The new principle is presented related to construction of the multipoint optoelectronic flame detection system based on the sum-total of non-coordinate optoelectronic sensors located in a special way at the protected technogenic object. The system allows to detect the flame and determine its spatial coordinates throughout the volume of the protected premises of complex geometric shape. The analysis was conducted concerning the methods for solving the problem of determining the flame spatial coordinates: multifactorial polynomial regression; regression by neural networks; numerical methods for solving nonlinear equations.

Optimal approach for determination of flame spatial coordinates is proposed, which is based on obtaining the polynomial regression function of the output signals of non-coordinate of the optoelectronic sensors. As a result, the calculation process is simplified, high speed is ensured while maintaining the required error even in dustiness of atmosphere of the protected object. The method was developed for adapting flame detection for the protected object of complex geometric shape, which allows to determine the optimal number of non-coordinate optoelectronic sensors, their spatial location and orientation. For this, the numerical methods were used for searching the minimum target function in order to detect the flame and determine its spatial coordinates with a given error over the entire volume of the protected technogenic object. The task of minimizing the target function is solved by the method of coordinate descent or the gradient method with step breaking. The results are shown related to the adaptation of the system to the conditions of the test objects in the form of rectangular conjunction of the coal mine workings. The required error of flame coordinates determination when using fourth-order polynoms for conjugating in the form of the direct intersection is ensured through the use of nine sensors for conjugating in the form of a direct branch — eight sensors, for conjugating in the form of a junction of the working at right angle — seven sensors.

- Ayruni A.T., Klebanov F.S., Smirnov O.V. Explosion hazard of the coal mines. Moscow: Gornoe delo, 2011. 262 p. (In Russ.).
- Kosterenko V.N., Timchenko A.N. Factors affecting the occurrence of methane gas explosions and coal dust in the mines.
*Gornyy informatsionno-analiticheskiy byulleten (nauchno-tekhnicheskiy zhurnal) = Mining Information and Analytical Bulletin (scientific and technical journal)*. 2011. № 7. pp. 368–377. (In Russ.). - GOST R 54777—2011. Automatic systems for localization and suppression of methane-dust-air mixture explosions in coal mines. General technical requirements. Test methods. Available at: http://docs.cntd.ru/document/1200092245 (accessed: August 14, 2019). (In Russ.).
- Analysis of the systems operating in the coal mines for localizing explosions, and the evaluation of their use efficiency: report on scientific research work. Available at: http://asvplv.ru/doc/expert_mvk.pdf (accessed: August 14, 2019). (In Russ.).
- Sidorenko A.I. Optoelectronic device for detecting fire sources and determining their two-dimensional coordinates: thesis ... of the Candidate of Technical Sciences. Biysk, 2015. 154 p. (In Russ.).
- Çetin A.E., Merci B., Günay O., Töreyin B.U., Verstockt S. Methods and Techniques for Fire Detection: Signal, Image and Video Processing Perspectives. Academic Press, 2016. 95 p. DOI: 10.1016/C2014-0-01269-5
- Icove D.J., Lyster C.T., Banwarth D.M. Passive microwave system and method for protecting a structure from fire threats. Patent US 2011/O155397 A1 United States A62C2/00. Published: June 30, 2011.
- Pesatori A., Norgia M. Infrared image system for fire location. Measurement. 2013. Vol. 46. № 10. pp. 4172–4178. DOI: 10.1016/j.measurement.2013.07.040
- Rao M.R., Borah S., Ramanna S.K., Kamruddin P.U., Rynkiewicz A., Weston C. System and method for detecting fire location. Patent WO2015092691 A1 United States G08B 17/12. Published: June 25, 2015.
- Gornostaev R.V., Osavelyuk P.A., Melnik A.A. Application of the new fire fighting technologies.
*Vestnik Sankt-Peterburgskogo universiteta GPS MChS Rossii = Vestnik of Saint-Petersburg University of State Fire Service of EMERCOM of Russia*. 2009. № 4. pp. 23–28. (In Russ.). - Töreyin B.U., Dedeoǧlu Y., Güdükbay U., Çetin A.E. Computer vision based method for real-time fire and flame detection. Pattern Recognition Letters. 2006. Vol. 27. Iss. 1. pp. 49–58. DOI: 10.1016/j.patrec.2005.06.015
- Lisakov S.A., Pavlov A.N., Sypin E.V. Application of neural networks to determine the coordinates of the seat of fire by multipoint electro-optical system. Proceedings of 15th International Conference of Young Specialists on Micro/Nanotechnologies and Electron Devices (EDM). Novosibirsk, 2014. pp. 265–269. DOI: 10.1109/EDM.2014.6882526
- Zyryanova M.N., Lisakov S.A., Pavlov A.N., Sypin E.V. Application of numerical simulation for solving the problem of determining the spatial coordinates of the fire seat by multipoint electro-optical system. Proceedings of 16th International Conference of Young Specialists on Micro/Nanotechnologies and Electron Devices (EDM). Novosibirsk, 2015. pp. 347–352. DOI: 10.1109/EDM.2015.7184561
- Lisakov S.A., Pavlov A.N., Sypin E.V., Leonov G.V. Determination of control points quantity and their location in protected object for high-speed multipoint electro-optical system for fire detection and determine its spatial coordinates. Proceedings of 18th International Conference of Young Specialists on Micro/Nanotechnologies and Electron Devices (EDM). Novosibirsk, 2017. pp. 371–377. DOI: 10.1109/EDM.2017.7981776