Abstract A previously developed two-row vacuum-type seeder showed promising singulation performance but its operational productivity was limited due to its low row capacity. This study aimed to design, develop, and evaluate a four-row vacuum-type soybean seeder to improve sowing capacity while maintaining high precision. The seeder, mounted on and powered by a walking tractor, used a centrifugal suction blower to generate vacuum pressure for four seed metering devices. Seed disks were driven by a chain-and-sprocket mechanism linked to the tractor’s wheel shaft. Bench tests were conducted at two seed disk speeds (12.7 and 26.1 rpm) and six blower speeds (3500, 4000, 4500, 5000, 5500, and 6000 rpm) using two soybean varieties. Singulation efficiency and the percentages of multiple and missing seeds were recorded. Results showed that higher vacuum pressures (>1.70 kPa) at lower disk speeds (12.7 rpm) significantly improved singulation, achieving 99–100% efficiency with minimal missed and multiple seeds (<1.25%) and no seed damage. Higher disk speeds (26.1 rpm) reduced accuracy. Field tests at three blower speeds (5000–6000 rpm) and two tractor speeds (Low-1 and Low-2) confirmed consistent seed spacing (19.2–19.6 cm) and optimal planting depth (4.8–4.9 cm), with minimal variation due to wheel slip. The best field performance (98% singulation) was achieved at Low-1 tractor speed and 6000 rpm blower speed. Increasing tractor speed enhanced field capacity from 0.16 to 0.31 ha·h⁻¹. The results validate the prototype’s effectiveness for precise and efficient soybean planting.

Abstract This study presents the results of experimental tests on single flare (SF) terminations in brake tubes with diameters of ∅4.75 mm and ∅6.35 mm, and wall thickness of 0.9 mm and 1 mm, respectively, using the FALCON toolset. The research aimed at determining the optimal range of initial tube protrusion that results in a flare diameter compliant with the BN-90 3617-09 standard. The paper provides a detailed description of the testing methodology It presents raw measurement data, statistical calculations, and their analysis. Conformity or nonconformity with the specification was assessed according to the guidelines outlined in the PN-EN ISO 14253-1:2018-02 standard. For tubes with the diameter of ∅4.75 mm and the wall thickness of 0.9 mm, the average measured flare diameters for the initial protrusion in the range of 4.4–5.0 mm fall within the tolerance specified by the standard. However, when accounting for measurement uncertainty, the required flare diameters are obtained for the initial protrusion in the range of 4.6–5.0 mm. In the case of tubes with the diameter of ∅6.35 mm and the wall thickness of 1.0 mm, the average measured flare diameters meet the standard requirements for the initial protrusion of 4.6–5.0 mm, whereas, considering the measurement uncertainty, the required flare diameters are obtained for the initial protrusion in the range of 4.8–5.0 mm.

Abstract The aim of this paper was to analyze the primary and secondary order of influencing factors and to establish a BP neural network prediction model with different hydraulic performance indicators. Particle swarm optimization algorithm was used to test for the optimal hydraulic performance of the nozzle and validated by experiment. The orthogonal design covered the 200–300 kPa range, while the PSO algorithm selected 540 kPa based on the full dataset. The sprinkler was raised at a height of 1.4 m from the ground level in a square configuration. The optimal parameter combination for the square layout achieved higher uniformity and controlled maximum kinetic energy relative to the average sprinkling intensity. Results from the experiment gave 4.71 mm·h −1 , 83.28% and 0.0078 W·m −2 for average sprinkler intensity, CU and maximum kinetic energy, respectively. Compared with the rangeanalysis- based scheme, the optimized configuration reduced average irrigation intensity while improving uniformity and kinetic energy performance. The maximum error between experimental results and optimization results was 3.29%, indicating that the optimization model is feasible and reliable. This study proposes a practical optimization framework for the design and operation of sprinkler irrigation systems.

