Energy-efficient stockyard scheduling in dry bulk ports with belt conveyor sharing

Distributed acoustic sensing (DAS) has attracted considerable attention across various fields, and artificial intelligence (AI) technology plays a vital role in DAS applications for event recognition and denoising. Existing AI models require real-world data (RWD), whether labeled or not, for training, which is contradictory to the reality of limited available event data in practical scenarios. Here, a physics-informed DAS neural network paradigm is proposed, which eliminates the need for real-world event data during training. By physically modeling the target events along with real-world and DAS system constraints, physical functions are derived to train a generative network for the synthesis of DAS event data. A DAS noise-removal network is then trained using the generated data to effectively eliminate background noise in DAS measurements. The effectiveness of the proposed paradigm is demonstrated in two applications: event identification based on a public DAS spatiotemporal dataset, and belt conveyor fault monitoring based on DAS time-frequency data. In both cases, the paradigm achieves comparable or superior performance to data-driven networks trained with RWD. Owing to the incorporation of physical information and the ability to remove background noise, the proposed approach shows strong generalization capability across different sites within the same application. Notably, a fault diagnosis accuracy of 91.8% is achieved in a real belt conveyor field using networks transferred from a simulation test site, without using any fault event data from the target field during training. The proposed paradigm offers a potential solution to the critical challenges of limited data availability and intense noise in practical DAS applications.

Belt conveyor idlers frequently fail under high-load harsh conditions, causing system shutdowns. Existing deep learning-based fault diagnosis methods suffer from insufficient frequency resolution and poor dynamic adaptability. To address this, this paper proposes a fault diagnosis framework based on adaptive frequency-band KAN: First, an adaptive frequency-band Mel filter bank designed based on fault mechanisms enhances resolution in critical fault frequency bands through non-uniform frequency-axis remapping and third-order peak detection. Second, a temporal convolutional network is integrated to expand the receptive field and capture cross-period features. A Kolmogorov-Arnold Networks (KAN) layer is introduced to dynamically analyze nonlinear coupling relationships in the frequency domain using learnable B-spline basis functions. This model achieves synergistic optimization of feature resolution enhancement and dynamic modeling, significantly improving diagnostic accuracy and cross-condition generalization capability for roller faults. Under actual conveyor roller operating conditions, fault prediction accuracy reaches 81.25%, fully validating the model's adaptability to real-world industrial scenarios.

Influence of composite impact loads on the failure behavior of bearings in carriage belt conveyors

Speed synchronization control of logistics belt conveyor based on feedforward fuzzy control algorithm

To address coal dust emissions at the feeding and discharging ports of belt conveyor transfer points, this study proposes a pressure-enhanced dust suppression system based on a supersonic spray fortress. A coupled numerical and physical model of the transfer point was established, incorporating assumptions of coal flow–induced airflow, tangential distribution of induced effects, and incompressible fluid behavior. The multiphase gas–particle–spray flow was characterized using the Stokes equation for particle motion and the standard k–ε turbulence model. The supersonic spray fortress was designed as an irregular hexahedral metal structure integrating gas–liquid pipelines and equipped with three supersonic atomizing nozzles arranged at composite angles (45° horizontal spacing and 45° downward inclination). A geometric model of the coal conveying system was constructed, and mesh independence verification was conducted to ensure the reliability of numerical simulations. Spray characteristics were experimentally investigated using a Winner319 laser particle size analyzer and a three-dimensional particle image velocimetry (3D-PIV) system, enabling measurement of droplet size parameters (SMD, V50, N50) and velocity under pressures ranging from 0.2 to 0.4 MPa. Field experiments were performed at a coal mine transfer point to monitor dust concentrations before and after system implementation. The results indicate that spray characteristics are highly sensitive to operating pressure. As pressure increases from 0.2 to 0.4 MPa, droplet SMD decreases from 65–70 μm to 30–45 μm, while initial velocity increases from 12 to 20 m/s; droplet size shows minimal variation with distance. The spray fortress significantly regulates the airflow field, reducing peak airflow velocity by more than 60% (from >10 to ≤4.8 m/s) and decreasing the proportion of high-velocity regions (>5 m/s) from 35% to ≤8%. In terms of dust behavior, peak dust velocity decreases by over 70% (from >20 to ≤6 m/s), accompanied by a substantial reduction in coarse particles (>10 μm). Field application results show that within a range of −10 to 15 m, the total dust removal efficiency reaches 51.4–93.4%, while respirable dust removal efficiency is 44.6–87.9%. The dust particle size distribution shifts toward finer fractions (0–2.5 μm), primarily due to the effective capture of medium and coarse particles by spray droplets. The supersonic spray fortress achieves synergistic regulation of airflow and dust transport through mechanisms such as momentum offset, spatial coverage, and collision capture. With optimized spray parameter configuration and strategic placement at the upper and lower ports of the chute, a full-process dust control system is established, providing a reliable and effective solution for dust mitigation at belt conveyor transfer points.

