Long-term change forecasting algorithm and object state analysis based on data from electric-inductive sensors

by

S.S. Kolmogorova1,2

1St. Petersburg State Electrotechnical University “LETI” named after V.I. Ulyanov (Lenin),

 RF, St. Petersburg, Prof. Popov St., 5, literature F

2St. Petersburg State Forest Engineering University named after S.M. Kirov,

 RF, St. Petersburg, Institutsky per. 5, literature U

 E-mail: ss.kolmogorova@mail.ru

DOI: 10.33075/2220-5861-2026-1-117-128

UDC 621.317.328                        

EDN: https://elibrary.ru/kytkpd

Abstract:     

The research work presents an original algorithm for long-term forecasting of changes and analysis of object states based on data from electrical induction sensors. The suggested method takes into account the physical principles of sensor operation and the characteristics of multicomponent signals with stationary and non-stationary noise, complex degradation trajectories, and multivariate effects. The algorithm includes multi-level processing of time series using spectral analysis, detection of phase transitions, adaptive local approximation of trajectories, and hybrid prediction. The work was carried out taking into account the specifics of environmental and industrial monitoring, using data from an electro-inductive sensor. The results of experiments on models of various objects showed high prediction accuracy, surpassing classical methods and modern general neural network approaches, as confirmed by MAE, RMSE, and R² determination coefficient metrics. The algorithm provides early detection of accelerated degradation phases, stability of predictions when the noise spectrum changes, and adaptability to long time intervals. The practical significance of the algorithm lies in the possibility of its implementation in intelligent systems for monitoring technical and natural objects to increase reliability, minimize accidents and optimize maintenance.

Keywords: electro-induction sensors, electrometric measurements, prediction algorithm, electric field, environmental monitoring, technical control

