The transition to the late stages of development at a significant number of large oil fields, as well as the need to develop hard-to-recover reserves, made necessitate the search for the optimal method of oil production. At the same time, despite the large number of works in this direction, the problems of the scientific and methodological substantiation for the choice of the optimal method of artificial lift remain relevant. Earlier, the author proposed criteria for assessing the operating efficiency of downhole pumping equipment and the success criteria for the application of new methods of artificial lift, as well as regression equations for determining the ranges of inefficient operation of sucker rod pump (SRP) units and electric submersible pump (ESP) units. At the same time, the classification of statistical data was made with assumptions about the independence and homogeneity of samples for the ranges of operating parameters. As a result of the conducted research, these assumptions were confirmed, as well as the stability of statistical conclusions when changing the boundaries of the ranges of operational parameters of downhole pumping equipment.
It has been established that for lowering depths of more than 1500 m low failure times, and the low values of flow coefficient for pumping rate is under 20 m3/day for the considered oil company are characteristic. These ranges correspond to the suboptimal range of operating parameters SRP and ESP. The combination of the optimal ranges method, taking into account economic efficiency, and statistical methods will allow to improve the existing methods of analysis of field data for their implementation in software products of monitoring and management of the wells.
The following statistical methods were used in the research: exploratory data analysis, Student test, Mann – Whitney test, single-factor analysis of variance, multiple linear regression analysis, generalized regression models, Kendall rank correlation, Spearman rank correlation, gamma rank correlation, permutation test.
References
1. Medvedev A.V., Povyshenie bezopasnosti i nadezhnosti ekspluatatsii oborudovaniya neftedobychi (Improving the safety and reliability of oil production equipment): thesis of doctor of technical science, Ufa, 2009.
2. Slepchenko S.D., Otsenka nadezhnosti UETsN i ikh otdel'nykh uzlov po rezul'tatam promyslovoy ekspluatatsii (Evaluation of the ESP reliability and their individual nodes according to the exploitation results of oil fields): thesis of candidate of technical science, Perm', 2011.
3. Mel'nichenko V.E., Otsenka vliyaniya osnovnykh tekhnologicheskikh kharakteristik dobyvayushchikh skvazhin na resurs pogruzhnykh elektrotsentrobezhnykh nasosov (Assessment of the impact of the main technological characteristics of producing wells on the resource of submersible electric centrifugal pumps): thesis of candidate of technical science, Moscow, 2017.
4. Kuchumov R.Ya., Uzbekov R.B., Optimizatsiya glubinnonasosnoy neftedobychi v usloviyakh Bashkirii (Optimization of bottomhole pumping in the conditions of Bashkiria), Ufa: Bashkirskoe knizhnoe izdatelstvo Publ., 1986, 160 p.
5. Shakirov A.M., Kompleksnyy podkhod k vyboru ratsional'nogo sposoba mekhanizirovannoy dobychi na neftyanom mestorozhdenii pri neopredelennosti vkhodnykh dannykh (An integrated approach to the choice of a rational method of machine mining in oil field with uncertainty of input data): thesis of candidate of technical science, Moscow, 2012.
6. Volkov M.G., Khalfin R.S., Topol'nikov A.S. et al., Approaches to justification of selection of the application field for new artificial lift method (In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2019, no. 3, pp. 96–100.
7. Lemeshko B.Yu., Lemeshko S.B., Postovalov S.N., Chimitova E.V., Statisticheskiy analiz dannykh, modelirovanie i issledovanie veroyatnostnykh zakonomernostey. Komp'yuternyy podkhod (Statistical data analysis, modeling and investigation of probability laws. Computer approach), Novosibirsk: Publ. of NSTU, 2011, 888 p.
8. Trevor H., Tibshirani R., Friedman J., The elements of statistical learning. Data mining, inference, and prediction, New York: Springer, 2009, 764 p.
