The paper describes computer-assisted injection well placement technique. The study focuses on the fields of Tatneft PJSC. The technique is based on the proxy models of production targets and has been implemented in Epsilon software package. Epsilon software package is designed to automate long-term planning of production enhancement operations on a variety of oil fields through generation of multiple development scenarios in proxy-models, estimation of production and economic performance and optimization of investment portfolio using high-performance computing and machine learning.
Epsilon includes a software module which uses a proxy model of the field for step-wise introduction of proposed production locations for drilling based on irregular well pattern having the highest possible well spacing density and complying with technological and economic constraints (generation of drilling schedule). The algorithms designed by the authors are used to select proposed injection well locations from a set of drilling sites "rejected" due to geological or economic criteria at the stage of drilling schedule generation. The first algorithm solves the problem of selecting the limited optimal number of proposed injection wells from a set of "rejected" locations, considering the certain constraints (minimum and maximum spacing to drilled or proposed production wells, presence of responding wells, effects on a limited (maximum allowable) number of responding wells. The algorithm is implemented. It was developed via the Python 3.6 programming language. Optimization problem is solved using lpSolveAPI R-interface. The second algorithm provides iterative division of the entire target region into squares of a given area and determination of candidates for conversion to injection from "rejected" locations within each square. For regions that contain no drilled injection wells or "rejected" locations in the vicinity of proposed production wells, proposed injection wells are selected from proposed production wells. The algorithm was developed using C++ programming language.
References
1. Certificate of state registration of a computer program no. 2020661783 RF. Estimating Performance of System Investment in Long-term Oil production using Neuronet (Epsilon), Authors: Nasybullin A.V., Girfanov R.G., Denisov O.V., Lazareva R.G., Latifullin F.M., Sattarov R.Z., Khafizov R.R., Chirikin A.V., Sharifullina M.A.
2. Khisamov R.S., GanievB.G., Galimov I.F. et al., Computer-aided generation of development scenarios for mature oil field (In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2020, no. 7, pp. 22–25, DOI: https://doi.org/10.24887/0028-2448-2020-7-22-25.
3. Zvezdin E.Ju., Mannapov M.I., Nasybullin A.V. et al., Stage-wise optimization of project well pattern using oil reserves evaluation program module (In Russ.), Neftjanoe hozjajstvo = Oil Industry, 2019, no. 7, pp. 28–31, DOI: https://doi.org/10.24887/0028-2448-2019-7-28-31
4. Certificate of state registration of a computer program no. 2021680284 RF. Epsilon 1.1, Authors: Latifullin F.M., Sattarov Ram. Z., Khafizov R.R., Sharifullina M.A.
5. Taha H.A., Operations research: An introduction, Prentice Hall, 2006, 838 p.
