NEURAL NETWORK SIGNAL PROCESSING: THE TASKS WITHOUT “DEEP LEARNING”

DOI: 10.47026/1810-1909-2023-2-151-160

УДК 621.316.91

ББК 31.247

Leonid A. SLAVUTSKII, Elena V. SLAVUTSKAYA

Key words

artificial neural networks, direct propagation, signal processing in electrical engineering, approximation, error in determining parameters, transients

Abstract

The purpose of the study is to evaluate the possibility of using artificial neural networks for electrical signals processing.

Methods. The application of a multilayer perceptron for these purposes as the simplest neural feed forward network is considered. Its difference as the basis of neural network algorithms is that when analyzing dynamic processes, signal processing should be carried out in a “sliding time window”.

Results. It is shown that neural network processing makes it possible to approximate the shape of the signal with high accuracy and determine its parameters in real time. Using the example of periodic signals and transients in electrical circuits, accuracy assessments are made and the features of neural network processing are analyzed. The necessary sizes of the training sample of signals and the level of errors that occur when testing a neural network are discussed. Estimates of the required signal sampling frequency, the “sliding window” duration and the variation range of signal parameters when creating a training sample are given.

Conclusions. It is shown that the proposed approach does not require “deep learning” of neural networks with complex architecture. It enables to create a signals training sample based on simple analytical formulas and to control the neural network algorithm quality at intermediate stages of calculations.