Abstract The aim of the study is to investigate the performance of overburden mining and haulage complexes and its impact on the cost of overburden removal under conditions of horizontal deposits with a large overburden thickness. Analytical, statistical, and graphical-analytical research methods were applied. Dependencies were established for the costs of overburden excavation and transportation using cyclic mining and haulage complexes, including: open-pit excavators (mechanical shovels) with dump trucks, walking dragline excavators with dump trucks, as well as continuous mining systems with bucket-wheel excavators and conveyor transport. It was determined that in modern horizontal open pits for titanium-zirconium ore extraction with considerable overburden thickness, the most efficient mining and haulage equipment for overburden removal is the bucket-wheel excavator complex with a belt conveyor. Based on the calculation of minimum costs for excavation and transportation of overburden, it was found that the costs of overburden removal when using bucket-wheel complexes are on average 40% lower, with productivity exceeding that of cyclic complexes by a factor of 2-4. The conducted research made it possible to establish the dependencies of the unit cost of removing 1 m 3 of overburden and transportation costs on the productivity of overburden mining and haulage complexes.

Abstract Belt conveyors are widely utilized in material handling applications because of their high efficiency and substantial capacity. However, belt conveyor deviation typically leads to transmission system failures, which in turn impact production efficiency and may even cause serious safety incidents. Traditional sensor-based detection methods are sensitive to environmental noise, lighting changes, and material interference. Deep learning approaches face challenges such as high computational complexity, limited edge localization accuracy, and baseline drift due to camera position shifts. This paper proposes a deviation detection method that integrates deep learning with geometric constraints. The DC-YOLO-seg model, based on enhanced YOLOv11-seg, is combined with deformable convnets v3 (DCNv3) and coordinate attention (CA) mechanisms for high-precision instance segmentation of conveyor belts and rollers. Subsequently, the belt centerline was extracted using the Canny edge detection algorithm and random sample consensus (RANSAC) fitting method. The drive centerline was estimated based on the alignment of the roller centers, thereby quantifying the offset distance. This approach effectively reduces dependency on camera position and minimizes environmental interference. Experimental results on a self-constructed dataset demonstrate that DC-YOLO-seg achieves a Mask (mAP50-95) of 0.948, representing a 2.5% improvement over baseline models. The deviation detection error is generally maintained within 5 mm. This research provides a robust solution for intelligent operation and maintenance, establishing a foundation for cross-scenario generalization and real-time deployment.

Aiming at the severe floor heave occurring at the machine head section of the belt conveyor downhill roadway in the 1105 mining area of Banji Coal Mine, Bozhou, Anhui, a systematic investigation was conducted using mechanism analysis, numerical simulation, and on-site industrial testing to elucidate the multifactorial mechanism and control technique of floor heave in deep soft rock roadways. The dominant factors contributing to roadway floor heave were identified as high horizontal tectonic stress, weak floor lithology, high water content in the floor strata, and deficiencies in the original support system. Subsequently, the loosening zone of the roadway floor was detected, and the development characteristics of floor plastic deformation and deformation concentration were analyzed in detail. On this basis, the reasonable effective depth for grouting reinforcement was determined. Accordingly, an integrated floor heave control technique combining deep-hole high-pressure grouting and drainage was proposed to reinforce and strengthen the roadway structure. A numerical model was then established to comparatively analyze roadway deformation over a three-month period under conditions with and without reinforcement. The results indicated that the overall stress state of the surrounding rock was significantly improved after reinforcement, leading to effective control of floor heave deformation, while deformation of the roof and sidewalls was simultaneously mitigated. Furthermore, on-site monitoring results demonstrated that 30 d after grouting, floor subsidence, inwardconvergence of the two sides, and floor heave of the roadway all reduced quickly, and the deformation gradually stabilized. These findings confirm that the proposed floor heave control technique can effectively suppress deformation of the roadway surrounding rock and ensure safe and stable operation.

