Coal silo car loading control method and system

Abstract

The application discloses a coal silo loading control method and system, relates to the technical field of material loading control, and comprises the following steps: acquiring local space distribution data and material accumulation surface data in a carriage; calculating actual available space distribution in the carriage according to the local space distribution data; calculating a local average material accumulation rate according to the material accumulation surface data; predicting a coal overflow risk according to the actual available space distribution and the local average material accumulation rate; and controlling a double discharge port to discharge coal according to the coal overflow risk. The application can predict the overflow risk by combining space distribution and accumulation data, so as to realize loading control, and improve control precision and safety.

Description

Technical Field

[0001] This application relates to the field of material loading control technology, and in particular to a method and system for controlling the loading of coal silos. Background Technology

[0002] In large material handling facilities such as coal transfer stations, automated loading systems are used to improve operational efficiency and loading accuracy. These systems are typically equipped with dual discharge ports, using radar detection to acquire the overall dimensions of the truck bed and calculating the theoretical stacking height and total volume based on a preset standard coal density. During loading, existing methods primarily rely on weighing sensors located beneath the vehicle to monitor the total weight. As the total weight approaches the target value, the discharge port opening is gradually reduced for precise material replenishment, achieving accurate loading. However, in actual coal transportation operations, long-serving transport vehicles inevitably experience significant impacts and vibrations during daily coal loading and unloading and long-distance transport, leading to slight localized deformations in the truck bed structure. These subtle structural changes alter the actual usable volume and geometric regularity of the truck bed's interior. Existing methods may incorrectly calculate the total weight at which coal overflows due to these changes in internal volume, resulting in coal spillage, fire hazards, and other safety risks. This results in low control accuracy and low safety.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main purpose of this application is to propose a method and system for controlling the loading of coal silos, which can combine spatial distribution and stacking data to predict overflow risks, thereby achieving loading control and improving control accuracy and safety.

[0005] On the one hand, this application provides a method for controlling the loading of coal silos, including the following steps:

[0006] Acquire local spatial distribution data and material accumulation surface data inside the carriage;

[0007] Based on the local spatial distribution data, the actual usable space distribution inside the carriage is calculated;

[0008] Calculate the local average material accumulation rate based on the material accumulation surface data;

[0009] Based on the actual available space distribution and the local average material accumulation rate, the risk of coal spillage is predicted.

[0010] Based on the aforementioned coal spillage risk, coal feeding is controlled through both feed ports.

[0011] Preferably, the step of calculating the local average material accumulation rate based on the material accumulation surface data includes:

[0012] Based on the material accumulation surface data, identify local suspended coal dust layer areas inside the carriage;

[0013] Radar scanning was performed on the local suspended coal dust layer area to obtain radar echo data before airflow disturbance.

[0014] When injecting controlled airflow into the local suspended coal dust layer region, radar echo data of airflow disturbance is collected;

[0015] After injecting controlled airflow into the local suspended coal dust layer area, radar echo data after airflow disturbance is collected;

[0016] Based on the radar echo data before the airflow disturbance, the radar echo data during the airflow disturbance, and the radar echo data after the airflow disturbance, differential analysis is performed to calculate the radar sensing height difference.

[0017] If the difference in radar sensing height is greater than a preset stability threshold, the thickness of the suspended coal dust layer is calculated based on the radar echo data before the airflow disturbance and the radar echo data after the airflow disturbance.

[0018] The height of the solid coal accumulation surface is calculated based on the thickness of the suspended coal dust layer and the current radar sensing height.

[0019] The local average material accumulation rate is calculated based on the surface height of the solid coal accumulation.

[0020] Preferably, predicting coal spill risk based on the actual available space distribution and the local average material accumulation rate includes:

[0021] Acquire vehicle attitude data and vehicle position data;

[0022] Calculate the carriage posture change based on the vehicle attitude data and the vehicle position data;

[0023] Based on the change in the carriage's posture, the coordinate transformation of the material accumulation surface data is performed to correct the position of the material accumulation surface data in the carriage coordinate system;

[0024] Calculate the relative positional relationships based on the actual available spatial distribution and the material accumulation surface data after coordinate transformation;

[0025] If the relative positional relationship is greater than the preset positional deviation threshold, then the material accumulation surface data after coordinate transformation is locally re-matched according to the actual available space distribution.

[0026] Based on the actual available space distribution, the local average material accumulation rate, and the material accumulation surface data after local spatial rematching, the risk of coal spillage is predicted.

[0027] Preferably, predicting coal spill risk based on the actual available space distribution and the local average material accumulation rate includes:

[0028] Collect the local material accumulation height;

[0029] Calculate the local instantaneous material accumulation rate based on the local material accumulation height;

[0030] The accumulation rate deviation is calculated based on the local instantaneous material accumulation rate and the local average material accumulation rate.