Abstract Weed control with chemicals is a challenging process that should be performed in a rational way to reduce their negative impact on the surrounding environment. The growth of artificial intelligence algorithms encourages researchers to develop smart spraying robots that detect and spray weeds and distinguish them from the main crop which leads to sustainable use of these chemicals and achieves some of the sustainable development goals. However, few studies are available to comprehensively compare different versions of YOLO algorithm to detect weed. In this research, seven versions of YOLO algorithms were evaluated for their performance to detect and spray four types of weeds, namely, Cultivated licorice ( Glycyrrhiza glabra L.), Dyer’s Croton ( Chrozophora verbascifolia ), Lambsquarters ( Chenopodium album L.), and Puncturevine ( Tribulus terrestris L.) using a locally manufactured remotely controlled spraying robot. The results showed that YOLOv6n surpassed other algorithms which achieved the highest precision (0.89), recall (0.80), F1-score (0.84), mAp@0.50 (0.86), inference speed (18.83 fps), in addition to the field indicators including true positive rate (0.83), false negative rate (0.17), false positive rate (0.19), true negative rate (0.81).

Abstract High gas levels in poultry houses, namely carbon dioxide (CO₂) and ammonia (NH₃), can cause stress, impair respiration in the birds, reduce egg production, and increase their disease susceptibility. Therefore, this study presents an IoT-based monitoring system built on an Arduino ESP32 microcontroller and integrated with DHT22, MQ-135, and MQ-137 sensors for measuring key environmental parameters in poultry houses. The system monitors temperature, humidity, NH 3 , and CO 2 in both closed and open laying hen housing systems. Environmental data were stored in the cloud to enable remote, real-time monitoring and subsequent analysis. The results showed that closed housing systems are more suitable for regulating temperature and humidity while keeping gas levels within permissible limits inside poultry farms, thus leading to improved egg production. The proposed IoT-based monitoring system offers a practical tool for improving poultry management and reducing losses associated with poor air quality, ultimately helping farmers enhance production efficiency.

The accuracy and reliability of measurement results for controlled parameters are essential for ensuring the quality of engine repairs. However, current standards, such as GOST 10448-2014 (“Piston Internal Combustion Engines”) and GOST 14846-2020 (“Automobile Engines: Bench Test Methods”), do not specify requirements for the permissible error in the indirect measurement of engine power. This study aims to develop metrological recommendations to achieve specified levels of reliability for measurement data obtained during the acceptance testing of repaired engines. Using probability theory, the authors derived a formula to calculate power measurement errors and established permissible error limits for various ZMZ engine models. A comparative analysis was performed to compare these calculated limits with the actual errors observed during running-in tests on GOSNITI test benches (models KI-5274, KI-5540М, KI-5541М, KI-542М, and KI-2118А). The results indicate that the actual measurement errors remain within the calculated permissible range (derived from the GOST-based methodology). To enhance the reliability of power control, the authors developed recommendations for calculating acceptance limits with an offset relative to the nominal power value. Permissible error margins and acceptance limits were determined for various ZMZ engine models based on the required reliability levels. The application of this methodology will improve measurement quality and the overall reliability of control results during the acceptance testing of repaired engines.

Enhancing blade durability and determining optimal sharpening angles are critical for extending the service life and increasing the efficiency of feed choppers and industrial processing machinery. This study aims to determine the optimal sharpening angle for feed chopper blades subjected to through-hardening (volumetric quenching) followed by low-temperature tempering under abrasive wear conditions. Experimental samples were fabricated from U7 steel (GOST 1435-99) with sharpening angles of 10°, 20°, and 30°. Heat treatment produced a uniform fine-needle martensite structure across the entire cross-section. The resulting hardness was 740 HV, a 2.96-fold increase over the initial microhardness of 250 HV. Microstructural analysis revealed decarburizationinduced iron oxides at depths of up to 0.079 mm; consequently, suggesting that controlled-atmosphere furnaces should be used during quenching. Wear resistance was tested over a 100-hour period using a custom-made laboratory installation with quartz sand as the abrasive medium. The initial cutting-edge radius ranged from 20 to 35 μm. Under abrasive wear, the smallest increase in edge radius (blunting) was observed in the 10° blades (120 μm), while the 30° blades showed the greatest increase (185 μm). Conversely, the 30° sharpening angle exhibited the lowest overall width wear, while the 10° angle showed the highest. For cutting root crops without impact loads, the 10° sharpening angle is optimal. Furthermore, increasing the cutting speed from 5 to 7 m/s accelerated the dulling rate by 81.5%. These findings provide a technical basis for the design and maintenance of high-efficiency feed choppers and processing equipment

