The purpose of the research is the improving the efficiency of operation of sucker-rod pumping unit (SRPU) submersible equipment due to the operational diagnosis of the condition. The solved problem is the development of an algorithm for recalculating in the dynamogram the dependence of the power consumption of the rocking machine electric drive of moving the rod column suspension point (wattmeterogram) based on the analysis of the technological time series of the accumulated parameters. The analysis of the wattmetering data allows to determine condition of the surface and underground equipment of the SRPU. The advantages of wattmetering are the ease of measurement and the lack of additional equipment to record the force parameters at the rod column suspension point. The disadvantage is an increase in the performance requirements of the computational core of the controller of the SRPU control station for the implementation of diagnostic algorithms.
To date, the problem is in approximation of the functional relationship between the samples of the dynamogram and the wattmeterogram, which does not allow the using of well-known diagnostic methods by the dynamogram calculated from wattmetering data.
An algorithm for converting a wattmeterogram into a dynamogram is proposed based on the known time dependence of the rod position and the refined kinematic model of the rocking machine. Based on an array of instantaneous values of active power during a single period of swing, obtained uniformly at same intervals from the moment of the beginning of the plunger stroke up, and the physical parameters of the rocking machine, the stroke of the rod suspension point and the force at the rod string suspension point are calculated. To eliminate the disadvantages of the recalculation algorithm using the kinematic model, a neural network approximation is proposed for the functional dependence of the wattmeterogram and the dynamogram samples.
The possibility of monitoring and diagnosis of surface and underground equipment of the SRPU by using the module for wattmetering data intellectual preprocessing in the task of controlling the operating modes of a producing well is shown.
References
1. Chigvintsev S.V., Chigvintseva A.S., Virtual'nye datchiki dlya sistemy upravleniya shtangovoy glubinnoy nasosnoy ustanovkoy (Virtual sensors for the control system of a deep-well pumping unit), Collected papers “Elektrotekhnologii, elektroprivod i elektrooborudovanie predpriyatiy” (Electrotechnology, electric drive and electrical equipment of enterprises), Proceedings of II Vserossiyskoy nauchno-tekhnicheskoy konferentsii, 2009, p. 197–200.
2. Tagirova K.F., Vul'fin A.M., Sabitov A.R., Akhmetov N.B., Updating of the diagnostics of well sucker-rod pumping units on the basis of intelligent analysis of dynamometric data (In Russ.), Avtomatizatsiya, telemekhanizatsiya i svyaz' v neftyanoy promyshlennosti, 2014, no. 11, pp. 23-28.
3. Khakim'yanov M.I., Pachin M.G., Monitoring of sucker rod pump units on result of the analysis wattmeter cards (In Russ.), Neftegazovoe delo = The electronic scientific journal Oil and Gas Business, 2011, no. 5, pp. 26–36.
4. Krichke V.O., Measuring information system for wells equipped with pumping jack IIS-SK (In Russ.), Avtomatizatsiya i telemekhanizatsiya v neftyanoy promyshlennosti, 1976, no. 11, pp. 16–18.
5. Aliev T.M., Ter-Khachaturov A.A., Avtomaticheskiy kontrol' i diagnostika skvazhinnykh shtangovykh nasosnykh ustanovok (Automatic monitoring and diagnostics of downhole sucker rod pumping units), Moscow: Nedra Publ., 1988, 232 p.
6. Dregotesku N.D., Glubinnonasosnaya dobycha nefti (Oil production using deep-well pumping unit), Moscow: Nedra Publ., 1966, 294 p.
7. Osovskiy S., Neyronnye seti dlya obrabotki informatsii (Neural networks for information processing), Moscow: Finansy i statistika Publ., 2004, 344 p.