[0031] Based on the stacking rate deviation and the changing trend of the discharge port opening, the local average material stacking rate is corrected to obtain the target material stacking rate.

[0032] Based on the target material accumulation rate and the actual available space distribution, the risk of coal spillage is predicted.

[0033] Preferably, controlling the coal feeding from both feed ports based on the coal spill risk includes:

[0034] Obtain material properties;

[0035] Based on the material characteristics, calculate the mapping relationship between the discharge port opening and the coal discharge flow rate;

[0036] Monitor the actual flow rate at the feed inlet;

[0037] Calculate the expected discharge flow rate based on the discharge port opening and the mapping relationship;

[0038] Calculate the flow deviation based on the actual flow rate at the discharge port and the expected discharge flow rate;

[0039] The opening of the feed inlet is adjusted based on the coal spillage risk and the flow deviation.

[0040] After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

[0041] Preferably, controlling the coal feeding from both feed ports based on the coal spill risk includes:

[0042] Based on the coal spillage risk, determine the opening adjustment amount corresponding to each feed port;

[0043] The mutual influence between each discharge port and the material accumulation inside the car is analyzed to obtain the mutual influence coefficient.

[0044] Based on the mutual influence coefficient, the opening adjustment amount corresponding to each feed port is corrected;

[0045] Based on the corrected opening adjustment amount, adjust the opening of each feed port and control the coal feeding through both feed ports.

[0046] Preferably, the analysis of the mutual influence between each discharge port and the material accumulation inside the carriage to obtain the mutual influence coefficient includes:

[0047] Acquire surface morphology data of materials already piled up inside the carriage;

[0048] Based on the surface morphology data, local accumulation features inside the carriage are identified, including slopes and pits;

[0049] Based on the local accumulation characteristics, calculate the diffusion path and landing point distribution of the feed bundle;

[0050] Based on the diffusion path and the distribution of landing points, the mutual influence between each discharge port and the material accumulation inside the carriage is analyzed to obtain the mutual influence coefficient.

[0051] Preferably, controlling the coal feeding from both feed ports based on the coal spill risk includes:

[0052] Obtain end-of-loading overflow risk information, which includes remaining available space and local stacking height;

[0053] If the remaining available space is less than a preset space threshold and the local stacking height is greater than a preset critical value, then the material stacking trend is calculated based on the local instantaneous material stacking rate.

[0054] Based on the coal spillage risk and the material accumulation trend, the discharge port opening is adjusted to reduce the step size of the discharge port opening adjustment and increase the frequency of the discharge port opening adjustment.

[0055] After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

[0056] Preferably, controlling the coal feeding from both feed ports based on the coal spill risk includes:

[0057] Obtain material stacking characteristics;

[0058] The changing trend of the material's packing characteristics is analyzed to obtain the changing trend of the material's packing characteristics;

[0059] Based on the changing trend of the material's stacking characteristics, determine the material diffusion parameters;

[0060] Based on the material diffusion parameters and the material accumulation characteristics, the diffusion range is predicted using a material diffusion model.

[0061] The opening of the discharge port is adjusted according to the diffusion range.

[0062] After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

[0063] On the other hand, this application provides a coal silo loading control system, including:

[0064] The data acquisition module is used to acquire local spatial distribution data and material accumulation surface data inside the carriage;

[0065] The spatial distribution calculation module is used to calculate the actual usable space distribution inside the carriage based on the local spatial distribution data.

[0066] The accumulation rate calculation module is used to calculate the local average material accumulation rate based on the material accumulation surface data.

[0067] The spillover risk prediction module is used to predict the coal spillover risk based on the actual available space distribution and the local average material accumulation rate.

[0068] The material feeding control module is used to control the coal feeding from the dual feeding ports based on the coal overflow risk.

[0069] This application includes at least the following beneficial effects: First, it acquires local spatial distribution data and material accumulation surface data inside the car body. Then, based on the local spatial distribution data, it calculates the actual usable space distribution inside the car body and calculates the local average material accumulation rate based on the material accumulation surface data. Next, based on the actual usable space distribution and the local average material accumulation rate, it predicts the risk of coal spillage. Finally, based on the risk of coal spillage, it controls the coal feeding through the dual discharge ports. This allows for the prediction of spillage risk by combining spatial distribution and accumulation data, thereby achieving loading control and improving control accuracy and safety.

