Prompt forecasting of heat flows under fire conditions in a vertical steel tank having an ANFIS protective wall

Introduction. Tanks and tank farms are widespread in many constituent entities of the Russian Federation and are among the most important elements of the model for the production, treatment, transportation and processing of oil and oil products. It is relevant both at the Russian and global levels to ensure that fire safety is arranged for tank farms to reduce highest risk levels according to the risk-based safety model. In the context of information and communication technology (ICT) developments and introduction of ICT into the operation and management of various facilities, over the past decades advanced methods have emerged for predicting the occurrence and development of emergency situations at facilities and enhancing management decisions on containment and elimination of emergency situations including fires.

Goals and objectives. In this paper, the authors present a model that they developed to promptly forecast heat flows using artificial neural networks. The forecast model will improve the safety of fire brigade personnel responsible for extinguishing fires inside ground-based vertical steel tanks having protective walls. In the research, the authors place special emphasis on identifying dependence between the heat flow and the wind load. Methods. To achieve this goal, the authors arranged and conducted a series of experiments, collected experimental data on heat flows, and created training and test samples.

Results. Dependences between heat flows and environmental factors were identified by constructing adaptive neuro-fuzzy inference systems or adaptive network-based fuzzy inference systems (ANFIS). Various types of membership functions, optimisation and system generation methods were compared and it was found out that for ANFISs, prediction of heat flows with regard to and disregarding wind loads were optimal, if subcluster and hybrid optimisation methods were used, as they had lowest error values for samples.

Discussion. The analysis shows that wind speed and tank location can rise temperatures of the air, tank wall and petrol. Therefore, despite the complexity of the analysis, the regard for all these factors makes it possible to identify a safe distance between a burning tank and firefighters.

Conclusions. The research made it possible to develop a model for prompt heat flow forecasting with the help of artificial intelligence elements (ANFIS). The results obtained in the course of the work make it possible to increase the efficiency of prompt forecasting of the dynamic behaviour of fire inside tanks and tank farms and optimize managerial decision-making by responsible persons.

1. Krasnov A.V., Sadykova Z.Kh., Perezhogin D.Yu., Mukhin I.A. Statistics of emergency accidents in the refining and petrochemical industry in 2007-2016. Oil and Gas Business. 2017; 6:179-191. URL: https://www.elibrary.ru/item.asp?id=30784379 (rus.).

2. Demekhin F.V., Tarantsev A.A., Belov D.I. On the problem of fire extinguishing in reservoirs having ring-type protective walls. Vestnik Sankt-Peterburgskogo Universiteta GPS MCHS Rossii. 2013; 2:68-75. URL: https://vestnik.igps.ru/wp-content/uploads/V52/13.pdf (rus.).

3. Petrova N.V., Cheshko I.D., Galishev M.A. Analysis of the expert practice of fire research into oil storage facilities. Vestnik Sankt-Peterburgskogo Universiteta GPS MCHS Rossii. 2016; 3:40-46. URL: https://vestnik.igps.ru/wp-content/uploads/V83/7.pdf (rus.).

4. Blinov V.I., Khudyakov G.N., Petrov I.I. On the mechanism of extinguishing the combustion of oil products in tanks by mixing them with air. Moscow, Information collection TsNIIPO, 1958; 34-38. (rus.).

5. Petrov I.I., Reutt V.Ch. Critical conditions for extinction of the flame of an oil product when it is stirred in a tank. Moscow, Information collection TsNIIPO, 1959; 58-61. (rus.).

6. Volkov O.M., Proskuryakov G.A. Fire safety at the enterprises performing transportation and storage of oil and oil products. Moscow, Nedra publ., 1981; 256. (rus.).

7. Abramov Yu.A., Basmanov A.E. Minimizing fire damage in tank farms. Pozharovzryvobezopasnost/ Fire and Explosion Safety. 2007; 4:59-65. URL: https://www.elibrary.ru/item.asp?id=12512034

8. Zabetakis M.G., Burgess D.S. Research on hazards associated with production and handling of liquid hydrogen. Bureau of Mines, Washington, D.C., 1961. DOI:10.2172/5206437

9. Markstein G.H., De Ris J. Wall-fire radiant emission. Part 1: Slot-burner flames, comparison with jet flames. Symposium (International) on Combustion. 1991; 23(1):1685-1692. DOI: 10.1016/S00820784(06)80443-1

10. Markstein G.H., De Ris J. Wall-fire radiant emission — Part 2: Radiation and heat transfer from porous-metal wall burner flames. Symposium (International) on Combustion. 1992; 24(1):1747-1752. DOI: 10.1016/S0082-0784(06)80204-3

11. Drysdale D. An introduction to fire dynamic. John Wiley & Sons, 2011; 512.

12. Babrauskas V. Estimating large pool fire burning rates. Fire Technology. 1983; 19(4):251-261. DOI: 10.1007/BF02380810

13. Munoz M., Arnaldos J., Casal J., Planas E. Analysis of the geometric and radiative characteristics of hydrocarbon pool fires. Combustion and Flame. 2004; 139(3):263-277. DOI: 10.1016/j.combustflame.2004.09.001

14. Shvyrkov S.A., Vorobyev V.V., Ibatulin R.K. The danger of a cascading fire spreading in a tank farm with a wind effect. Technology of Technosphere Safety. 2017; 4(74):8. URL: http://agps-2006.narod.ru/ttb/2017-4/26-04-17.ttb.pdf (rus.).

15. Rubtsov D.N., Shalymov M.S. About the spread of fire in a tank of the ‘glass in the glass’ type containing oil and oil products. Technology of Technosphere Safety. 2016; 3(67):8. URL: http://agps-2006.narod.ru/ttb/2016-3/23-03-16.ttb.pdf (rus.).

16. Schälike S., Mishra K.B., Malow M., Berger A., Wehrstedt K.D., Schönbucher A. Mass burning rate of a large TBPB pool fire — experimental study and modeling. Chemical Engineering Transactions. 2013; 31:853-858. DOI: 10.3303/CET1331143

17. Shalymov M.S. Influence of thermal loads fire oil tanks on the body neighboring heated tank. Technology of Technosphere Safety. 2015; 2(60):8. URL: http://agps-2006.narod.ru/ttb/2015-2/16-02-15.ttb.pdf (rus.).

18. Anwar Z., Afzal H., Bibi N., Abbas H., Mohsin A., Arif O. A hybrid-adaptive neuro-fuzzy inference system for multi-objective regression test suites optimization. Neural Computing and Applications. 2019; 31(11):7287-7301. DOI: 10.1007/s00521-018-3560-8

19. Zaichenko Yu.P., Sevaee F. The investigation of fuzzy neural network ANFIS in the problem of macroeconomic indexes forecasting. System Research & Information Technologies. 2005; 1:100-112. URL: http://dspace.nbuv.gov.ua/handle/123456789/13765 (rus.).

20. Leitea R.M., Centenob F.R. Effect of tank diameter on thermal behavior of gasoline and diesel storage tanks fires. Journal of Hazardous Materials. 2018; 342:544-552. DOI: 10.1016/j.jhazmat.2017.08.052

21. Yun-Wei Zhanga, Zhao-Lin Gua, Zan-She Wanga, Chuck Wah Francis Yua. Revisit of wind flows between wind tunnels and real canyons – the viewpoint of Reynolds dynamic similarity. Indoor and Built Environment. 2013; 22(3):467-470. DOI: 10.1177/1420326X12489064