Bakhodir Bekimbetov

Mr. Bakhodir Bekimbetov

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2026 in Processes
Verified
Journal Article
Open access
PDF file
Jasur SevinovJasur Sevinov
Jasur Sevinov
Barnokhon TemerbekovaBarnokhon Temerbekova
Barnokhon Temerbekova
GBGulnora Bekimbetova
Gulnora Bekimbetova
Ulugbek MamanazarovUlugbek Mamanazarov
Ulugbek Mamanazarov
BBBakhodir Bekimbetov
Bakhodir Bekimbetov
The study focuses on the design of an intelligent information-control system (ICS) for metallurgical production, aimed at robust forecasting of technological parameters and automatic self-adaptation under noise, anomalies, and data drift. The proposed architecture integrates a hybrid LSTM–DNN model with low-discrepancy hypercube sampling using Sobol and Halton sequences to ensure uniform coverage of operating conditions and the hyperparameter space. The processing pipeline includes preprocessing and temporal synchronization of measurements, a parameter identification module, anomaly detection and correction using an ε-threshold scheme, and a decision-making and control loop. In simulation scenarios modeling the dynamics of temperature, pressure, level, and flow (1 min sampling interval, injected anomalies, and measurement noise), the hybrid model outperformed GRU and CNN architectures: a determination coefficient of R2 > 0.92 was achieved for key indicators, MAE and RMSE improved by 7–15%, and the proportion of unreliable measurements after correction decreased to <2% (compared with 8–12% without correction). The experiments also demonstrated accelerated adaptation during regime changes. The scientific novelty lies in combining recurrent memory and deep nonlinear approximation with deterministic experimental design in the hypercube of states and hyperparameters, enabling reproducible self-adaptation of the ICS and increased noise robustness without upgrading the measurement hardware. Modern metallurgical information-control systems operate under non-stationary regimes and limited measurement reliability, which reduces the robustness of conventional forecasting and decision-support approaches. To address this issue, a hybrid LSTM–DNN architecture combined with low-discrepancy hypercube probing and anomaly-aware data correction is proposed. The proposed approach is distinguished by the integration of hybrid neural forecasting, deterministic hypercube-based adaptation, and anomaly-aware data correction within a unified information-control loop for non-stationary industrial processes.

28 April 2023 in Chernye Metally
Verified
Journal Article
Closed access
Barnokhon TemerbekovaBarnokhon Temerbekova
Barnokhon Temerbekova
Ulugbek MamanazarovUlugbek Mamanazarov
Ulugbek Mamanazarov
BBBakhodir Bekimbetov
Bakhodir Bekimbetov
ZIZh. M. Ibragimov
Zh. M. Ibragimov
The scheme of interaction of intellectual support tools in the systems of automated control and management of the structural system of intellectual control in mining and metallurgical complexesis disclosed. Practical implementation of quality management of technological parameters in complex thermal technological processes on the basis of modern cloud technologies implies a detailed knowledge base of the principles of construction and operation of the system, the specifics of using different algorithms, the features of actuators and the controlled object. The expediency and ways of solving the problem of filtering, distorted by disturbances and noise of the measuring signal on the basis of methods of the theory of fuzzy sets are substantiated. The algorithm of control of reliability of the measuring information in thermal technological processes of mining and metallurgical productions which is based on existence in control system of certain information redundancy of the data and consisting of separate interacting with each other blocks: "Preliminary control", "Transformation of equations", "Equations" and "Calculation xi " of corrected values xi = xi* + vi which satisfy to the initial equations and are most reliable" is offered.