[0070] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description and the accompanying drawings. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0072] Figure 1 A flowchart illustrating a coal silo loading control method provided in this application embodiment;

[0073] Figure 2This is a schematic diagram of the structure of a coal silo loading control system provided in an embodiment of this application;

[0074] In the diagram: 301 is the data acquisition module, 302 is the spatial distribution calculation module, 303 is the stacking rate calculation module, 304 is the overflow risk prediction module, and 305 is the material feeding control module. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0076] In modern industrial production, particularly in large material handling facilities like coal transfer stations, automated loading systems have become indispensable for achieving higher operational efficiency and loading accuracy. These systems typically utilize advanced sensing technologies, such as radar, to accurately perceive the status of transport vehicles and, based on this, achieve precise loading of bulk materials through automated control. However, in actual operation, due to various complex factors, these automated systems still face unexpected challenges that often affect loading quality and efficiency, and may even pose safety hazards.

[0077] At a large coal transfer station, an advanced automated coal silo loading system was introduced to improve loading efficiency and reduce manual intervention. The core of this system is a coal silo equipped with dual discharge ports, capable of simultaneously loading coal into the freight cars of transport vehicles. Before loading begins, the transport vehicles enter a designated loading area. At the entrance of this area, a high-precision radar detection device is installed. Once the vehicle comes to a stop, the radar system automatically activates, performing a 360-degree scan of the freight car. Through the reflection and reception of radar waves, the system accurately acquires the internal dimensions of the car, including its length, width, and internal depth when empty. This dimensional data is then transmitted to the central control system. Based on a preset standard coal density and this dimensional information, the control system calculates the theoretical stacking height and total volume of coal in the car when the target loading weight is achieved. Once all parameters are confirmed to be correct, the control system simultaneously opens the valves of both discharge ports, beginning the loading of coal into the car. Throughout the loading process, weighing sensors beneath the vehicle continuously monitor the vehicle's real-time total weight and feed this data back to the control system. As the monitored total weight gradually approaches the preset target loading weight, the control system, based on a pre-set algorithm, progressively reduces the valve openings of the two discharge ports to slow down the material feeding speed and perform precise replenishment, ensuring loading accuracy. Finally, when the vehicle's total weight precisely reaches the target value, the control system completely closes the valves of both discharge ports, marking the completion of one automated loading operation. This design aims to achieve efficient and precise coal loading, optimizing the operational process.

[0078] However, in actual coal transportation operations, automated loading processes encounter some unexpected complexities. Firstly, the freight cars of transport vehicles, especially those that have served under harsh conditions for extended periods, inevitably endure significant impacts and vibrations during daily coal loading and unloading and long-distance transport. These continuous mechanical stresses, along with the abrasive effects of the coal itself, can cause slight deformations in the car body structure that are difficult to detect with the naked eye. For example, the side panels or floor of the car body may no longer be perfectly flat, but may exhibit slight inward or outward concavity, or minute twisting in localized areas. While these subtle structural changes usually do not immediately affect the overall structural strength or basic load-bearing capacity of the vehicle, they subtly alter the actual usable volume and geometric regularity of the car body's interior.

[0079] While existing radar detection devices demonstrate high accuracy in measuring the overall dimensions of a vehicle compartment, their design and data processing logic are typically optimized based on an idealized, regular rectangular compartment model. When the radar beam scans the interior of the compartment, it acquires a series of discrete point cloud data reflecting distance information to the compartment's interior walls. However, when converting this raw point cloud data into compartment dimensions usable for loading calculations, the control system often uses a simplified geometric model for fitting. For example, it might determine the "average depth" of the compartment by averaging multiple depth measurements or directly calculate the outer contour dimensions of the compartment. For minor local deformations in the compartment's side panels or floor, such as a side panel recessing inward by several centimeters in a certain area, the radar system may not be able to capture these subtle shape changes with sufficient detail. Even if the radar sensors themselves can detect these minute deformations, these subtle local irregularities are often smoothed out during subsequent data processing and model fitting, leading to a discrepancy between the "usable volume" calculated by the system and the actual volume reduced due to local deformation. In short, when performing loading calculations, the control system may mistakenly assume that the interior of the carriage is a standard, geometrically regular rectangular space. However, due to local deformations, the actual usable space in some areas of the carriage is smaller than the theoretically calculated value.

[0080] The difference between the "theoretical volume" and the "actual volume" does not immediately become apparent in the initial stages of loading. When the control system calculates the theoretical coal pile height corresponding to the target loading weight based on the (slightly inaccurate) dimensional information provided by radar and the preset standard coal density, it assumes that the car has sufficient space to accommodate the calculated amount of coal. However, as coal begins to pour into the car from the double discharge ports, the effects of this localized deformation gradually become apparent. As a bulk material, coal preferentially fills the available space below it under gravity. In areas with less available space, such as the recessed parts of the car's side panels, the coal reaches its localized "full load" state more quickly, meaning the surface of the coal pile rises faster than expected. In other areas of the car, where the space is relatively larger, the coal pile accumulates more slowly. This results in the coal pile surface inside the car no longer exhibiting the expected flat or uniformly rising shape during loading, but rather displaying localized highs and lows. For example, if a side panel in the middle of the carriage has a noticeable indentation, the coal pile directly below that area may be higher than the coal piles on either side, forming a local "bulge".

