Machine Learning of Spatially Varying Decline Curves for the Duvernay Formation
- Alexander Bakay (Stanford University) | Jef Caers (Stanford University) | Tapan Mukerji (Stanford University) | Patrick Miller (Repsol) | Cheryl Cartier (Repsol) | Arnulfo Briceno (Repsol)
- Document ID
- Society of Petroleum Engineers
- SPE Annual Technical Conference and Exhibition, 30 September - 2 October, Calgary, Alberta, Canada
- Publication Date
- Document Type
- Conference Paper
- 2019. Society of Petroleum Engineers
- Machine learning, Shale reservoirs, Duvernay, Decline curves, Geostatistics
- 14 in the last 30 days
- 281 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 9.50|
|SPE Non-Member Price:||USD 28.00|
The focus of this paper is on Duvernay shale formation in Alberta, Canada. The objective is to provide, based on existing data of production, completion and geological parameters, an automated machine- learning approach to determine the spatial variation in decline type curves for gas production. This model will enable the prediction and uncertainty quantification of production profiles for new target wells or areas in the basin.
The project is based on publicly available monthly production data from most of the producing wells of the Duvernay formation. We use k-means to cluster 273 wells, using geological parameters (thickness, porosity, etc.), completion parameters (horizontal section length, proppant volume, etc.), spatial location, fluid window, and production curves. Based on the clustering results, a machine learning classification is used to draw distinct geographic regions, within which the combination of geological, completion, and production factors is fairly similar. A support vector machine approach is used to create maps of clusters and quantify its uncertainty.
In addition, functional classification and regression trees (CART) is used to indicate the most important/sensitive factors that should be used for clustering.
The results show that the unsupervised method, k-means, performs equally as well as the supervised CART method. The methodology is flexible and allows for quick changes in the variables used in clustering; the transfer to another dataset or basin is straightforward.
|File Size||1 MB||Number of Pages||13|
Esmaili, S. and Mohaghegh, S. D. 2016. Full field reservoir modeling of shale assets using advanced data-driven analytics. Geoscience Frontiers 7 (1): 11-20. 10.1016/j.gsf.2014.12.006.
Ghurav, A. 2017. Horizontal Shale Well EUR Determination Integrating Geology, Machine Learning, Pattern Recognition and MultiVariate Statistics Focused onthe Permian Basin. Presented at the SPE Liquids-Rich Basins Conference - North America, Midland, Texas, 13-14 September. SPE-187494-MS. 10.2118/187494-MS.
Grujic, O. 2017. Subsurface Modeling with Functional Data. PhD thesis, Stanford University, Stanford, California (September 2017). -. 2018. An R package that implements the methods for growing regression trees with functional output data. https://github.com/ogru/fTree. Accessed May 5, 2019.
Patel, S. 2017. SVM (Support Vector Machine) - Theory. https://medium.com/machine-learning-101/chapter-2-svm-support-vector-machine-theory-f0812effc72. Accessed May 5, 2019.
Piech,C.2013.ArtificialIntelligence:PrinciplesandTechniques. http://stanford.edui-cpiech/cs221/handouts/kmeans.html. Accessed May 5, 2019.
Wachtmeister, H., Lund, L., Aleklett, K.. 2017. Production Decline Curves of Tight Oil Wells in Duvernay Shale. Natural Resources Research 26 (3): 365-377. 10.1007/s11053-016-9323-2.