⅳ wells, respectively. The productivity classification trend distribution map of fracturing horizontal wells in Jimsar shale reservoir is drawn, and the high-production potential area is mainly distributed in the southeast of the reservoir. The analysis of the constructed posterior probability model shows that the optimization of fracturing construction parameters can improve the probability of high well production rate. The study provides a guiding basis for the subsequent large-scale fracturing reconstruction of horizontal wells for shale oil development.
图3 训练集与测试集性质分析(a)储层厚度;(b)压裂液用量;(c)簇间距;(d)含油饱和度;(e)水力压裂强度Fig. 3 Analysis of the nature of the training set (blue) and test set (orange). (a) Reservoir thickness, (b) fracturing fluid dosage, (c) cluster spacing, (d) oil saturation, and (e) fracturing strength.
[1]张宸嘉,曹剑,王俞策,等. 准噶尔盆地吉木萨尔凹陷芦草沟组页岩油富集规律[J]. 石油学报,2022,43 (9):1253-1268. ZHANG Chenjia, CAO Jian, WANG Yuce, et al. Shale oil enrichment law of Lucaogou Formation in Jimsar Sag in Junggar Basin [J]. Acta Petrolei Sinica, 2022, 43(9): 1253-1268. (in Chinese)
[2]贾曙光,王军,王玲玲,等. 吉木萨尔凹陷芦草沟组页岩油甜点地震预测方法[J]. 新疆石油地质,2020, 41(5):535-541.JIA Shuguang, WANG Jun, WANG Lingling, et al. Seis‐mic prediction method of shale oil dessert in the Lucaogou Formation of Jimsar Sag [J]. Xinjiang Petroleum Geology, 2020, 41(5): 535-541. (in Chinese)
[3]王林生,叶义平,覃建华,等. 陆相页岩油储层微观孔喉结构表征与含油性分级评价——以准噶尔盆地吉木萨尔凹陷二叠系芦草沟组为例[J]. 石油与天然气地质,2022,43(1):149-160. WANG Linsheng, YE Yiping, QIN Jianhua, et al. Micro‐pore throat structure characterization and oiliness grading evaluation of continental shale oil reservoirs: a case study of the Permian Lucaogou Formation in Jimsar sag in Jung‐gar Basin [J]. Oil&Gas Geology, 2022,43(1):149-160. (in Chinese)
[4]赵振峰,李楷,赵鹏云,等.鄂尔多斯盆地页岩油体积压裂技术实践与发展建议[J].石油钻探技术, 2021,49(4):85-91. ZHAO Zhenfeng, LI Kai, ZHAO Pengyun, et al. Practice and development suggestions for volumetric fracturing technology for shale oil in the Ordos basin [J]. Petroleum Drilling Techniques, 2021, 49(4): 85-91. (in Chinese)[5]吴承美,许长福,陈依伟,等.吉木萨尔页岩油水平井开采实践[J].西南石油大学学报自然科学版, 2021,43(5):33-41. WU Chengmei, XU Changfu, CHEN Yiwei, et al. The horizontal well exploitation practice of Jimsar Shale Oil [J]. Journal of Southwest Petroleum University Science &Technology Edition, 2021, 43(5): 33-41. (in Chinese)
[6]刘文锋,张旭阳,张小栓,等. 致密油多段压裂水平井产能预测方法[J]. 新疆石油地质,2019,40(6):731-735. LIU Wenfeng, ZHANG Xuyang, ZHANG Xiaoshuan, et al. Production capacity prediction method of tight oil multi-stage fracturing horizontal well [J]. Xinjiang Petro‐leum Geology, 2019, 40(6): 731-735. (in Chinese)
[7]马先林,周德胜,蔡文斌,李宪文,何明舫. 基于可解释机器学习的水平井产能预测方法[J]. 西南石油大学学报自然科学版,2022,44(4):81-90. MA Xianlin, ZHOU Desheng, CAI Wenbin, et al. Horizon‐tal well capacity prediction method based on explainable machine learning [J]. Journal of Southwest Petroleum University Natural Science Edition, 2022, 44(4): 81-90. (in Chinese)