[0081] This localized, uneven accumulation can trigger a serious and insidious problem even before the vehicle's total weight reaches the preset target: in a certain area of ​​deformation, due to the rapid accumulation of coal, the top of the coal pile may already be very close to or even exceed the height of the truck bed's side panels. At this point, although the total weight reported by the weighing system below the vehicle may still be far below the preset target loading weight, coal continuing to fall from that chute will overflow directly from the side of the truck bed due to insufficient local space, scattering onto the loading platform or the ground. Because the control system's closed loop relies solely on the vehicle's total weight, it cannot detect the localized accumulation height and material distribution within the truck bed. Therefore, it will continue to operate according to the predetermined loading logic until the total weight reaches or is close to the target, thus exacerbating the overflow problem.

[0082] Coal spills not only directly waste valuable materials and increase the workload and cost of on-site cleanup, but more importantly, the scattered coal dust severely pollutes the environment and may pose safety hazards such as fires and explosions. Even without obvious spills, such localized excessive accumulation can cause coal to scatter from the top or sides of the carriage due to inertia during transportation, especially when the vehicle is bumpy, turning, or braking, causing leakage and further aggravating material loss and environmental pollution. In addition, if such localized excessive accumulation occurs for a long time in a specific load-bearing area of ​​the carriage, it may accelerate fatigue damage to the carriage structure in that area, shorten the vehicle's service life, and increase maintenance costs. Existing control methods, while achieving accurate loading based on total weight, create new and more hidden problems due to their "blind spots" regarding the actual usable space inside the carriage and the real-time accumulation pattern. This means that localized spills may occur before the total weight is reached, or the carriage structure may age faster.

[0083] In the scenario of automated loading of coal silos with dual feed ports based on radar detection of vehicle parameters, there are complex situations such as local structural deformation of the silo due to long-term service, deviation of the radar initial scan from the actual usable volume, and the inability of existing control to perceive local stacking height by relying only on the total weight. It is necessary to obtain the actual usable space distribution inside the silo and the real-time stacking height of the coal in real time during loading, and intelligently adjust the feeding strategy of the dual feed ports accordingly. This can effectively avoid local overflow before the total weight is reached, ensure that the coal is safely and completely loaded and accurately reaches the target total weight, while eliminating material waste and environmental pollution.

[0084] In view of this, this application, by acquiring local spatial distribution data inside the car body, can accurately calculate the actual usable space distribution inside the car body. Simultaneously, by acquiring material accumulation surface data and calculating the local average material accumulation rate, it can monitor the dynamic accumulation of coal inside the car body in real time and identify areas with excessively rapid accumulation. Based on this, this application combines the actual usable space distribution and the local average material accumulation rate to predict the risk of coal spillage. This prediction is based on a comprehensive understanding of the actual situation inside the car body, rather than solely relying on the total weight. This application can dynamically adjust the opening of the dual discharge ports according to the risk level, achieving refined control of coal discharge and effectively preventing local spillage.

[0085] The embodiments of this application will be explained in detail below with reference to the accompanying drawings:

[0086] Figure 1 This is an optional flowchart of a coal silo loading control method provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S105.

[0087] Step S101: Obtain local spatial distribution data and material accumulation surface data inside the carriage;

[0088] Step S102: Calculate the actual usable space distribution inside the carriage based on the local spatial distribution data;

[0089] Step S103: Calculate the local average material accumulation rate based on the material accumulation surface data;

[0090] Step S104: Based on the actual available space distribution and local average material accumulation rate, predict the risk of coal spillage;

[0091] Step S105: Based on the risk of coal spillage, control the coal feeding through both feed ports.

[0092] Steps S101 to S105 as shown in the embodiments of this application can combine spatial distribution and stacking data to predict overflow risks, thereby achieving loading control and improving control accuracy and safety.

[0093] In some embodiments, steps S101-S105 can first acquire local spatial distribution data and material accumulation surface data inside the wagon. For example, a LiDAR scanner installed on the top of the silo or the side of the wagon can be used to perform high-frequency scanning of the wagon's interior to acquire three-dimensional point cloud data. This point cloud data contains precise geometric information of the wagon's walls, i.e., local spatial distribution data. Simultaneously, the LiDAR can continuously scan the coal accumulation surface to acquire real-time material accumulation surface data. In another embodiment, multiple 3D vision sensors (such as depth cameras) can work collaboratively to construct a three-dimensional model of the wagon's interior and the three-dimensional morphology of the coal accumulation surface through multi-view image fusion technology. These sensors can be configured to operate continuously during loading to ensure the real-time nature and accuracy of the data. It is understood that local spatial distribution data refers to detailed three-dimensional data about the spatial geometry inside the wagon acquired by sensors (such as LiDAR, 3D vision systems, etc.). This data reflects the actual contour of the wagon's interior, including any possible local deformation areas. Material accumulation surface data refers to the three-dimensional morphological data of the coal accumulation surface inside the car body, which is monitored in real time by sensors. This data can reflect the real-time height, slope, and local unevenness of the coal accumulation.

