The principles of constructing artificial neural networks for a quality control system for the operation of ship equipment related to environmental protection are considered. The concentration of harmful substances in exhaust gases and bilge waters depends on many factors related to both the condition of the equipment and external conditions. Analytically describing this dependence is extremely difficult, therefore, it is proposed to use artificial neural networks to monitor the state of equipment. The paper describes how to create a neural network such as a self-organizing feature map and methods for its training.
environmental protection, monitoring, artificial neural network, training of the neural network, vector quantization
1. Zinenko H.H. Metodika kontrolya sostava otrabotannyh gazov sudovyh malooborotnyh dvigateley v ekspluatacii / H.H. Zinenko, N.I. Nikolaev, V.E. Panamarev // Vestnik Gosudarstvennogo universiteta morskogo i rechnogo flota imeni admirala S. O. Makarova. -2015. -№ 3 (31). - S. 108-116.
2. Kruglov V.V. Iskusstvennye neyronnye seti. Teoriya i praktika / V.V. Kruglov, V.V.Borisov,- M.: Goryachaya liniya-Telekom, 2002 - 382 s.
3. Kohonen T. Samoorganizuyuschiesya karty,- M.: BINOM. Laboratoriya znaniy, 2008,- 655 s.
4. Osovskiy S. Neyronnye seti dlya obrabotki informacii- M.: Finansy i statistika, 2002,- 343 s.
5. Tynde Y., Buzo A., Gray R. An algorithm for vector quantizer desingn // IEEE Trans. Comm., 1980,- Vol. 28,-Pp. 84-95.
6. Studenikin D.E. Estimation of vessel's movement with the aid of fuzzy logic based hierarchy systems (in english) [tekst] / D.E. Studenikin, E.V. Khekert, M.A. Modina // Morskie intellektual'nye tehnologii. -2018,-T. E- № 1 (39).-S. 205-208.
7. Karakaev A.B., Lukanin A.B., Hekert E.V. Razrabotka metodologii, metodov i modeley analiza vliyaniya razlichnyh variantov postroeniya struktury i rezhimov podderzhaniya i vosstanovleniya rabotosposobnosti sudovyh elektroenergeticheskih sistem (chast' 1) // Ekspluataciya morskogo transporta,- 2016,- № 3(80).- S.54-60.
