建立了一种基于人工智能的致密砂岩气储集层重复压裂井筛选方法,并进行了算例分析。该方法以模糊逻辑为基础,通过语言模糊性的数学表示来处理语言的不精确性和主观性,是1个基于模糊集理论、模糊规则和模糊推理的计算系统。用5个指数分别表征与重复压裂井选择问题相关的水力裂缝质量、储集层特征、初始条件、操作参数、产量,每个指数又包含3个相关参数。将每个指数/参数的值划分为低、中、高3个类别,针对每个指数/参数的每个类别定义梯形隶属函数,并定义所有相关规则。先将某个指数的相关参数输入到基于规则的模糊推理系统中,输出该指数的值,再另建1个模糊推理系统,将储集层指数、操作指数、初始条件指数和产量指数作为输入参数,重复压裂潜力指数作为输出参数,从而筛选重复压裂井。利用已发表文献中的数据验证了该方法的有效性。图11表3参25
ARTUN Emre
,
KULGA Burak
. 基于模糊推理的致密砂岩气储集层重复压裂井选择方法[J]. 石油勘探与开发, 2020
, 47(2)
: 383
-389
.
DOI: 10.11698/PED.2020.02.17
An artificial-intelligence based decision-making protocol is developed for tight gas sands to identify re-fracturing wells and used in case studies. The methodology is based on fuzzy logic to deal with imprecision and subjectivity through mathematical representations of linguistic vagueness, and is a computing system based on the concepts of fuzzy set theory, fuzzy if-then rules, and fuzzy reasoning. Five indexes are used to characterize hydraulic fracture quality, reservoir characteristics, operational parameters, initial conditions, and production related to the selection of re-fracturing well, and each index includes 3 related parameters. The value of each index/parameter is grouped into low, medium and high 3 categories. For each category, a trapezoidal membership function all related rules are defined. The related parameters of an index are input into the rule-based fuzzy-inference system to output value of the index. Another fuzzy-inference system is built with the reservoir index, operational index, initial condition index and production index as input parameters and re-fracturing potential index as output parameter to screen out re-fracturing wells. This approach was successfully validated using published data.
[1] DONG Z, HOLDITCH S, MCVAY D, et al.Global unconventional gas resource assessment[R]. SPE 148365-PA, 2012.
[2] REEVES S, HILL D, HOPKINS C, et al.Restimulation technology for tight gas sand wells[R]. SPE 56482-MS, 1999.
[3] REEVES S, HILL D, TINER R, et al.Restimulation of tight gas sand wells in the Rocky Mountain Region[R]. SPE 55627-MS, 1999.
[4] CROWELL R F, JENNINGS A R.A diagnostic technique for restimulation candidate selection[R]. SPE 7556-MS, 1978.
[5] ROUSSEL N P, SHARMA M M.Selecting candidate wells for refracturing using production data[R]. SPE 146103-PA, 2013.
[6] UDEGBE E, MORGAN E, SRINIVASAN S.Big-data analytics for production-data classification using feature detection: Application to restimulation-candidate selection[R]. SPE 187328-PA, 2019.
[7] MOORE L P, RAMAKRISHNAN H.Restimulation candidate selection methodologies and treatment optimization[R]. SPE 102681-MS, 2006.
[8] SHELLEY R F.Artificial neural networks identify restimulation candidates in the Red Oak Field[R]. SPE 52190-MS, 1999.
[9] MOHAGHEGH S, REEVES S, HILL D.Development of an intelligent systems approach for restimulation candidate selection[R]. SPE 59767-MS, 2000.
[10] OBERWINKLER C, ECONOMIDES M J.The definitive identification of candidate wells for refracturing[R]. SPE 84211-MS, 2003.
[11] KULGA B, ARTUN E, ERTEKIN T.Characterization of tight-gas sand reservoirs from horizontal-well performance data using an inverse neural network[J]. Journal of Natural Gas Science and Engineering, 2018, 59: 35-46.
[12] VINCENT M C.Refracs: Why do they work, and why do they fail in 100 published field studies?[R]. SPE 134330, 2010.
[13] ZADEH L.Fuzzy sets[J]. Information and Control, 1965, 8(3): 338-353.
[14] KULGA B, ARTUN E, ERTEKIN T.Development of a data-driven forecasting tool for hydraulically fractured, horizontal wells in tight-gas sands[J]. Computers & Geosciences, 2017, 103: 99-110.
[15] MOHAGHEGH S.Mapping the natural fracture network in Utica Shale using artificial intelligence (AI)[R]. URTEC 2669739-MS, 2017.
[16] MOHAGHEGH S.Shale analytics: Data-driven analytics in unconventional resources[M]. Berlin: Springer, 2016.
[17] MAMDANI E H, ASSILIAN S.An experiment in linguistic synthesis with a fuzzy logic controller[J]. International Journal of Man-Machine Studies, 1975, 7(1): 1-13.
[18] BATURONE I, BARRIGA A, JIMENEZ-FERNANDEZ C, et al.Microelectronic design of fuzzy logic-based systems[M]. 1st ed. Boca Raton, Florida: CRC Press, 2000.
[19] ROSS T J.Fuzzy logic with engineering applications[M]. 2nd ed. Hoboken, New Jersey: John Wiley and Sons, Ltd., 2004.
[20] WAGNER C, MILLER S, GARIBALDI J M.A fuzzy toolbox for the R programming language[C]//IEEE International Conference on Fuzzy Systems Proceedings. Taipei: IEEE, 2011: 27-30.
[21] ROTHKOPF B, CHRISTIANSEN D, GODWIN H, et al.Texas panhandle granite wash formation: Horizontal development solutions[R]. SPE 146651-MS, 2011.
[22] CASERO A, ADEFASHE H, PHELAN K.Open hole multi-stage completion system in unconventional plays: Efficiency, effectiveness and economic[R]. SPE 164009-MS, 2013.
[23] CASTRO L, BASS C, PIROGOV A, et al.A comparison of proppant placement, well performance, and estimated ultimate recovery between horizontal wells completed with multi-cluster plug & perf and hydraulically activated frac ports in a tight gas reservoir[R]. SPE 163820-MS, 2013.
[24] WEI Y, XU J.Unconventional oil and gas resources handbook[M]. Amsterdam: Elsevier, 2016: 449-473.
[25] ELY J, BROWN T, REED S.Optimization of hydraulic fracture treatment in the Williams Fork Formation of the Mesaverde Group[R]. SPE 29551-MS, 1995.