[0094] Then, based on the local spatial distribution data, the actual usable space distribution inside the carriage is calculated. A 3D reconstruction algorithm can be used to process the point cloud data and generate an accurate 3D model of the carriage's interior. By performing volume calculations and spatial analysis on this 3D model, the actual usable space distribution in different areas inside the carriage can be obtained. For example, multiple small voxel units can be divided, and the usable volume of each voxel unit can be calculated, thus forming a refined usable space distribution map. This method accurately reflects the changes in actual usable space caused by local deformation of the carriage, rather than being based on an idealized rectangular model. Simultaneously, based on the material accumulation surface data, the local average material accumulation rate is calculated. Time series analysis of the accumulation surface data can be performed, comparing the changes in the accumulation surface height of the same area at different time points to calculate the local instantaneous material accumulation rate in that area. To obtain a more stable "local average material accumulation rate," a moving average or weighted average can be applied to the local instantaneous material accumulation rate to eliminate the influence of short-term fluctuations. For example, the interior of the car can be divided into several virtual regions. The coal accumulation surface height in each region can be monitored, and the average height increase of each region within a certain time window can be calculated, thus obtaining the local average material accumulation rate for that region. It can be understood that the actual usable space distribution refers to the actual volume and shape information of the interior of the car that can be used to accumulate coal in different regions, calculated based on local spatial distribution data; it takes into account the local deformation of the car. The local average material accumulation rate refers to the average rate at which the coal accumulation height changes over time within a specific region inside the car, reflecting the coal filling efficiency of that region.

[0095] Based on the actual available space distribution and the local average material accumulation rate, the risk of coal spillage can be predicted by inputting these factors into a risk assessment model. This model can be a simulation model based on physical laws or a predictive model based on machine learning. For example, a simulation model can simulate the coal accumulation process over a future period based on the current available space and material accumulation rate in each area, and predict which areas' coal accumulation height is likely to reach the height of the wagon's side panels before reaching the target total weight. A machine learning model can be trained using historical loading data (including spillage events) to learn the complex relationship between the actual available space distribution, the local average material accumulation rate, and spillage risk, thereby predicting the spillage risk under the current loading conditions. It can be understood that coal spillage risk refers to the probability of coal spilling from the wagon during loading, assessed based on the actual available space distribution and the local average material accumulation rate. This risk can be qualitative (high, medium, low) or quantitative (spillage probability).

[0096] Finally, based on the risk of coal spillage, the dual discharge ports are controlled to discharge coal. For example, the opening of the dual discharge ports can be dynamically adjusted according to the risk level or probability. For instance, if the spillage risk is high in a certain area, the discharge control module can correspondingly reduce the opening of the corresponding discharge port or temporarily close it to reduce the material accumulation rate in that area. Simultaneously, to ensure overall loading efficiency, the opening of the discharge ports corresponding to other lower-risk areas can be appropriately increased. This control strategy enables refined management of coal discharge, avoiding spillage caused by excessive local accumulation. For example, when a high spillage risk is predicted in the front left side of the car body, the system can reduce the opening of the left discharge port while maintaining or slightly increasing the opening of the right discharge port to balance the material accumulation inside the car body. It is understood that dual discharge ports refer to the two or more outlets typically equipped in a coal silo for discharging coal into the car body. By independently controlling these discharge ports, refined adjustment of the coal discharge position and flow rate can be achieved.

[0097] Through the above technical solution, this embodiment can solve the problems of uneven loading and overflow caused by local deformation of the car body, improving the accuracy and safety of loading. By real-time monitoring and prediction of the local material accumulation rate, this embodiment can detect potential overflow risks earlier and take preventive measures. At the same time, this embodiment is more flexible and intelligent, able to dynamically adjust the material feeding strategy according to the real-time status inside the car body, optimizing loading efficiency. Therefore, this embodiment has significant technological advancements and practical value in improving coal loading efficiency, reducing material waste, reducing environmental pollution, and enhancing operational safety.

[0098] In some embodiments, step S103, calculating the local average material accumulation rate based on the material accumulation surface data, may include, but is not limited to, the following steps:

[0099] Based on the surface data of the material accumulation, identify the local suspended coal dust layer area inside the car;

[0100] Radar scanning was performed on a localized area of ​​suspended coal dust to obtain radar echo data before airflow disturbance.

