Method for evaluating technical condition of aggregates based on artificial intelligence
Authors:
M. l. Horbiichuk, Dr. Sc. (Tech.), Prof., orcid.org/0000-0002-8586-1883, Ivano-Frankivsk National Technical University of Oil and Gas, Ivano-Frankivsk, Ukraine, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
O. T. Bila, orcid.org/0000-0003-2245-7434, Ivano-Frankivsk National Technical University of Oil and Gas, Ivano-Frankivsk, Ukraine, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Y. I. Zaiachuk, Cand. Sc. (Tech.), Assoc. Prof., orcid.org/0000-0001-8705-2724, Ivano-Frankivsk National Technical University of Oil and Gas, Ivano-Frankivsk, Ukraine, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
T. V. Humeniuk, Cand. Sc. (Tech.), Assoc. Prof., orcid.org/0000-0003-2610-2550, Ivano-Frankivsk National Technical University of Oil and Gas, Ivano-Frankivsk, Ukraine, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Abstract:
Purpose. Enhancing operational efficiency of natural gas pumping units by applying artificial intelligence methods to assess their technical condition.
Methodology. The artificial neural networks theory, in particular, the counter-propagation network with two layers of Kohonen and Grossberg, was used to recognize aggregate states. The structure choice and calculation of the separation line coefficients was carried out using genetic algorithms.
Findings. The task of evaluating the aggregate technical state is formed as a pattern recognition task. An analysis of literary sources has shown that the problem of pattern recognition relates to difficult-to-formulize tasks, and their solution requires new approaches, based on artificial intelligence methods. The artificial neural networks of counter propagation are proposed to use for recognizing the aggregate technical states. The article shows advisability of using the LVQ-network type, which has two layers of Kohonen and Grossberg in its composition, for this assessment. The efficiency of the network is confirmed by a test case. A genetic algorithm that allows choosing both the polynomial structure and its parameters is used to construct a diving line separating one class of signs from another. The technical condition of the lubricating system of the natural gas pumping aggregate is estimated, as an example of the developed methodology application.
Originality. The AI-based method for assessing the technical state of gas-pumping units has been further developed to evaluate their state in operating mode and, based on this, develop effective algorithms for optimal loading of parallel operating aggregates.
Practical value. Algorithmic and software testing was developed on the test case, based on the proposed method for assessing the aggregate technical condition. The proposed method, which effectively solves the problem of partitioning the signs planes into classes each of which characterizes its certain state, has been shown as an example of the technical state evaluation of the gas turbine engine lubrication system.
References.
1. Bulatov, A. I. (2015). Concept of quality of drilled oil and gas wells. Drilling and oil, (12), 15-19.
2. Dobrotvor, I. G., Chykhira, I. V., & Yarema, I. T. (2017). Conditions of oscillation of kinetics characteristics of microstructures of composites. Bulletin of the Khmelnytskyi National University. Technical Sciences, (4), 52-55.
3. Gorbiychuk, M. I., & Pashkovskyi, B. V. (2016). Method of determining the generalized coefficient of technical state of a gas pumping unit on the basis of fuzzy logic and genetic algorithms. Methods and devices of quality control, 2(37), 102-107.
4. Zhao, Z., Chu, L., Tao, D., & Pei, J. (2018). Classification with label noise: a Markov chain sampling framework. Data Mining and Knowledge Discovery, 1-37.
5. Gershilov, A. A. (2014). Mathematical methods for decision making. Moscow: Publishing house of Moscow State Technical University. N. E. Bauman.
6. Marchenko, O. O. (2016). Machine-learning methods for recognizing named entities of text. Programming problems, (2-3) (Special issue), 150-157.
7. Walkman, Yu. R., & Stepashko, P. V. (2015). Towards building an ontology of intellectual modeling. Inductive modeling of complex systems: Collection of research papers, (7), 101-115.
8. Zygmunt Lech Warsza, & Zabolotnii, S. W. (2018). Polynomial estimation of measurand parameters for samples from non-Gaussian distributions based on higher order statistics. Advanced Mathematical and Computational Tools in Metrology and Testing XI, 383-400.
