DOI 10.17586/0021-3454-2024-67-2-200-204
UDC 004.8
HYBRID NEURAL NETWORK MODELS FOR MONITORING TIME SERIES DATA OF COMPLEX OBJECTS
The Joint Institute for Informatics Problems of the National Academy of Sciences of Belarus, Laboratory of Information Security Problems; Leading Researcher;
B. V. Sokolov
St. Petersburg Institute for Informatics and Automation of Russian Academy of Sciences; Deputy Director for R&D; Professor
W. Zhang
College of Information Engineering of Zhejiang University of Technology, Dean of the College and Director of International Cooperation Department ;
M. Fu
College of Information Engineering of Zhejiang University of Technology, Deputy Director of International Cooperation Department;
Reference for citation: Skobtsov V. Yu., Sokolov B. V., Zhang W.-A., Fu M. Hybrid neural network models for monitoring time series data of complex objects. Journal of Instrument Engineering. 2024. Vol. 67, N 2. P. 200—204 (in Russian). DOI: 10.17586/0021-3454-2024-67-2-200-204.
Abstract. The problem of monitoring the state of complex objects of various natures based on classification and regression analysis of time series data is considered. Hybrid neural network models of classification and regression analysis are developed and studied using data on the functioning of three types of systems: spacecraft, information system and economic system, presented in the form of time series. For all types of systems, the proposed hybrid models demonstrate an advantage in accuracy. A genetic algorithm is developed for the automatic search of hybrid neural network models, with the help of which models of varying complexity are generated with an accuracy no lower than for models developed manually. As a result of the search, it is noted that the generated hybrid neural networks show results close to the maximum value of the fitness function. The fact is considered as experimental confirmation of the constructed solution to be close to optimal for certain search parameters.
Abstract. The problem of monitoring the state of complex objects of various natures based on classification and regression analysis of time series data is considered. Hybrid neural network models of classification and regression analysis are developed and studied using data on the functioning of three types of systems: spacecraft, information system and economic system, presented in the form of time series. For all types of systems, the proposed hybrid models demonstrate an advantage in accuracy. A genetic algorithm is developed for the automatic search of hybrid neural network models, with the help of which models of varying complexity are generated with an accuracy no lower than for models developed manually. As a result of the search, it is noted that the generated hybrid neural networks show results close to the maximum value of the fitness function. The fact is considered as experimental confirmation of the constructed solution to be close to optimal for certain search parameters.
Keywords: monitoring of complex objects, hybrid neural networks, classification, regression, time series data
Acknowledgement: Research carried out on this topic was carried out with partial financial support from the budget topic FFZF-2022-0004.
References:
Acknowledgement: Research carried out on this topic was carried out with partial financial support from the budget topic FFZF-2022-0004.
References:
- Okhtilev M.Yu., Mustafin N.A., Miller V.E., Sokolov B.V. Journal of Instrument Engineering, 2014, no. 11(57), pp. 7–15. (in Russ.)
- Chen H., Zhang Zh. 2020 IEEE International Instrumentation and Measurement Technology Conference (I2MTC): Proceedings of the International Conference. May 25–28, 2020, DOI: 10.1109/I2MTC43012.2020.9129183.
- Zhao N. IEEE Access, 2021, vol. 9, рр. 15561–15569, DOI: 10.1109/ACCESS.2021.3052937.
- Skobtsov V.Yu. Applied Aspects of Information Technology, 2021, no. 4(4), pp. 299–310.
- Skobtsov V.Yu., Sokolov B.V. Bulletin of the VSU. System analysis and information technology, 2022, no. 3, pp. 99—114. (in Russ.)
- Skobtsov V.Y., Stasiuk A. Artificial Intelligence Application in Networks and Systems. CSOC 2023. Lecture Notes in Networks and Systems, 2023, vol. 724, pp. 800–811.
- Kaiming H. et al. Conference on Computer Vision and Pattern Recognition, 2015, https://arxiv.org/abs/1512.03385.