10 July 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
BTBarnokhon Temerbekova
Barnokhon Temerbekova
BBBakhodir Bekimbetov
Bakhodir Bekimbetov
GBGulnora Bekimbetova
Gulnora Bekimbetova
Ulugbek MamanazarovUlugbek Mamanazarov
Ulugbek Mamanazarov
This paper proposes a hybrid neural network architecture combining a recurrent LSTM module and a fully connected DNN module with an integrated noise immunity criterion to improve the robustness of predictive control and decision making in dynamic systems operating under uncertainty. The key idea is to incorporate a formalized probabilistic measure of robustness directly into the loss function, allowing the model to account for stochastic disturbances, sensor failures, and parametric variations. A representative training set is generated using hypercube probing based on low-variance sequences, ensuring uniform coverage of the disturbance space with reduced computational costs. Numerical experiments on a dynamic simulation bench demonstrate improved forecast accuracy, robustness (≥0.95), and computational efficiency compared to the basic LSTM and LSTM–DNN architectures. The developed architecture can be considered as a general mechanism for robust learning and control for a wide class of process-intensive systems, including manufacturing, energy, and chemical ones.

21 September 2026 in Journal of Automation, Mobile Robotics and Intelligent Systems
Verified
Journal Article
Open access
PDF file
Barnokhon TemerbekovaBarnokhon Temerbekova
Barnokhon Temerbekova
GBGulnora Bekimbetova
Gulnora Bekimbetova
Ulugbek MamanazarovUlugbek Mamanazarov
Ulugbek Mamanazarov
BBBakhodir Bekimbetov
Bakhodir Bekimbetov
Erroneous and noisy measurements significantly distort the assessment of the stability of technological lines in industrial information and control systems. The paper proposes an algorithm for operational dispatch control of technological complexes based on the assessment of noise immunity using a hybrid LSTM–DNN model and a fuzzy set apparatus. The hybrid model predicts time dependencies and failure probabilities, while the fuzzy inference system combines objective sensor data and subjective expert assessments into a single reliability indicator. The use of trapezoidal membership functions and defuzzification by the center of gravity method ensures a smooth mapping between linguistic and numerical variables. An analysis of the monotonic dependence of the interference immunity coefficient on system performance and conditions for maintaining stability in linearly connected storage lines was carried out. Modelling has shown that the proposed neuro-fuzzy approach increases the adaptability and stability of the system to non-stationary disturbances compared to probabilistic methods. The developed model can be integrated into industrial dispatch control systems for real-time decision-making.

2 December 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
Jasur SevinovJasur Sevinov
Jasur Sevinov
Barnokhon TemerbekovaBarnokhon Temerbekova
Barnokhon Temerbekova
GBGulnora Bekimbetova
Gulnora Bekimbetova
Ulugbek MamanazarovUlugbek Mamanazarov
Ulugbek Mamanazarov
BBBakhodir Bekimbetov
Bakhodir Bekimbetov
The study focuses on the design of an intelligent information-control system (ICS) for metallurgical production, aimed at robust forecasting of technological parameters and automatic self-adaptation under noise, anomalies, and data drift. The proposed architec-ture integrates a hybrid LSTM–DNN model with low-discrepancy hypercube sampling using Sobol and Halton sequences to ensure uniform coverage of operating conditions and the hyperparameter space. The processing pipeline includes preprocessing and tem-poral synchronization of measurements, a parameter identification module, anomaly de-tection and correction using an ε-threshold scheme, and a decision-making and control loop. In simulation scenarios modeling the dynamics of temperature, pressure, level, and flow (1-min sampling interval, injected anomalies and measurement noise), the hybrid model outperformed GRU and CNN architectures: a determination coefficient of R² > 0.92 was achieved for key indicators, MAE and RMSE improved by 7–15%, and the proportion of unreliable measurements after correction decreased to < 2% (compared with 8–12% without correction). The experiments also demonstrated accelerated adaptation during regime changes. The scientific novelty lies in combining recurrent memory and deep non-linear approximation with deterministic experimental design in the hypercube of states and hyperparameters, enabling reproducible self-adaptation of the ICS and increased noise robustness without upgrading the measurement hardware. The practical signifi-cance lies in improving the reliability of measurement information and the stability of control loops; the proposed methodology is transferable to related technological systems within the Industry 4.0 domain.