Article 4315

Title of the article



Shcherbakova Anna Alekseevna, Postgraduate student, Penza State University (40 Krasnaya street, Penza, Russia),
Solov'ev Vladimir Aleksandrovich, Doctor of engineering sciences, professor, sub-department of instrument engineering, Penza State University (40 Krasnaya street, Penza, Russia),
Artamonov Dmitriy Vladimirovich, Doctor of engineering sciences, professor, sub-department of autonomous information and control systems, Penza State University (40 Krasnaya street, Penza, Russia),

Index UDK



Background. Application of artificial neural networks for processing of digital signals with spectral absorptivity values of commercial grade fuels, obtained by components mixing, refers to analytical tools and can be used, for example, for identification of components, control and maintenance of desired concentration and detonation resistance values of components in production of commercial grade fuels. The purpose of this work is to develop a structure, an objective function and an activation function of the artificial neural network for fuels’ components identification and composition determination by spectral absorptivity.
Materials and methods. The problems of gasoline’s components identification, determination of composition and detonation resistance were solved using the theory of artificial neural networks. Using the method of mathematical simulations the authors found objective functions of artificial neural network for composition identification, neiron activation functions. The problem of artificial neural network training was solved by the method of backpropagation of error.
Results. Using the VBA Microsoft Excel package the authors simulated the developed mathematical model for training the artificial neural network for composition identification and determination by spectral absorptivity. The researchers submitted an application for the invention "Method of gasoline’s components identification of composition definition in real time" on 22.04.2014.
Conclusions. Artificial neural network for components identification and composition determination by spectral absorptivity in industrial information and measuring systems is designed to ensure reliability of identification, to increase accuracy of commercial grade gasoline’s composition and detonation resistance determination directly in production technological processes that will make it possible to promptly amend the fuel production technology in real time.

Key words

identification, artificeal neural network, objective function, neuron activation function, spectral absorptivity.

Download PDF

1. Bakeev K. A. Process Analytical Technology. Blackwell Publishing Ltd. 2005, 445 p.
2. Rodionova O. E., Pomerantsev A. L. Khemometrika v analiticheskoy khimii [Chemometrics in analytical chemistry]. Moscow: In-t khim. fiziki RAN, 2008, 61 p.
3. Sharaf M. A., Illmen D. L., Koval'ski B. P. Khemometrika [Chemometrics]. Leningrad: Khimiya, 1989, 272 p.
4. Khaykin S. Neyronnye seti [Neural networks]. Moscow: Vil'yams, 2006, 1104 p.
5. Nikiforov I. K. Neyrosetevoy impedansnyy metod i ustroystva identi-fikatsii i opredeleniya parametrov zhidkikh nefteproduktov: dis. kand. tekhn. nauk [Neural network impedance technique and devices for liquid oil products’ parameters identification and determination: dissertation to apply for the degree of the candidate of engineering sciences]. Cheboksary: ChGU im. I. N. Ul'yanova, 2005, 172 p.
6. Smyshlyaeva Yu. A. Modelirovanie protsessa prigotovleniya vysokooktanovykh benzinov na osnove uglevodorodnogo syr'ya v apparatakh tsirkulyatsionnogo tipa: dis. kand. tekhn. nauk [Simulation of producing high-octance gasolines on the basis of hydrocarbon raw stock in circulation-type devices: dissertation to apply for the degree of the candidate of engineering sciences]. Tomsk: NITPU, 2011, 144 p.
7. Dzhelen B., Sirstad T. Primenenie VBA i makrosov v Microsoft Excel [Application of VBA macros in Microsoft Excel]. Moscow: Vil'yams, 2006, 624 p.
8. Vesnin V. L., Muradov V. G. Izvestiya Samarskogo nauchnogo tsentra Rossiyskoy akademii nauk [Proceedings of Samara scientific center of the Russian Academy of Sciences]. 2009, vol. 11, no. 3, pp. 29–32.


Дата создания: 28.12.2015 13:30
Дата обновления: 28.12.2015 16:02