**S.L. Golofast, Dr. Sci. (Eng.), Prof., trasser@inbox.ru
OOO Gazprom Proyektirovaniye, Saint Petersburg, Russia
**

Currently the approach based on risk criteria is used as a methodological basis for preventing accidents and emergencies in potentially hazardous facilities. The initial step at the implementation of this approach for extended linear objects is the analysis of technical risk, the indices of which are determined by the relevant methods of reliability theory. One of the main steps of technical risk analysis is the preliminary assessment of failure probability of the linear sections of trunk oil and gas pipelines, which shall be performed at the design and operation stages of the objects under study. Calculation of the probability of failure is carried out in accordance with the regulatory framework, the basis of which are the methods of the probability theory and mathematical statistics. At the probabilistic calculations the stresses emerging in the pipeline wall are considered as random values. The limiting stresses, which depending on the calculation norms take the ultimate strength, the yield strength or the fatigue limit, are also the random values. The correctness of the results of technical risk analysis depends on the laws of these random values distribution, which are adopted at the calculations of failure probability.

The accumulated experience shows that in most cases it is impossible to formulate the conditions for referring the selection at processing the results of measurements and experimental studies of the random values under consideration to any parametric law, and also to identify the distribution law using fitting criterion. The basis for determining the parameters of the law of distributing the random values at calculating the probability of failure-free operation are the selections of stresses arising in the pipeline wall and the limiting stresses allowed by the pipe material. In consequence of this, the distribution of the random values is reduced both from the left and from the right, which leads to a change in the statistical characteristics of the distribution.

The listed problems show that the existing methods for determining the probability of failure-free operation of linear sections of the trunk pipelines based on the application of methods of probability theory and classical (parametric) mathematical statistics have a number of significant limitations. They result in systematic unknown errors when analyzing the technical risk of the objects under study.

1. Syzrantsev V.N., Golofast S.L. Probabilistic assessment of pipelines strength reliability. Truboprovodnyy transport: teoriya i praktika = Pipeline Transport: Theory and Practice. 2011. № 5 (27). pp. 14–22. (In Russ.).

2. Syzrantsev V.N., Novoselov V.V., Sozonov P.M., Golofast S.L. Assessment of safety and strength reliability of the trunk pipelines using nonparametric statistics. Novosibirsk: Nauka, 2013. 172 p. (In Russ.).

3. Bertsche B., Lechner G. Zuverlässigkeit im Fahrzeug- und Maschinenbau. Berlin: Springer, 2004. 495 p.

4. Filatov A.A., Golofast S.L. Problems of assessment of gas pipelines strength reliability. Gazovaya promyshlennost = Gas Industry. 2015. № 7 (725). pp. 45–48. (In Russ.).

5. Kapur K., Lamberson L. System reliability and design. Мoscow: Мir, 1980. 604 p. (In Russ.).

6. Kharionovskiy V.V. Reliability and service life of gas pipeline constructions. Мoscow: Nedra, 2000. 466 p. (In Russ.).

7. STO Gazprom 2-2.3-184—2007. Methods for calculating and justifying the safety factor and stability of the trunk gas pipelines at the stage of operation and maintenance. Available at: http://gostrf.com/norma_data/58/58957/index.htm (accessed: February 20, 2018). (In Russ.).

8. O`Connor P.D.T. Practical Reliability Engineering. John Wiley & Sons, 2001.

9. Haibach E. Betriebsfestigkeit: Verfahren und Daten zur Bauteilberechnung. Berlin: Springer, 2002.

10. O’Connor P.D.T., Kleyner A. Practical Reliability Engineering. 5th Ed. John Wiley & Sons, Ltd, 2012. 485 p.

11. Dzhonson N., Lion F. Statistics and experiment planning in engineering and science. Methods of data processing. Мoscow: Мir, 1980. 610 p. (In Russ.).

12. Birillo I.N., Yakovlev A.Ya., Teplinskiy Yu.A., Bykov I.Yu., Voronin V.N. Assessment of the strength resource of gas pipes with corrosion damage. Мoscow: TsentrLitNefteGaz, 2008. 168 p. (In Russ.).

13. Syzrantsev V.N., Nevelev Ya.P., Golofast S.L. Adaptive methods for recovering the probability density function. Izvestiya vysshikh uchebnykh zavedeniy. Mashinostroenie = Izvestiya of Higher Educational Institutions. Mechanical Engineering. 2006. № 12. pp. 3–11. (In Russ.).

14. Filatov A.A., George M.S. Influence of operating conditions of the gas pipeline on the indices of its strength reliability. Nauka i tekhnika v gazovoy promyshlennosti = Science and Technology in Gas Industry. 2013. № 2 (54). pp. 75–82. (In Russ.).