Creation of intelligent block of neural network prediction leakage current values
DOI:
https://doi.org/10.31548/energiya1(65).2023.115Abstract
Means of monitoring the magnitude the leakage current have proven themselves as an effective technical method monitoring the state of the insulation the electric motor. The use of technical means that allow not only to fix, but also to predict the dangerous values the leakage current, make it possible to inform the service personnel about the possible danger in advance. Thanks to this, the time for simple electrical equipment is reduced and it becomes possible to carry out maintenance, repair or replacement electric motors during a technological pause without waiting for their complete failure. The use of neural networks for predicting the reliability of electric motors has proven to be effective for predicting these complex processes.
Based on the data conducted passive experiment, two neural networks were synthesized. A comparison the operating features a neural network based on technological parameters and a neural network based on the theory of time series forecasting indicates the need to combine them to obtain a better forecast the leakage current value. This led to the need to create a selection criterion and synthesize a hybrid neural network that will work according to this criterion.
Key words: leakage current, selection criterion, hybrid neural network
References
Zagirnyak, M., Prus, V., Somka, O. (2015). Reliability Models of Electric Machines with Structural Defects Proceedigs 2015 16th International Conference on “Computational Problems of Electrical Engineering”
CPEE – 2015. Lviv, 249-251.
Gerasymenko, V., Kozyrskyi, V., Maiborodina, N., Kovalov, O. (2019). Mathematical Model Changing the Value of the Process of Leakage Current in 0.38 kV Networks. Modern Development Paths of Agricultural Production. Trends and Innovations. Cham: Springer International Publishing, 339 – 348.
Zaiets N., Kondratenko I. (2019) Development of an Intelligent System for Predicting the Reliability of Electric Motors. IEEE 39th International Conference on Electronics and Nanotechnology (ELNANO). 2019, Kyiv, 614-619.
Lysenko. V. P., Reshetiuk, V. M., Shtepa, V. M., Zaiets, N. A. (2014). Systemy shtuchnoho intelektu: nechitka lohika, neironni merezhi, nechitki neironni merezhi, henetychnyi alhorytm [Artificial intelligence systems: fuzzy logic, neural networks, fuzzy neural networks, genetic algorithm]. Kyiv, 336.
Gerasymenko, V. P. (2022). Rozrobka kryteriyu vyboru prohnozovanykh znachenʹ strumu vytoku syntezovanykh neyromerezh [Development of a criterion for selecting predicted values of the leakage current of synthesized neural networks]. Enerhetyka i avtomatyka, 5. 52-61.
Kondratenko, I. P., Zaiets, N. A., Shtepa, V. M. (2020). Naukovi osnovy keruvannia elektrotekhnichnymy kompleksamy neperervnykh vyrobnytstv iz prohnozuvanniam neshtatnykh sytuatsii: monohrafyia [Scientific bases of management of electrotechnical complexes of continuous productions with forecasting of abnormal situations: monograph]. Kyiv: Printeko, 256.
Lysenko, V. P., Zayets, N. A. Shtepa, V. M., Dudnyk, A. O. (2011). Neiromerezheve prognozuvannia chasovih riadiv temperaturi navkolishniogo prirodnogo seredovishcha [Neural network forecasting of time series of external temperature], Bioresursy і pryrodokorystuvannia, №3 – 4, 102 – 108.
Gerasymenko, V., Vasylenko, V., Maiborodina, N., Kovalov, O. (2022). Neyromerezheve prohnozuvannya strumu vytoku na osnovi tekhnolohichnykh parametriv [Neural network forecast of leak current based on technological parameters]. Enerhetyka i avtomatyka, 3, 109 – 118.
Gerasymenko, V., Vasylenko, V., Maiborodina, N., Kovalov, O. (2022). Neyromerezheve prohnozuvannya strumu vytoku na osnovi teoriyi chasovykh ryadiv [Neural network prediction of leakage current based on the theory of time series forecasting ]. Enerhetyka i avtomatyka, 4, 86 – 93.
Downloads
Published
Issue
Section
License
Relationship between right holders and users shall be governed by the terms of the license Creative Commons Attribution – non-commercial – Distribution On Same Conditions 4.0 international (CC BY-NC-SA 4.0):https://creativecommons.org/licenses/by-nc-sa/4.0/deed.uk
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