The evolving energy landscape, characterized by decarbonization, the integration of renewable energy sources (RES), and the growing demand for grid reliability and energy efficiency, necessitates a profound transformation of power grids, particularly in rural areas. Conventional methods of power grid management often prove insufficient for these challenges, driving the adoption of artificial intelligence (AI) solutions. This study provides a systematic analysis of key AI applications in power grids, examining the algorithms employed and practical implementation examples from both international and domestic contexts. Drawing on a comprehensive review of literature, the authors have identified four primary application areas: load forecasting, power system and grid optimization, fault detection and equipment monitoring, and optimal resource management within power grids. Within these domains, effective algorithms include deep learning techniques such as LSTM, GRU, and CNN, along with machine learning models like SVM and various metaheuristic methods. Practical examples highlight the diverse deployment of AI, adapting to national power system specificities. For instance, countries with a high share of renewable energy sources (RES) often prioritize AI for load forecasting, while in Russia, the focus is on automating the monitoring and diagnostics of extensive rural grids through computer vision and UAVs. AI is instrumental in the design of Smart Grids, enabling the digital transformation of power infrastructure to enhance efficiency, resilience, and adaptability. However, successful AI integration requires addressing challenges related to reliability, cybersecurity, and the explainability of automation-driven decision-making.

Cercospora leaf spot (CLS) of sugar beet, caused by Cercospora beticola Sacc., is a highly destructive plant disease that can reduce yields by up to 40% and significantly impair root crop quality. This study aimed to develop and validate a quantitative disease assessment method utilizing UAV-based multispectral imaging and semantic segmentation. Field trials were conducted in 2023-2024 on commercial sugar beet crops (Agrofirma Start OOO, Buzdyak District, Republic of Bashkortostan). Plots, measuring 20 × 6 rows (≈21.6 m²), included both control and artificially inoculated treatments. UAV imagery was acquired using a Geoscan ChatGPT equipped with a Pollux multispectral camera (Blue, Green, Red, Red-edge, NIR) at an altitude of approximately 30 m. A U-Net model was trained on 420 annotated image tiles (512 × 512 px), using a 6:2:2 split for training, validation, and testing. The model incorporated spectral indices (NDVI, NDRE, MCARI, NSVDI) in addition to geometric features derived from the normal vectors of a Digital Surface Model (DSM). The developed integrative algorithm achieved an overall multiclass classification accuracy of 88.6%. Specifically, an F1-score of 46.0% was obtained for the ‘infected plants’ class, outperforming Partial Least Squares Discriminant Analysis (PLS-DA) by 18.6 percentage points. F1-scores reached 92.5% for ‘Healthy vegetation’ and 68.7% for ‘Soil/Background.’ This methodology confirms the strong applicability of U-Net for the diagnostic segmentation of Cercospora outbreaks, significantly enhancing the objectivity of crop monitoring. The integration of spectral and geometric features proved crucial in improving the detection of weakly expressed disease symptoms.

The actual service life of tractor engines of traction class 1.4 in the Tomsk region falls significantly short of the warranty period and exhibits high dispersion. Statistical analysis reveals that the mean operating time until the first overhaul does not exceed 7,000 engine hours, with a standard deviation of 1,707 hours and a coefficient of variation of 0.24. The service life of new engines until the first overhaul varies by more than a factor of 2.8. To address the challenge of predicting failure modes based on cumulative operating time, this study employs artificial neural networks (ANNs). The research objective was to train an ANN to identify the most likely cause of engine failures using durability data collected under routine operating conditions of traction class 1.4 tractor engines. The authors developed an intelligent failure diagnostics system using Python and the PyTorch framework. The Matplotlib module was used for visualization, NumPy for matrix operations, and sklearn for input data normalization. The ANN uses a fully connected (dense) architecture consisting of an input layer (one neuron), a hidden layer (10 neurons), and an output layer (four neurons). The model was trained on a dataset from 25 Minsk Motor Plant engines (type 4Ch(N) 11/12.5). Based on the “operating time” input parameter, the model generates a probability distribution across four failure categories: the crank mechanism, the lubrication system, the fuel system, and the cooling system. Initial testing yielded a prediction accuracy of 60%. Future research will focus on fine-tuning the artificial neural network by expanding the training dataset to achieve a target accuracy of 80%.