DOI: 10.17587/prin.17.595-602
LSTM Network Architecture based on Programmable Logic Controllers
M. E. Shevchenko, Postgraduate Student, mike.shevchenko1@yandex.ru,
M. V. Shcherbakov, Professor, Head of the Chair, maxim.shcherbakov@vstu.ru,
Volgograd State Technical University, Volgograd, 400005, Russian Federation
Corresponding author: Mikhail E. Shevchenko, Postgraduate Student, Volgograd State Technical University, Volgograd, 400005, Russian Federation, E-mail: mike.shevchenko1@yandex.ru
Received on March 26, 2026
Accepted on June 16, 2026
The article is devoted to the problem of short-term forecasting of technological parameters at the level of a programmable logic controller (PLC). Real-time analysis of process signals is infeasible due to latency in the network infrastructure between industrial control system (ICS) levels, as well as delays in measurement channels. The paper proposes implementing an LSTM neural network model at the PLC level, which makes it possible to optimize data handling time and promptly detect deviations from normal operating modes. An implementation of the LSTM model architecture on a PLC is presented for forecasting time series of process parameters. The LSTM model, developed in the high-level Python language, was trained on 110 hours of historical data sampled at one-minute intervals. Training yielded weight coefficients and neuron bias values that were used to generate forecasts on the PLC. An LSTM model library was developed in Structured Text (ST) as defined by the IEC 61131-3 standard. For the obtained results, forecast metrics were computed — both on the upper-level server and on the PLC — against the test data, and forecasting performance indicators are reported. The results demonstrate the feasibility of implementing neural network models, in particular LSTM, at the PLC level without adversely affecting system performance.
Keywords: forecasting, neural networks, PLC, time series, IEC 61131-3, LSTM
pp. 595—602
For citation:
Shevchenko M. E., Shcherbakov M. V. LSTM Network Architecture based on Programmable Logic Controllers, Programmnaya Ingeneria, 2026, vol. 17, no. 11, pp. 595—602. DOI: 10.17587/prin.17.595-602 (in Russian).
References:
- Valencia-Palomo G., Rossiter J. A. Programmable logic controller implementation of an auto-tuned predictive control based on minimal plant information, ISA transactions, 2011, vol. 50, no. 1, pp. 92—100. DOI: 10.1016/j.isatra.2010.10.002.
- Krupa P., Limon D., Alamo T. Implementation of model predictive control in programmable logic controllers, IEEE Transactions on Control Systems Technology, 2020, vol. 29, no. 3, pp. 1117— 1130. DOI: 10.1109/TCST.2020.2992959.
- Kolodin A. A., Elshin V. V. Development and study of a controller based on a predictive model, Vestnik Samarskogo gosu-darstvennogo texnicheskogo universiteta. Seriya: Texnicheskie nauki, 2021, vol. 29, no. 1 (69), pp. 36—45. DOI: 10.14498/tech.2021.1.3 (in Russian).
- Dyakonov N. A., Logunova O. S. Process Control Systems Based on Predictive Analytics: Design, Elektrotekhnicheskie sistemy i kompleksy, 2021, no. 1 (50), pp. 58—64. DOI: 10.18503/2311-8318-2021-1(50)-58-64 (in Russian).
- Kavaliauskas Z., Blaziflnas G., Sajev I. et al. Development and Optimization of an Automated Industrial Wastewater Treatment System Using PLC and LSTM Neural Network, Applied Sciences, 2025, vol. 15, no. 16, article 8990. DOI: 10.3390/app15168990.
- Jagnade S. A., Patwardhan S. C., Kumar K. et al. Optimization of Continuous Steel Annealing Operations Using Model Predictive Control at Tata Steel, India, INFORMS Journal on Applied Analytics, 2025, vol. 55, no. 1, pp. 48—65. DOI: 10.1287/inte.2024.0183.
- Mennilli R., Mazza L., Mura A. Integrating Machine Learning for Predictive Maintenance on Resource-Constrained PLCs: A Feasibility Study, Sensors, 2025, vol. 25, no. 2, article 537. DOI: 10.3390/s25020537.
- Zhang X. Design and application of PLC control system based on neural network, Procedia Computer Science, 2025, vol. 261, pp. 1005—1011. DOI: 10.1016/j.procs.2025.04.493.
- Dolidze A. N. The use of neural networks in programmable logic controllers as an opportunity to abandon classical programming, Inzhenernyj vestnik Dona, 2025, no. 10 (130), pp. 215—229 (in Russian).
- Gafarov F. M., Galimyanov A. F. Artificial Neural Networks and Applications, Study Guide, Kazan, Kazan University Press, 2018, 121 p. (in Russian).
- Bauer H., Hoppner S., Iatrou Ch. et al. Hardware implementation of an OPC UA server for industrial field devices, IEEE Transactions on very large scale integration (VLSI) systems, 2021, vol. 29, no. 11, pp. 1998—2002. DOI: 10.1109/TVLSI.2021.3117401.
- Zelyaev A. E. Forecasting Time Series Using Recurrent Neural Networks, Matematicheskoe i komp'yuternoe modelirovanie v ekonomike, strahovanii i upravlenii riskami, sb. statej, 2025, p. 31 (in Russian).