Automated technological complexes in the oil and gas industry are characterized by increasing requirements for the continuity of their equipment. The development of technology, the increasing complexity of the processes, digitalization of production - these are the main reasons for the changes. At the same time, the exponentially growing volume of information circulating in the control system creates a stable platform for using modern data analysis methods in order to identify hidden patterns in them. The revealed patterns, in turn, allow conclusions to be drawn about the causes of emergency and pre-emergency events that occurred at technological facilities. Such an analysis is currently carried out manually, since it requires a highly qualified expert whose functions are difficult to formalize. The essence of the proposed solution is that modern intelligent systems allow, if not completely exclude a person, then significantly reduce his labor costs.
The developed software package implements an approach called RCA (root cause analysis) using the Data Mining method. This is an intensively developing method of analyzing large amounts of information, different from the classical methods of mathematical statistics. It allows detecting previously unknown, non-trivial, but practically useful and accessible interpretations of knowledge necessary for decision-making in raw data. The result of the work is a program written in the Delphi language, which provides the expert with a number of tools that significantly reduce his labor costs and expand opportunities. Among them: the tool for event processing, which allows you to sel ect the most vulnerable positions in the analyzed data file; the tool for constructing histograms of the distribution of alarms over time; the tool for analyzing the dependencies of events fr om previous events; the tool for analyzing failures and determining their criticality.
Further development of the work involves the expansion of the list of supported information systems, as well as the complication of data processing algorithms. Thus, the creation of a common base of failures of an enterprise equipped with an automated expert system will make it possible to predict the development of failure situations at facilities. This will provide an early response to such events and will significantly improve the reliability of automated technological systems in the oil and gas industry.
References
1. Baffers J., Kose M.A., Ohnsorge F., Stocker M., The Great plunge in oil prices: causes, consequences, and policy responses, policy research, World bank group, 2015, 60 p.
2. Kozlov V.V., Lapik N.V., Popova N.V., Optimization of operational costs when operating an electric-centrifugal pumping unit (In Russ.), Avtomatizatsiya telemekhanizatsiya i svyaz' v neftyanoy promyshlennosti, 2016, no. 6, pp. 43–48.
3. Gao C., Rajeswaran R., Nakagawa E., A literature review on smart well technology, SPE 106011-MS, 2007.
4. Barsegyan A.A., Kupriyanov M.S., Stepanenko V.V., Kholod I.I., Metody i modeli analiza dannykh: OLAP i Data Mining (Methods and models of data analysis: OLAP and Data Mining), St. Petersburg: BKhV-Peterburg Publ., 2004, 336 p.
5. MacLennan J., ZhaoHui Tang, Crivat B., Data Mining with Microsoft SQL Server 2008, Wiley Publishing, 2009.