JI Lei, LI Juhua, XIAO Jialin. Application of random forest algorithm in the multistage fracturing stimulation of shale gas field [J]. Petroleum Geology&Oilfield Develop‐ment in Daqing, 2020, 39(6): 168-174. (in Chinese)
[9]周志华.机器学习[M].北京:清华大学出版社, 2016:12-13. ZHOU Zhihua. Machine learning [M]. Beijing: Tsinghua University Press, 2016: 12-13. (in Chinese)
[10]姜毅.基于贝叶斯网络模型的泥石流灾害经济损失风险评估[D].昆明:云南大学,2019. JIANG Yi. Risk assessment of economic loss in debris flow disasters based on Bayesian network [D]. Kunming:Yunnan University, 2019. (in Chinese)
[11]刘瑞.基于贝叶斯网络的洪水灾害风险评估与建模研究[D].上海:华东师范大学,2016. LIU Rui. A study on modeling of flood risk evaluation based on Bayesian networks [D]. Shanghai: East China Normal University, 2016. (in Chinese)
[12]杨立洪,李琼阳,李兴耀.基于信息值的相关属性约减——加权二分类朴素贝叶斯算法研究[J].统计与决策,2018,34(2):23-26. YANG Lihong, LI Qiongyang, LI Xingyao. Research on relevant attribute reduction: weighted binary classification naive Bayesian algorithm based on value of information [J]. Statistics and Decision, 2018, 34(2): 23-26. (in Chi‐nese)
[13]朱璇,袁彬,同元辉,等.致密低渗油藏压裂井网气驱深度学习预测模型[J].深圳大学学报理工版, 2022,39(5):559-566. ZHU Xuan, YUAN Bin, TONG Yuanhui, et al. Deep-learning-based proxy model for forecasting gas flooding performance of fractured well pattern in tight oil reservoirs [J]. Journal of Shenzhen University Science and Engi‐neering, 2022, 39(5): 559-566. (in Chinese)
[14]王武亮,江辉.基于多任务贝叶斯压缩感知的电能质量信号重构[J].深圳大学学报理工版,2021,38 (1):77-84. WANG Wuliang, JIANG Hui. Power quality signal reconstruction based on multitask Bayesian compressive sensing [J]. Journal of Shenzhen University Science and Engineering, 2021, 38(1): 77-84. (in Chinese)
[15]李楚进,付泽正.对朴素贝叶斯分类器的改进[J].统计与决策,2016(21):9-11. LI Chujin, FU Zezheng. Improvement of naive Bayesian classifier [J]. Statistics and Decision, 2016 (21): 9-11. (in Chinese)
[16]王娟,梁立星,武璠,等.数据挖掘技术在数字化油气藏研究与决策中的应用[J].钻采工艺,2018,41 (3):107-109. WANG Juan, LIANG Lixing, WU Fan, et al. Application of data mining technology in digital reservoir study and strategy planning [J]. Drilling and Production Technology, 2018, 41(3): 107-109. (in Chinese)
[19]吴璐.基于SVM-RFE特征选择的规则提取方法[J].微型电脑应用,2021,37(9):150-154. WU Lu. Rule extraction method based on SVM-RFE feature selection [J]. Microcomputer Applications, 2021, 37(9): 150-154. (in Chinese)
[20]于营,杨婷婷,杨博雄.混淆矩阵分类性能评价及Python实现[J].现代计算机,2021(20):70-73,79. YU Ying, YANG Tingting, YANG Boxiong. Confusion matrix classification performance evaluation and Python implementation [J]. Modern Computer, 2021 (20): 70-73, 79. (in Chinese)
[21]满涛,赵兴玲.CT在肺结核患者治疗中的指导价值及效能ROC曲线研究[J].当代医学,2021,27(28):139-140. MAN Tao, ZHAO Xingling. Study on the guiding value and efficacy ROC curve of CT in the treatment of patients with pulmonary tuberculosis [J]. Contemporary Medicine, 2021, 27(28): 139-140. (in Chinese)
[22]徐兵祥,白玉湖,陈岭,等.页岩油气产能预测新思路及方法流程[J].天然气技术与经济,2019,13 (5):36-42. XU Bingxiang, BAI Yuhu, CHEN Ling, et al. New ideas and workflow to predict shale-oil and shale-gas produc‐tivity [J]. Natural Gas Technology and Economy, 2019, 13 (5): 36-42. (in Chinese)