Abstract
Indicators of the object actual technical condition at some discrete point in time carry mainly information about the functioning of the object in the past and do not allow us to say about the object behavior in the upcoming period of operation. At the same time, the effectiveness of diagnosis increases significantly when the task of predicting changes in the state of an object at future points in time is solved. The ability to predict unmanageable aspects of the object operation before making a decision to put it into repair allows you to make the best choice about the timing and scope of repairs. The purpose of the study is to create models for predicting the technical condition of gas piston units, taking into account the influence of external factors, based on an array of historical data on the technical condition. Time series regression models were used in the development of models. Models have been created for predicting changes in the technical condition of gas piston units based on time series regression models, including a linear regression model, a regression model using the support vector machine method, a Gaussian process regression model and regression trees. Conclusions are drawn about the highest accuracy of the Gaussian process regression model for describing the process of changing the technical condition of gas piston units. The proposed methods made it possible to predict fairly accurately the change in the technical condition of gas piston units, taking into account the influence of external factors such as the maintenance interval, the quality of maintenance and repair, seasonal changes in the composition of associated petroleum gas and sudden changes in loads. The results can be used as a basis for the creation of predictive maintenance systems for gas piston units.
Keywords
technical condition, forecasting, gas piston unit, time series regression model, machine learning, linear regression model, support vector machine, Gaussian process, regression trees
1. Liu J., Wang W., Ma F., Yang Y., Yang C. A data-model-fusion prognostic framework for dynamic system state forecasting. Engineering Applications of Artificial Intelligence. 2012, no. 4(25), pp. 814–823. doi: 10.1016/j.engappai.2012.02.015
2. Ryabov A.A. Existing methods for estimating the residual life of oil and gas processing equipment. Neftegazovoe delo [Petroleum Engineering], 2016, no. 1, pp. 198-220. (in Russian)
3. Say Van Kvong, Shcherbakov M.V. A method for predicting the remaining resource based on data processing of multi-object complex systems. Prikaspiyskiy zhurnal: upravlenie i vysokie tekhnologii. [Caspian Journal: Control and High Technologies], 2019, no. 1 (45), pp. 33-44. (in Russian)
4. Serebryakov A.V., Steklov A.S., Titov V.G. Monitoring i prognozirovanie tekhnicheskogo sostoyaniya avtonomnykh elektrotekhnicheskikh kompleksov [Monitoring and forecasting of technical condition of autonomous electrical complexes]. Nizhniy Novgorod, Nizhny Novgorod State Technical University named after R.E. Alekseev Publ., 2018. 175 p. (in Russian)
5. Hanachi H., Liu A., Banerjee Y., Koul A. A physics-based modeling approach for performance monitoring in gas turbine engines. IEEE Transactions on Reliability. 2015, vol. 64, no. 1, pp. 197-205. doi: 10.1109/TR.2014.2368872
6. Huang Z., Xu Z., Wang W., Sun Y. Remaining useful life prediction for a nonlinear heterogeneous Wiener process model with an adaptive drift. IEEE Transactions on Reliability. 2015. Vol. 64. No. 2. Pp. 687-700. doi: 10.1109/TR.2015.2403433
7. Yaguo L., Naipeng L., Szymon G., Jing L., Stanislaw R., Jacek D. A Model-Based Method for Remaining Useful Life Prediction of Machinery. IEEE Transactions on Reliability. 2017, vol. 65, no. 3, pp. 1314-1326. doi: 10.1109/TR.2016.2570568
8. Liu J., Saxena A., Goebel K., Saha B., Wang W. An adaptive recurrent neural network for remaining useful life prediction of lithium-ion batteries. Annual Conference of the Prognostics and Health Management Society. Portland, Oregon, 2010, vol. 2(1), pp. 1-29. doi: 10.36001/phmconf.2010.v2i1.1896
9. Liu J., Wang W., Golnaraghi F. A multi-step predictor with a variable input pattern for system state forecasting. Mechanical Systems and Signal Processing. 2009, vol. 23(5), pp. 1586–1599. doi: 10.1016/j.ymssp.2008.09.006
10. Khoury E., Deloux E., Grall A., Berenguer C. On the Use of Time-Limited Information for Maintenance Decision Support: A Predictive Approach under Maintenance Constraints. Mathematical Problems in Engineering. 2013, 983595. doi: 10.1155/2013/983595
11. Mathew V., Toby T., Singh V., Rao B.M., Kumar M.G. Prediction of Remaining Useful Lifetime (RUL) of Turbofan Engine using Machine Learning. IEEE, 2017. doi: 10.1109/ICCS1.2017.8326010
12. Demidenko E.Z. Lineynaya i nelineynaya regressii [Linear and nonlinear regression]. Moscow, Finansy i statistika Publ., 1981. 302 p. (in Russian)
13. 13. Cristianini N., Shawe-Taylor J. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods. UK, Cambridge University Press, 2000. doi: 10.1017/S0263574700232827
14. Kotova D., Panov M. Gaussian processes for active learning in a classification problem. Informatsionnye tekhnologii i sistemy 2019 (ITiS 2019): Sbornik trudov 43-y mezhdistsiplinarnoy shkoly-konferentsii IPPI RAN, Perm', 17–22 sentyabrya 2019 goda [Proceedings of the 43rd Interdisciplinary School-Conference of the IPPI RAS "Information Technologies and Systems 2019 (ITiS 2019)"]. Perm, A.A. Harkevich Institute of Information Transmission Problems of the Russian Academy of Sciences Publ., 2019, pp. 471-478. (In Russian)
15. Breiman L., Friedman J., Olshen R.A., Stone C.J. Classification and regression trees. New York, Routledge, 2017. doi: 10.1201/9781315139470
16. Rajput J., Man S., Lal K., Khanna M., Sarangi A., Mukherjee J., Singh S. Data-driven reference evapotranspiration (ET0) estimation: a comparative study of regression and machine learning techniques. Environment, Development and Sustainability. 2023. doi: 10.1007/s10668-023-03978-4
Zayniev A.V., Shaydullin V.F., Khakimyanov M.I. Forecasting the Technical Condition of Gas Piston Units. Elektrotekhnicheskie sistemy i kompleksy [Electrotechnical Systems and Complexes], 2024, no. 1(62), pp. 51-55. (In Russian). https://doi.org/10.18503/2311-8318-2024-1(62)-51-55