[0101] When injecting controlled airflow into a local suspended coal dust layer area, radar echo data of airflow disturbance is collected;

[0102] After injecting controlled airflow into a local suspended coal dust layer area, radar echo data after airflow disturbance is collected.

[0103] Differential analysis is performed based on radar echo data before airflow disturbance, radar echo data during airflow disturbance, and radar echo data after airflow disturbance to calculate the radar sensing height difference.

[0104] If the difference in radar sensing height is greater than the preset stability threshold, the thickness of the suspended coal dust layer is calculated based on the radar echo data before and after the airflow disturbance.

[0105] The height of the solid coal accumulation surface is calculated based on the thickness of the suspended coal dust layer and the current radar sensing height.

[0106] Calculate the local average material accumulation rate based on the surface height of the solid coal accumulation.

[0107] In some embodiments, the coal feeding process generates a large amount of coal dust. These suspended coal dust layers may interfere with the accurate acquisition of material accumulation surface data, thereby affecting the calculation accuracy of the local average material accumulation rate. This could lead to inaccurate predictions of coal spill risk, which in turn affects the control effect of the dual discharge ports and increases the risk of coal spillage. To address this, local suspended coal dust layer areas inside the truck bed can be identified based on the material accumulation surface data. For example, this can be achieved by analyzing abnormal reflection, scattering, or attenuation characteristics in the material accumulation surface data. For instance, lidar or millimeter-wave radar can be used to scan the inside of the truck bed. A significant decrease in echo signal intensity or the presence of multiple reflections may indicate the presence of a suspended coal dust layer. Alternatively, image information acquired by visual sensors can be combined with image processing algorithms to identify blurred areas of coal dust.

[0108] Then, radar scanning was performed on the local suspended coal dust layer area to obtain radar echo data before airflow disturbance. Without applying external airflow disturbance to this area, the raw echo data acquired by the radar sensor reflects the initial state of the material accumulation surface, including the coal dust layer. When a controlled airflow was injected into the local suspended coal dust layer area, radar echo data was collected during airflow disturbance. This can be achieved by applying a controlled airflow to the area via an injector, causing a certain degree of disturbance to the suspended coal dust layer, and collecting radar echo data during this process. This controlled airflow can be a pre-set airflow with preset pressure, flow rate, and direction, its purpose being to distinguish the solid material surface from the suspended coal dust layer. After injecting the controlled airflow into the local suspended coal dust layer area, radar echo data was collected after the airflow disturbance. After the controlled airflow action ended, radar echo data was collected again. At this time, some of the suspended coal dust may have been dispersed or settled, and the radar echo data will more closely approximate the true surface of the solid material.

[0109] Then, based on the radar echo data before, during, and after the airflow disturbance, differential analysis is performed to calculate the radar sensing height difference. This aims to quantify the impact of suspended coal dust layers on radar signals by comparing echo data at different time points (before, during, and after the disturbance). For example, the differences in echo signal intensity or distance data before and during the disturbance, during and after the disturbance, or before and after the disturbance can be calculated to reflect the dynamic changes in the coal dust layer. The radar sensing height difference refers to the change in surface height sensed by the radar under different disturbance conditions.

[0110] If the difference in radar sensing height exceeds a preset stability threshold, it indicates the presence of a significant suspended coal dust layer in the area, and its impact on radar measurements is not negligible. The preset stability threshold is an empirical value or a parameter determined experimentally, used to judge whether the coal dust layer is sufficiently stable or thick, requiring correction. The thickness of the suspended coal dust layer is calculated based on radar echo data before and after airflow disturbance. The thickness of the suspended coal dust layer can be estimated by comparing the difference in radar sensing height before and after the disturbance. For example, subtracting the radar sensing height after the disturbance from the radar sensing height before the disturbance provides an approximate thickness of the suspended coal dust layer.

[0111] Finally, the surface height of the solid coal accumulation is calculated based on the thickness of the suspended coal dust layer and the current radar sensing height. Subtracting the calculated suspended coal dust layer thickness from the current radar sensing height yields a more accurate estimate of the actual solid coal accumulation surface height. The current radar sensing height refers to the height measured by radar before or during airflow disturbance. Based on the solid coal accumulation surface height, the local average material accumulation rate is calculated. By monitoring the change of this height over time and combining it with information such as the material flow rate, the local average material accumulation rate for that area can be accurately calculated.