REFERENCES

  1. Biryukov S.V. Issledovanie elektroindukcionnogo diskovogo datchika napryazhennosti elektricheskogo polya (Research on an electro-induction disc sensor for measuring electric field strength). Omskij nauchnyj vestnik, 2023, No. 2(186), pp. 134–139. DOI 10.25206/1813-8225-2023-186-134-139.
  2. Trushnikov K.P., Romodin A.V., and Mishurinskikh S.V. K voprosu o razrabotke metodiki dolgosrochnogo prognozirovaniya elektro-potrebleniya v elektricheskih setyah g. Permi (On the development of a methodology for long-term forecasting of electricity consumption in the electrical networks of Perm). Fundamental’nye issledovaniya, 2016, No. 11–3, pp. 604–608. URL: https://fundamental-research.ru/article/view?id=41025 (November 23, 2025).
  3. Kolmogorova S.S. Sistema sbora i analitiki ekologicheskih dannyh dlya operativnogo obnaruzheniya ekstremal’nyh znachenij s ispol’zovaniem elektroindukcionnyh datchikov (System for collecting and analysing environmental data for the rapid detection of extreme values using electro-induction sensors). Sistemy kontrolya okruzhayushchej sredy, 2025, No. 3(61), pp. 75–83. DOI 10.33075/2220-5861-2025-3-75-83. EDN VGMHCW.
  4. Kolmogorova S.S., Biryukov S.V. Use of an electro-inductive sensor for tracking moving and stationary observation objects / S. S. Kolmogorova, // Omskij nauchnyj vestnik, 2023. – No. 2(186). – Pp. 140-146. – DOI 10.25206/1813-8225-2023-186-140-146. – EDN XDGDZN.
  5. Kolmogorova S.S., Kolmogorov A.S., Baranov D.S., and Mokryak A.V. Platforma kontrolya elektromagnitnogo polya dlya obespecheniya bezopasnosti truda i promyshlennyh ob”ektov (Electromagnetic field control platform for ensuring occupational safety and industrial facilities). Bezopasnost’ truda v promyshlennosti, 2022, No. 2, pp. 58–63. DOI 10.24000/0409-2961-2022-2-58-63.
  6. Kolmogorova S.S., Baranov D.S., Kolmogorov A.S., and Biryukov S.V. Avtomatizaciya obrabotki signalov datchikov napryazhennosti elektricheskogo polya, vhodyashchih v informacionno-izmeritel’nuyu sistemu (Automation of processing signals from electric field intensity sensors included in the information and measurement system). Omskij nauchnyj vestnik, 2019, No. 4(166), pp. 66–70. DOI 10.25206/1813-8225-2019-166-66-70. EDN IBOAUG.
  7. Kolmogorova S.S., Ivanov S.A., and Pavlov V.S. Modeli prognozirovaniya dlya sistemy sbora bol’shih potokovyh dannyh s raspredelennyh elektroindukcionnyh datchikov. Nejrosetevye tekhnologii (Forecasting models for a system for collecting large streaming data from distributed electrical induction sensors. Neural network technologies). Informacionnye tekhnologii, 2025, Vol. 31, No. 7, pp. 356–363. DOI 10.17587/it.31.356-363. EDN VJFZCU.
  8. Kolmogorova S.S. and Golubyatnikova N.O. O primenenii regulyarizacii struktury bol’shih dannyh dlya raspredelennoj sistemy sbora i prognozirovaniya parametrov ob”ektov nablyudenij (On the application of big data structure regularization for a distributed system for collecting and forecasting observation object parameters). Vestnik Voronezhskogo gosudarstvennogo tekhnicheskogo universiteta, 2022, Vol. 18, No. 5, pp. 91–99. DOI 10.36622/VSTU.2022.18.5.012. EDN ZDQFUO.
  9. Polyakhov N.D. and Prikhodko I.A. Prognozirovanie elektropotrebleniya na osnove metoda opornyh vektorov (Wang Yefeng Forecasting electricity consumption based on the support vector method). Izvestiya Sankt-Peterburgskogo gosudarstvennogo elektrotekhnicheskogo universiteta, 2014, No. 10, pp. 26–30.
  10. Sergeev N.N. and Matrenin P.V. Povyshenie tochnosti prognoza elektropotrebleniya promyshlennogo predpriyatiya metodami mashinnogo obucheniya s pomoshch’yu otbora znachimyh priznakov iz vremennogo ryada (Improving the accuracy of electricity consumption forecasts for industrial enterprises using machine learning methods by selecting significant features from time series). iPolytech Journal, 2022, No. 3, pp. 487–498. DOI 10.21285/1814-3520-2022-3-487-498.
  11. Mohammed N.A. and Al-Bazi A. An Adaptive Backpropagation Algorithm for Long-Term Electricity Load Forecasting, Neural Computing and Applications, pp. 1–15, https://link.springer.com/content/pdf/10.1007/s00521-021-06384-x.pdf, 11 August 2021. DOI: 10.1007/S00521-021-06384-X.
  12. Luo Y., Cai Y., Yuanhang Q., Chen H., and Wang S. Long Short-Term Power Load Forecasting Algorithm Using Long Short-Term Memory Neural Network with Density-Based Spatial Clustering, IEEE International Conference Computer and Communications, 1 December 2019. DOI: 10.1109/ICCC47050.2019.9064382.
  13. Zhang R., Yuan M., Chen Y. F., Wang Y. P., Sun Y., and Zhao L. Long Short-Term Memory for Short Term Load Forecasting with Singular Spectrum Analysis and Whale Optimisation Algorithm, pp. 1164–70, 23 March 2023. DOI: 10.1109/AEEES56888.2023.10114086.
  14. Hu Y., Wei R., Yang Y., Li H., Huang Z., Liy Y., He C., and Lu H. Performance degradation prediction using LSTM with optimized parameters. Sensors, 2022, Vol. 22, No. 6, 2407 p. DOI 10.3390/s22062407.
  15. Xiao K., Xueqian L., Zhan X., Xiaorui Q., Chen Y., Zhiyun J., Zhang, Y., and Peng W. Data Feature Processing and Improved Informer for Medium and Long-Term Electrical Power Forecasting, Proc. 2024 IEEE 2-nd International Conference on Power Science and Technology (ICPST), May 9, 2024. DOI: 10.1109/ICPST61417.2024.10602192.
  16. Loskutov A.Yu. Analiz vremennyh ryadov. Kurs lekcij (Analysis of Time Series. Lecture Course). Moscow: Physics Department of Moscow State University (year unknown), 113 pp.
  17. Golyandina N. and Zhigljavsky A. Singular Spectrum Analysis for Time Series. Berlin: Springer, 2013, 120 p.
  18. Basseville M. and Nikiforov I.V. Detection of Abrupt Changes: Theory and Application. Englewood Cliffs, NJ: Princeton, NJ: Prentice Hall, 1993, 528 p.
  19. Box G.E.P. and Cox D.R. An analysis of transformations. Journal of the Royal Statistical Society. Series B (Methodological), 1964, Vol. 26, No. 2, pp. 211–252.
  20. Kolmogorova S.S. Konstruktivnoe reshenie postroeniya ellipsoidal’nogo elektroindukcionnogo datchika s chuvstvitel’nymi elementami v forme szhatogo sferoida dlya ocenki parametrov elektricheskogo polya (Design solution for constructing an ellipsoidal electro-inductive sensor with sensitive elements in the form of a compressed spheroid for evaluating electric field parameters). Pribory i sistemy. Upravlenie, kontrol’, diagnostika, 2025, No. 11, pp. 9–18. DOI: 10.25791/pribor.11.2025.1627.

Loading