The sorting of coal gangue is of great significance for improving coal quality, avoiding environmental pollution, and reducing labor costs. The image-based coal gangue sorting method has been proposed by a large number of researchers, but the complexity of the environment, the speed and accuracy of coal gangue detection and recognition methods, and the performance of hardware equipment all pose challenges to the accuracy of coal gangue sorting. This paper discusses the research and application of deep-learning methods in the field of coal gangue detection and proposes an improved YOLOv7 coal gangue detection model for ordinary GPU devices with large computing power and memory. In response to the feature redundancy problem of the YOLOv7 model in coal gangue detection tasks, FasterNet was introduced to improve the backbone network of YOLOv7, reducing redundant calculations and memory access, making the model more effective in extracting features. In response to the requirements for detection speed in high-speed motion of belt conveyors, VoVGSCSP was introduced to improve the efficient layer aggregation network (ELAN) of YOLOv7 neck, further enhancing the detection speed of the model. The experimental results show that when the belt speed is 0.6 m/s, the improved model’s mAP0.5 is similar to YOLOv7, FPS increases from 9 frames per second to 18 frames per second, coal gangue sorting rates reach 91.1%, and coal misselection rates are 4.8%. The proposed coal gangue detection and recognition method based on improved YOLOv7 has increased the detection speed of the recognition model and promoted the improvement of coal gangue sorting efficiency.

Enhanced RT-DETR for lightweight and accurate detection of non-coal foreign objects on belt conveyors

Synthesizing drive structural diagrams is typically a one-dimensional (narrowly focused) problem, traditionally solved for specific designs of lifting-and-transport machine mechanisms. This article proposes an algorithm for synthesizing drum mechanism drive structural diagrams based on a feed forward neural network architecture. This algorithm enables the synthesis of a generalized mechanism drive under conditions of design object differentiation, i.e., the use of a single synthesis algorithm for drum mechanisms of various lifting-and-transport machines (elevators, hoists, boom cranes, overhead cranes, belt conveyors, etc.). The architecture of the proposed neural network assumes supervised learning based on adjusting the weighting coefficients of connections between corresponding neurons, with data divided into sets for training, validation, and testing. A dedicated genetic algorithm for synthesizing the drum mechanism drive structural diagram for a specific type of lifting-and-transport machine is embedded in the structure of each neuron.

This study was motivated by the occurrence of a belt conveyor tear in Area C26 at the Bontang Coal Terminal, which is managed by PT Indominco Mandiri, a subsidiary of PT Indo Tambangraya Megah Tbk. The incident caused significant disruptions to the coal loading process and resulted in substantial operational losses. This research aims to analyze the operational risks arising from the incident and to identify its causes, consequences, and the effectiveness of existing risk controls. The study employs a qualitative descriptive research design using the Bowtie Analysis approach, supported by field observations, semi-structured interviews, and document analysis. The findings indicate that the damage was influenced by several factors, including belt wear, conveyor misalignment, environmental conditions, the potential ingress of foreign objects, and limitations in early detection systems. The incident led to a seven-day suspension of loading activities, increased operational costs due to demurrage, and elevated occupational safety risks. The application of Bowtie Analysis generated recommendations to strengthen both preventive and mitigative controls, such as integrating real-time monitoring sensors to enhance operational reliability

