A leading cause of accidents during the landing phase of a flight lies in a considerable altitude loss by an aircraft as a result of the impact of a microburst of wind. One of the significant factors focuses primarily on the need to simultaneously satisfy various requirements regarding conditions of environmental disturbances and a wide range of systemic changes. The paper presents an algorithm for synthesizing an optimal controller that solves the mixed H2/H∞ control problem for the stabilization of aircraft in glide-path landing mode in the presence of uncertainty. Firstly, the principles of multi-criteria optimization are presented, and the mixed H2/H∞ problem is interpreted as the synthesis of a system with optimal quadratic performance, subject to its readiness to operate with the worst disturbance. Then, the ensuing section expounds upon the mathematical depiction of the vertical trajectory of aircraft, duly considering the perturbations imposed by wind phenomena. Subsequently, the effectiveness of mixed H2/H∞ control is confirmed compared to autonomous H2 or H∞ regulators through simulation outcomes acquired from the created system. Optimization based on a hybrid (mixed) criterion allowed combining the strengths of locally optimal systems based only on H2 or H∞ theory.

One of the most dangerous parts of the flight is the landing phase, as most accidents occur at this stage. In order to reduce the effect of the low-level wind shear on the longitudinal motion of the aircraft in the glide path landing mode (task) a robust H- control is proposed. Dynamic models of the plane and wind shear are built. H2 and H∞ synthesis methods are investigated for the task of aircraft flight control in a vertical plane during landing under conditions of undefined disturbances. Both control methods allow to reduce height deviation significantly. However, suboptimal control H∞ provides better quality of transition processes both in height and speed than optimal control H2. The results of simulation of the synthesized system confirm the effectiveness of H∞ - control for increasing robust stability to uncertainties caused by wind disturbances.

A strategy for the design of robust control of tracking electromechanical systems based on 𝐻∞ synthesis is proposed. Proposed methods are based on the operations on frequency characteristics of control systems designed and developed using the MATLAB robust control toolbox. Determination of the singular values for a transfer matrix of the control system reduces the disturbances and guarantees its stability margin. For selecting the weighted transfer functions, the basic recommendations are formulated. The efficiency of the proposed approach is verified by robust control of an elastically coupled two-mass system whose parameter values are adjusted by matching them with the parameters of one of the supplied robots. The simulation results confirm that the proposed strategy of design of robust control of two-mass elastic coupling system using the 𝐻∞ synthesis is very efficient and significantly reduces the perturbation of parameters of the controlled plant.

The detection of anomalies and the prediction of corrosion defects in oil and gas pipelines constitute critical tasks for ensuring industrial safety and improving operational reliability. This study addresses the problem of regression-based prediction of corrosion defect levels (CR—corrosion defect) using operational process parameters. Machine learning methods, including Decision Tree, Random Forest, LightGBM, and CatBoost, were employed to develop predictive models. Data preprocessing was performed, including feature standardization and hyperparameter tuning using KFold cross-validation. Model performance was primarily evaluated using the Root Mean Square Error (RMSE) on both training and test datasets, as this metric is more sensitive to large prediction errors, which is particularly important in the context of corrosion defect analysis. Additionally, Mean Squared Error (MSE), Mean Absolute Error (MAE), and the coefficient of determination (R²) were used to provide a comprehensive assessment of model accuracy and robustness. Experimental results demonstrate that the CatBoost model achieved the best performance, yielding the lowest RMSE on the test dataset (0.02040) with a close value on the training dataset (0.01682), indicating strong generalization capability. Furthermore, this model outperformed the others in terms of MSE, MAE, and R² on the test dataset (MSE = 0.000418, MAE = 0.006319, R² = 0.695544). The obtained results confirm the effectiveness of ensemble methods and gradient boosting algorithms for regression modeling of corrosion defect development processes in pipeline systems.

Pipelines are widely used for the transportation of water, oil, gas, and other liquid and gaseous media. The occurrence of leaks in pipeline systems leads to significant losses of natural resources and may pose a threat to the environment and public safety. In this regard, the problem of timely leak detection and localization remains one of the most important challenges in the oil and gas industry. This paper addresses the problem of determining the leak location in a pipeline based on pressure measurement data using artificial intelligence methods. An approach to leak localization is proposed based on approximating the relationship between pressure drop, leak flow rate, and the distance to the leak location using a radial basis function neural network implemented in the MATLAB environment. Experimental data were obtained during testing of a leak detection system under simulated leak conditions with different operating modes. The simulation results demonstrate that the developed model provides an effective approximation of the studied relationship and allows the leak location to be determined with a root-mean-square error of approximately 1.9 km for medium and large leaks, confirming the potential of neural network-based methods for pipeline monitoring applications.

Фаза посадки во время полета, вероятно, является самой опасной частью, так как большинство аварий происходит на этой стадии. Для уменьшения влияния низковысотного сдвига ветра на продольное движение самолета в режиме посадки по глиссаде (заданию) предлагается робастное – управление. Построены динамические модели самолета и сдвиг ветра. Исследуются и методы синтеза для задачи управления полетом самолета в вертикальной плоскости при посадке в условиях неопределенных возмущений. Оба управления позволяют значительно уменьшить отклонение высоты. Однако, – субоптимальное управление обеспечивает лучшее качество переходных процессов как по высоте, так и по скорости, чем – оптимальное управление. Результаты моделирования синтезируемой системы подтверждают эффективность – управления для повышения робастной устойчивости к неопределенностям, вызванным возмущениями ветра.