References

1. Andreev O.N., Slavutskiy A.L., Alekseev V.V. Strukturnyy analiz elektrotekhnicheskikh signalov pri rekurrentnom ispol’zovanii mnogosloynogo perseptrona [Electrical signals’ structural analysis with recurrent use of a multilayer perceptron]. Elektrotekhnika, 2022, no 8, pp. 41–44.
2. Kruglov V.V., Borisov V.V. Iskusstvennye neironnye seti. Teoriya i praktika [Neural networks. Theory and practice]. Moscow, Goryachaya liniya Telekom Publ., 2001, 382 p.
3. Minskiy M., Peypert S. Perseptrony [Perceptrons]. Moscow, Mir Publ., 1971, 261 p.
4. Gorban’ A.N., Dunin-Barkovskiy V.L., Kirdin A.N. et al. Neyroinformatika [Neuroinformatics]. Novosibirsk, Nauka Publ., 1998, 296 p.
5. Slavutsky A.L., Slavutsky L.A., Alekseev V.V. et al. Neyrosetevoy algoritm vosstanovleniya v real’nom vremeni signala promyshlennoy chastoty pri nelineynykh iskazheniyakh [A Neural-Network Algorithm for Real-Time Recovery of an Industrial-Frequency Signal Upon Nonlinear Distortions]. Russian Electrical Engineering, 2021, vol. 92, no. 8, pp. 429–432. DOI: 10.3103/S1068371221080113.
6. Petrov A.P. O vozmozhnostyakh pertseptrona [On the capabilities of the perceptron]. Izvestiya AN SSSR. Tekhnicheskaya kibernetika, 1964, no. 6, pp. 25–57.
7. Slavutskaya E., Slavutskiy L. O vybore struktury iskusstvennykh neyrosetey i algoritmov analiza psikhodiagnosticheskikh dannykh [On choosing the artificial neural networks structure and the algorithms for psycho diagnostic data analyzing]. Kazan pedagogical journal, 2020, no. 5(142), рp. 202–210. DOI: 34772/KPJ.2020.142.5.026.
8. Haykin S. Neural Networks. New Jersey, 1999 (Russ. ed.: Neyronnyye seti: polnyy kurs. 2-ye izd. Moscow, Vil’yams Publ., 2006, 1104 p.).
9. Andreev O.N., Slavutskiy A.L., Slavutskii A. Neural network in a sliding window for power grids signals structural analysis. IOP Conference Series: Earth and Environmental Science, 2022, vol. 990. DOI: 10.1088/1755-1315/990/1/012054.
10. Antonov V.I., Naumov V.A., Fomin A.I. et al. Adaptive structural analysis of input signals of digital and relay protection and automation. Russian Electrical Engineering, 2015, vol. 86, pp. 391–397. DOI: 10.3103/S1068371215070032.
11. Basodi, Zhang H., Pan Y. Gradient amplification: An efficient way to train deep neural networks. Big Data Mining and Analytics, 2020, vol. 3(3), pp. 196–207.
12. Billings S.A. Nonlinear system identification: NARMAX methods in the time, frequency, and spatio-temporal domains. John Wiley & Sons, 2013, 555 p. DOI 10.1002/9781118535561.
13. Bhattacharya B., Sinha А. Intelligent Fault Analysis in Electrical Power Grids. In: IEEE 29^th International Conference on Tools with Artificial Intelligence (ICTAI). Boston, 2017, pp. 985–990. DOI: 10.1109/ICTAI.2017.00151.
14. Burton B., Harley R.G. Reducing the computational demands of continually online-trained artificial neural networks for system identification and control of fast processes. IEEE Transactions on Industry Applications, 1998, vol. 34(3), pp. 589–596.
15. Bychkov A., Slavutskii L., Slavutskaya E. Neural Network for Pulsed Ultrasonic Vibration Control of Electrical Equipment. International Ural Conference on Electrical Power Engineering (UralCon), 2020, pp. 24–28. DOI: 10.1109/UralCon49858.2020.9216248.
16. Coury D.V., Oleskovicz M., Aggarwal R.K. An ANN routine for fault detection, classification and location in transmission lines. Electrical Power Components and Systems, 2002, no. 30, pp. 1137–1149.
17. Dharmendra K., Moushmi K., Zadgaonkar A.S. Analysis of generated harmonics due to transformer load on power system using artificial neural network. International journal of electrical engineering, 2013, vol. 4(1), pp. 81–90.
18. Grossberg S. A Path Toward Explainable AI and Autonomous Adaptive Intelligence: Deep Learning, Adaptive Resonance, and Models of Perception, Emotion, and Action. Neurorobot, 2020, vol. 14. DOI: https://doi.org/10.3389/fnbot.2020.00036.
19. Ince T., Kiranyaz S., Eren L., Askar M., Gabbou M. Real-Time Motor Fault Detection by 1-D Convolutional Neural Networks. IEEE Transactions on Industrial Electronics, 2016, vol. 63, vol. 11, pp. 7067–7075. DOI: 10.1109/TIE.2016.2582729.
20. Ivanov S.O., Nikandrov M., Lariukhin Neuro Algorithm Accuracy Evaluation for The Anomalies Detecting in Overcurrent Protection Operation. International Ural Conference on Electrical Power Engineering (UralCon), 2021, pp. 116–120. DOI: 10.1109/UralCon52005.2021.9559614.
21. He Z., Lin S., Deng Y., Li X., Qian Q. A rough membership neural network approach for fault classification in transmission lines. International Journal of Electrical Power and Energy Systems, 2014, vol. 61, pp. 429–439.
22. Jain A., Thoke A.S., Patel R.N. Fault classification of double circuit transmission line using artificial neural network. International Journal of Electrical Systems Science and Engineering, 2008, vol. 1, pp. 230–235.
23. Jamil M., Kalam A., Ansari A.Q., Rizwan M. Generalized neural network and wavelet transform based approach for fault location estimation of a transmission line. Applied Soft Computing, 2014, vol. 19, pp. 322–332.
24. Keerthipala W.L., Low Tah Chong, Tham Chong Leong. Artificial neural network model for analysis of power system harmonics. IEEE International Conference on Neural Networks, 1995, vol. 2, pp. 905–910.
25. Kozma R., Ilin R., Siegelmann H. T. Evolution of Abstraction Across Layers in Deep Learning Neural Networks. Computer Science, 2018, vol. 144, pp. 203–213. DOI: 10.1016/j.procs.2018.10.520.
26. Kulikov A.L., Loskutov A.A., Mitrovic M. Improvement of the technical excellence of multiparameter relay protection by combining the signals of the measuring fault detectors using artificial intelligence methods. International Scientific and Technical Conference Smart Energy Systems (SES-2019), 2019, vol. 124. DOI: 1051/e3sconf/201912401039.
27. Leroux S., Verbelen T., Simoens P. et al. Iterative neural networks for adaptive inference on resource-constrained devices. Neural Comput & Applic, 2022, 34, pp. 10321–10336. DOI: 10.1007/s00521-022-06910-5.
28. Leroux S., Bohez S., Verbelen T. et al. Resource-constrained classification using a cascade of neural network layers. International Joint Conference on Neural Networks (IJCNN), 2015, pp. 1–7, DOI: 10.1109/IJCNN.2015.7280601.
29. Niekerk C.R., Rens A.P.J., Hoffman A.J. Identification of types of distortion sources in power systems by applying neural networks. 6^th IEEE AFRICON, 2002, vol. 2(2), pp. 829–834.
30. Oh J.W., Jeong J. Convolutional neural network and 2-D image based fault diagnosis of bearing without retraining. 3^rd International Conference (ICDA 2019), 2019, pp. 134–138. DOI: 1145/3314545.3314563.
31. Osowski S. Neural network for estimation of harmonic components in a power system. IEEЕ Proceedings on Generation, Transmission and Distribution, 1992, vol. 139(2), pp. 129–135.
32. Rosenblatt F. Principles of neurodymamics. Washington, Spartan books, 1962.
33. Rumelhart D.E., Hinton G.E., Williams R.J. Learning Internal Representations by Error Propagation. Parallel Distributed Processing. Cambridge, MA-MIT Press, 1986, vol. 1, pp. 318–362.
34. SamarasingheNeural Networks for Applied Sciences and Engineering: From Fundamentals to Complex Pattern Recognition. 1^st ed. Boca Raton, Auerbach Publications, 2006, 570 p.
35. Slavutskiy A., Slavutskii L., Slavutskaya E. Neural Network for Real-Time Signal Processing: the Nonlinear Distortions Filtering. International Ural Conference on Electrical Power Engineering (UralCon), 2021, pp. 84–88. DOI: 10.1109/UralCon52005.2021.9559619.
36. Schmidhuber J. Deep learning in neural networks: An overview. Neural networks, 2015, 61, pp. 85–117. DOI:10.48550/arXiv.1404.7828.

Information about the authors

Leonid A. Slavutskii – Doctor of Physical and Mathematical Sciences, Professor, Department of Automation and Control in Technical Systems, Chuvash State University, Russia, Cheboksary (lenya@slavutskii.ru; ORCID: https://orcid.org/0000-0001-6783-2985).

Elena V. Slavutskaya – Doctor of Psychological Sciences, Professor, Department of Psychology and Social Pedagogy, I.Ya. Yakovlev Chuvash State Pedagogical University, Russia, Cheboksary (slavutskayaev@gmail.com; ORCID: https://orcid.org/0000-0002-3759-6288).

For citations

Slavutskii L.A., Slavutskaya E.V. NEURAL NETWORK SIGNAL PROCESSING: THE TASKS WITHOUT “DEEP LEARNING”. Vestnik Chuvashskogo universiteta, 2023, no. 2, pp. 151–160. DOI: 10.47026/1810-1909-2023-2-151-160 (in Russian).

Download the full article