9. Zabolotnyi, S. V., Chepinoga, A. V., Bondarenko, Yu. Y., & Rud, M. P. (2018). Polynomial parameter estimates for exponentially distributed data. Visnyk NTUU “KPI”. Seriia Radiotekhnika Radioaparatobuduvannia, 75, 40-47.
10. Huang, Huajuan, Wei, Xiuxi, & Zhou, Yong-Quan (2018). Twin Support Vector Machines: A Survey. Neurocomputing.
11. Norkin, V. I. (2017). B&B solution technique for multicriteria stochastic optimization problems. Optimization Methods and Applications.
12. Bodyanskiy, Ye., Vynokurova, O., Savvo, V., Tatiana Tverdokhlib, T., & Mulesa, P. (2016). Hybrid Clustering-Classification Neural Network in Medical Diagnostics of Reactive Arthritis. International Journal of Intelligent Systems Technologies and Applications, 8(8), 1-9. https://doi.org/10.5815/ijisa.2016.08.01.
13. Nikolenko, S., Kadurin, A., & Arkhangelskaya, E. (2017). Deep learning. “Peter” Publishing House.
14. Horbiichuk, M. I., Kohutiak, M. I., & Zaiachuk, Y. I. (2008). Inductive method for constructing mathematical models of natural gas pumping units. Oil and gas industry, (5), 32-35.
15. Glybovets, M. M., Gulaeva, N. M., & Pasichnyk, M. M. (2015). A Parallel Genetic Algorithm for Creating a Work Schedule. Programming Problems, (2), 76-85.
16. Horbiichuk, M. I., & Shchupak, I. V. (2010). Diagnostic model of natural gas supercharger lubrication system. Scientific Bulletin of the Ivano-Frankivsk National Technical University of Oil and Gas, 3(25), 130-137.
Newer news items:
- Analysis of economic and mathematical modeling of industrial enterprise functioning at multicollinearity based on parameterization - 10/05/2020 08:18
- Innovation development of transition countries in the context of their participation in global value chains - 10/05/2020 08:16
- Illegal mining of mineral resources of the national importance: issues of subsoil protection with criminal-legal remedies - 10/05/2020 08:14
- The WTO toolkit in regulating interstate trade and economic cooperation - 10/05/2020 08:12
- Paradoxes of economic development: science and innovation in the modern world - 10/05/2020 08:11
- Information asymmetry as a determinant of competition - 10/05/2020 08:09
- Improving the quality of civil service management in Ukraine - 10/05/2020 08:07
- Technology for determining the effectiveness of international cooperation in the mining industry - 10/05/2020 08:05
- Formation of metallurgical enterprise sustainable development portfolio using the method of analyzing hierarchies - 10/05/2020 08:04
- Implications of total quality management in Ukrainian higher education institutions: international experience - 10/05/2020 08:02
Older news items:
- Distribution of inorganic nitrogen compounds in purification of storm wastewater of the engine-building manufactory - 10/05/2020 07:59
- Development of new composite cement based on waste rocks from Djebel Onk phosphate deposit (Tebessa-Algeria) - 10/05/2020 07:56
- Modeling the process of formation of stagnation zones at hazardous production facility with application of CFD technologies - 10/05/2020 07:55
- The environmental reliability of gas-fired boiler units by applying modern heat-recovery technologies - 10/05/2020 07:13
- Effect of filtering box parameters on the protective action of gas filters - 10/05/2020 07:11
- Calculation of static and dynamic losses in power IGBT transistors by polynomial approximation of basic energy characteristicss - 10/05/2020 07:09
- Modeling the development of machine-building industry on the basis of the fuzzy sets theory - 10/05/2020 07:07
- Effect of the chirp on interactions of pure-chirped solitons - 10/05/2020 07:04
- The use of deforming broaching for enhancing the efficiency of cutter chisels - 10/05/2020 07:02
- Development and research of thеrmоplаstiс methods for hardening details - 10/05/2020 07:00