[0112] To illustrate this technical solution more clearly, a specific example is used below. Suppose that during the loading process of a coal silo, a thick layer of suspended coal dust is generated in a certain area inside the silo due to coal feeding. Traditional radar measurements might directly identify the top of the dust layer as the material accumulation surface, leading to an overestimation of the calculated material accumulation height. To address this, the local suspended coal dust layer area can be identified first through radar scanning. Then, the system injects a controlled airflow at a preset pressure into this area, simultaneously collecting radar echo data before, during, and after the airflow disturbance. For example, before the disturbance, the radar sensing height is H1; after the disturbance, some coal dust is dispersed, and the radar sensing height becomes H2. Differential analysis is used to calculate the difference in radar sensing height |H1-H2|. If this difference is greater than a preset stability threshold, a significant coal dust layer is considered to exist. In this case, the thickness of the suspended coal dust layer is calculated as H1-H2. The height of the solid coal accumulation surface is corrected to H2, thus allowing the calculation of the local average material accumulation rate in this area based on a more accurate H2. In this way, reliable material accumulation data can be obtained even in environments filled with coal dust, ensuring the accuracy of loading control.

[0113] Through the above technical solution, this embodiment, by introducing airflow disturbance and multi-stage radar echo data differential analysis, achieves the identification, quantification, and stripping of suspended coal dust layers, thereby obtaining a more accurate solid coal accumulation surface height. This significantly improves the calculation accuracy of the local average material accumulation rate, thus making the prediction of coal spillage risk more accurate and reliable. This embodiment can avoid misjudgments caused by coal dust interference, effectively reduce the risk of coal spillage, improve the automation and intelligence level of the loading process, and reduce material loss and environmental pollution.

[0114] In some embodiments, in step S104, predicting the coal spill risk based on the actual available space distribution and the local average material accumulation rate may include, but is not limited to, the following steps:

[0115] Acquire vehicle attitude data and vehicle position data;

[0116] Calculate the carriage posture change based on vehicle attitude data and vehicle position data;

[0117] Based on the changes in the carriage's posture, coordinate transformation is performed on the material accumulation surface data to correct the position of the material accumulation surface data in the carriage coordinate system;

[0118] Calculate the relative positional relationships based on the actual available spatial distribution and the material accumulation surface data after coordinate transformation;

[0119] If the relative positional relationship is greater than the preset positional deviation threshold, then the material accumulation surface data after coordinate transformation will be locally re-matched according to the actual available spatial distribution.

[0120] Based on the actual available space distribution, local average material accumulation rate, and material accumulation surface data after local spatial rematching, the risk of coal spillage is predicted.

[0121] In some embodiments, changes in vehicle attitude or position can cause discrepancies between the material accumulation surface data and the actual usable space distribution within the vehicle compartment. This discrepancy can affect the accuracy of coal spill risk prediction, potentially leading to inaccurate predictions and increasing the likelihood of coal spills, resulting in material waste and environmental pollution. To address this, vehicle attitude and position data can be acquired first. For example, sensors installed on the vehicle, such as inertial measurement units (IMUs), GPS receivers, or visual sensors, can be used to collect real-time vehicle attitude information (e.g., pitch, roll, and yaw angles) and position information (e.g., longitude, latitude, and altitude). This data forms the basis for subsequent calculations of changes in the vehicle compartment's attitude.

[0122] Then, based on the vehicle attitude data and vehicle position data, the carriage pose change is calculated. Using the acquired vehicle attitude and position data, data fusion algorithms or kinematic models can be used to calculate the real-time position and attitude change of the carriage relative to a reference coordinate system (e.g., the ground coordinate system or the initial loading coordinate system). This can be represented as a series of translation and rotation transformation matrices. Based on the carriage pose change, the material accumulation surface data is then transformed to correct its position in the carriage coordinate system. Since the material accumulation surface data is usually acquired in the sensor's own coordinate system or a fixed coordinate system, its actual position inside the carriage will shift when the carriage pose changes. Therefore, it is necessary to use the calculated carriage pose change to transform the material accumulation surface data from its original coordinate system to the current carriage coordinate system, ensuring that the material accumulation surface data accurately reflects its true spatial position inside the current carriage.

[0123] Then, based on the actual available space distribution and the material stacking surface data after coordinate transformation, the relative positional relationship is calculated. After coordinate transformation, the material stacking surface data is compared with the actual available space distribution inside the carriage to assess the degree of alignment or deviation between the two. This relative positional relationship can be expressed as the degree of overlap, distance deviation, or geometric matching error between the two.

[0124] If the relative positional relationship exceeds a preset positional deviation threshold, then the material accumulation surface data after coordinate transformation is locally re-matched based on the actual available space distribution. This can be achieved through an Iterative Closest Point (ICP) algorithm, feature point matching, or optimization-based methods, aiming to fine-tune the precise position of the material accumulation surface data inside the carriage to achieve optimal alignment with the actual available space distribution. The preset positional deviation threshold is used to determine whether re-matching is necessary, balancing computational efficiency and matching accuracy.