Currently, matters of dependability in the design of technical systems with moving boundaries require an increasingly complete consideration of underlying dynamic phenomena. Aim. The aim of the study is to develop a mathematical model and an approximate analytical method for studying the transverse vibrations and resonant properties of a viscoelastic rope of variable length lying on an elastic foundation, taking into account energy dissipation. The relevance of the work is due to the widespread use of technical systems with moving boundaries (lifting mechanisms, flexible transmissions, railway contact networks, rail tracks, belt conveyors, drill strings, etc.), for which dynamic loads and resonance are dangerous. The existing methods do not allow for a complete consideration of a system of factors, i.e., changes in the object’s length, resistance of the medium, elastic properties of the foundation and internal friction. Methods. To solve the problem, the Kantorovich–Galerkin method, effective for systems with moving boundaries, was applied. The original boundary value problem for a partial differential equation was reduced to a system of ordinary differential equations. The solution procedure included the transition to dimensionless variables, the selection of coordinate functions in the form of eigenmodes and the application of the Galerkin procedure. The small parameter method was used to analyze non–stationary processes. In the considered model, the drag force of the rope movement is assumed to be proportional to the velocity, and the bending rigidity of the structure is also taken into account. Results. Calculation expressions are presented for the amplitude of oscillations corresponding to the n–th dynamic mode. Particular attention is paid to the study of the phenomena of steady–state resonance and passage through resonance. The solution covers the most common case in practice of the action of external disturbances on the moving boundary of the system. It is established that the amplitude significantly depends on the velocity of the boundary, dissipation parameters and the rigidity of the foundation. The conditions for steady–state resonance are determined for a certain ratio of the frequency of the external influence and the natural frequency of the system. The phenomenon of passage through resonance is studied. The resulting analytical expressions were verified by comparison with known special cases, confirming the method’s validity with an error of up to 5% for the fundamental modes. Conclusions. The resulting analytical expressions for the oscillation amplitude, steady–state resonance conditions, and resonance passage parameters enable the formulation of a number of practical recommendations for design engineers aimed at increasing the dependability and durability of technical systems with moving boundaries and preventing resonant failures in variable–length systems. Key applied problems solved using this model include fatigue life assessment, residual life prediction, and emergency prevention. Consideration of dissipation and an elastic foundation is critical for assessing resonant properties. To prevent resonance, it is recommended to optimize the boundary velocity, use materials with increased friction or dampers, and increase the foundation rigidity. The results have practical significance for improving the dependability of systems with moving boundaries. Research prospects are related to taking into account nonlinear effects and non–harmonic influences.

Belt conveyors are a critical element of production capacity in mining and manufacturing industries. The reliability, smoothness, and trouble-free operation of belt conveyor equipment directly impact the production performance and economic efficiency of the enterprise. In this paper, a belt conveyor is considered as a single electromechanical system. A mathematical model for this system was developed by decomposing the belt conveyor mechanism, identifying individual mechanical components most susceptible to failure. The resulting model represents an eight-mass system. Based on the resulting equations, a model of the conveyor's electromechanical drive system was created in the MATLAB Simulink environment. The modeling yielded diagnostic indicators of belt conveyor failure modes, such as roller jamming, receiving hopper blockage, and others. The results of a model-based study showed that the stator current of an asynchronous motor changes characteristically during emergency situations. The period, frequency, and shape of the current oscillations depend on the specific type of fault in the mechanical component of the belt conveyor. To confirm these results, field studies were conducted on a real-world object—a type HE-K belt conveyor. During the experiments, the RMS value of the motor stator current, the frequency and voltage of the power supply network, and the output frequency and voltage of the frequency converter were recorded. The results of the study confirm the feasibility of using changes in the stator current of an asynchronous motor to diagnose emergency operating conditions of a belt conveyor. This allows for increased reliability and efficiency of equipment operation, as well as a reduction in economic losses associated with emergency situations.

Distributed acoustic sensing (DAS), based on phase-sensitive optical time-domain reflectometry (<i>Φ</i>-OTDR), transforms optical fibers into distributed vibration sensors through Rayleigh backscattering, enabling real-time industrial monitoring with extensive coverage and high spatial resolution. This review systematically presents key advances and industrial applications made by the optical fiber sensing (OFS) group at University of Electronic Science and Technology of China (UESTC), which include a differential-frequency modulation scheme integrated with a polarization-multifrequency diversity fusion algorithm and achieve pε-level strain sensitivity and suppressed signal fading down to 0.1%, enabling high-fidelity and long-distance sensing using low-cost commercial DAS units. Based on the advanced sensing capability, our developed adaptive feature enhancement method combined with an incremental tree classifier achieves the remarkable 96.55% recognition accuracy for ten types of pipeline intrusion events while reducing retraining time by 98.5% and further attains 99.96% accuracy for five major intrusion types in real field deployments. For railway infrastructure monitoring, our RailFusion-DAS framework utilizes existing fiber-optic cables along the railway to precisely identify three typical track defects with the 98.73% accuracy. Furthermore, by implementing time-frequency analysis and a two-dimensional convolutional neural network classifier on an artificial intelligence (AI) hardware accelerator, we realize an on-chip AI-DAS system that achieves 98.7% accuracy in online fault detection for belt conveyor idlers.

This study investigated a change in the quantitative content and particle size distribution of curd dust during the production of cottage cheese with fat mass fractions of 0.2%, 5%, and 9%. Modern cottage cheese production lines are characterized by a high level of mechanization of technological processes, which enables high productivity. However, mechanical impact on curd grains leads to their destruction and the formation of curd dust, the particles of which remain in the whey after its separation. Losses of raw material in the form of curd dust affect the production cost of finished products, complicate further whey processing, and increase its environmental impact. Changes in the curd dust content and its particle size distribution in whey have been investigated at different stages of production, from cutting and stirring the curd coagulum in the curd-making vat to whey separation on a belt conveyor. The final average content of curd dust in the whey obtained during cottage cheese production was determined to be 4.78 kg/m3. It was established that, on average, 25% of curd dust is formed in the curd-making vat. The maximum amount of curd dust (62%) is formed during the transportation of curd grains from the curd-making vat to the heat exchanger. In the heat exchanger, 13% of curd dust is formed. The formation of curd dust in the rotary lobe pump is explained by the significant mechanical impact on curd grains in this equipment. To reduce the level of curd dust formation, heat exchangers with minimal hydraulic resistance should be used for cooling the curd grains. This allows the use of pumps with gentler operating characteristics (compared to rotary lobe pumps) for transporting curd grains from the curd-making vat to the heat exchanger. These measures could be applied in practice to reduce raw material losses during cottage cheese production on modern mechanized production lines

This paper presents a signal-processing methodology for assessing thermal stress and fatigue damage in IGBT modules. This study utilizes junction temperature data from operational frequency converters at a belt conveyor station rather than conventional approaches. These in situ measurements ensure that thermal profiles accurately reflect actual loading conditions. A reliability framework based on mission profiles assesses the contribution of each operational regime. We examine transient overloads, steady-state operation, and periods of low load specifically. We apply Miner’s rule and rainflow counts to the analyzed temperature profiles. This enables the assessment of accumulated damage in each operational segment. The primary finding indicates that a minimal duration of operational time constitutes the majority of total lifetime utilization. This disproportionate impact is attributable to transient overloads. This study quantitatively evaluates this phenomenon using Rainflow analysis to disaggregate mission profiles. The proposed framework enhances the precision of reliability engineering. It provides a valuable foundation for enhancing maintenance planning and control strategies in practical scenarios.

The speed-regulating magnetic coupler faces challenges in mechanism modeling for constant-torque-load soft start and multi-motor power balance. Moreover, system data contain singular noise values. To address these issues, an anti-singularity data-driven control strategy is proposed. This strategy enhances speed control accuracy and operational robustness. The method designs an exponentially stable air-gap regulation law using Lyapunov theory. A robust data model is constructed by estimating angular acceleration via piecewise least squares with singularity screening, followed by model extension using a generalized distance weighting factor, which enables the numerical solution of the control law. Experimental results demonstrate a speed control accuracy within 4%. In practical applications on long-distance belt conveyors, the strategy achieves a soft-start acceleration of ≤0.15 m/s2 with a 25 s start-up time, maintains power balance among motors within a 5% deviation, and improves energy efficiency by 17.6%. This work provides an effective data-driven solution for the high-performance control of magnetic couplers in complex industrial scenarios.

The safety of belt conveyor operation is of great importance during coal conveyance. This paper proposes a multi-task-based GSSA-YOLOM algorithm for monitoring the state of belt conveyors, which utilizes segmentation head to detect foreign objects and belt deviation, thereby balancing the trade-offs among multiple tasks. The detection neck is responsible for multi-scale feature fusion by incorporating the Asymptotic Feature Pyramid Network (AFPN) to achieve enhanced spatial perception. Then, Groupwise Separable Convolution (GSConv) is further introduced to simplify the network architecture, reducing computational complexity while maintaining sufficient detection accuracy for edge device deployment. Moreover, the SlideLoss and Soft-NMS functions are integrated to reduce the rate of false positives and missed detections. Comparison experiments were conducted, and the results indicate that the proposed GSSA-YOLOM model can improve mAP@50 by 3.4% compared with the baseline model while reducing the number of parameters by 27%, thereby satisfying coal mine safety monitoring requirements.