[0125] Finally, based on the actual available space distribution, local average material accumulation rate, and material accumulation surface data after local space rematching, the risk of coal spillage is predicted. Combining the actual available space distribution inside the car with the local average material accumulation rate allows for more accurate material accumulation status information and more reliable prediction of coal spillage risk.

[0126] To illustrate this technical solution more clearly, a specific example is used below. Assume that during the loading process of a coal silo, a train car initially tilts and shifts slightly due to uneven tracks or uneven stress on the suspension system. This application uses IMU and GPS sensors installed on the car to acquire the car's attitude and position data in real time. Based on this data, the system calculates the car's real-time pose change, for example, detecting a 0.5-degree tilt to one side and a 10-centimeter forward movement. Subsequently, the system uses this pose change information to perform coordinate transformation on the material accumulation surface data acquired by sensors such as lidar, accurately converting the material data from the sensor coordinate system to the coordinate system of the currently tilted and moved car. Next, the system compares the transformed material accumulation surface data with the actual usable space distribution inside the car to calculate the relative positional relationship. If the deviation between the two exceeds a preset positional deviation threshold, the system further executes a local spatial re-matching algorithm, such as through iterative optimization, to achieve optimal alignment between the material accumulation surface data and the actual usable space distribution inside the car. Ultimately, by combining precisely calibrated and rematched material accumulation surface data, actual available space distribution, and local average material accumulation rate, the system predicts a more accurate risk of coal spillage, adjusts the feeding strategy in a timely manner, and effectively avoids coal spillage caused by changes in the car's position.

[0127] Through the above technical solution, this embodiment effectively solves the problem of mismatch between material accumulation surface data and the actual usable space distribution in the wagon due to changes in vehicle posture and position, significantly improving the accuracy of coal spillage risk prediction. This embodiment enables the loading control system to adapt to dynamic loading environments, reducing the risk of coal spillage and thus improving the safety, efficiency, and economy of loading operations.

[0128] In some embodiments, in step S104, predicting the coal spill risk based on the actual available space distribution and the local average material accumulation rate may include, but is not limited to, the following steps:

[0129] Collect the local material accumulation height;

[0130] Calculate the local instantaneous material accumulation rate based on the local material accumulation height;

[0131] The accumulation rate deviation is calculated based on the local instantaneous material accumulation rate and the local average material accumulation rate.

[0132] Based on the deviation in the accumulation rate and the changing trend of the discharge port opening, the local average material accumulation rate is corrected to obtain the target material accumulation rate.

[0133] Based on the target material accumulation rate and the actual available space distribution, predict the risk of coal spillage.

[0134] In some embodiments, since the material accumulation rate may fluctuate instantaneously, and the adjustment of the discharge port opening directly affects the material discharge flow rate, relying solely on the local average material accumulation rate for prediction may not fully reflect the dynamic changes in material accumulation and the impact of discharge control, thus affecting the accuracy of spill risk prediction. Therefore, local material accumulation height can be collected first. For example, various sensors, such as LiDAR, ultrasonic sensors, or vision systems, can be used to acquire real-time data on the material accumulation height at different locations inside the vehicle compartment. This data can reflect the real-time morphology of the material accumulation surface.

[0135] Then, based on the local material accumulation height, the local instantaneous material accumulation rate is calculated. By comparing the local material accumulation height collected at different time points, the change in material accumulation height over a short period can be calculated, thus obtaining the material accumulation rate at the current moment or within a very short time period. For example, this can be obtained by performing time-difference processing on continuously collected local material accumulation height data. Based on the local instantaneous material accumulation rate and the local average material accumulation rate, the accumulation rate deviation is calculated. This allows for comparison between the real-time calculated local instantaneous material accumulation rate and the local average material accumulation rate calculated over a previous period, quantifying the difference between the two. This deviation indicates the fluctuation of the current material accumulation rate relative to the average level.

[0136] Then, based on the deviation in the accumulation rate and the trend of the discharge port opening, the local average material accumulation rate is corrected to obtain the target material accumulation rate. The trend of the discharge port opening refers to whether the discharge port opening is increasing, decreasing, or remaining constant, and the rate of change. Adaptive filtering algorithms, such as Kalman filtering, or rule-based expert systems can be used for correction. When the local instantaneous material accumulation rate deviates significantly from the average value, and the trend of the discharge port opening also supports this deviation (e.g., an increase in the discharge port opening leads to an increase in the local instantaneous material accumulation rate), the local average material accumulation rate is adjusted accordingly to better reflect the current actual accumulation situation, thus obtaining a more real-time and accurate target material accumulation rate.

[0137] Finally, based on the target material accumulation rate and the actual available space distribution, the risk of coal spillage is predicted. This target material accumulation rate combines historical averages with real-time dynamic changes, more accurately reflecting the material accumulation trend. Combined with the actual available space distribution inside the wagon, this allows for a more precise prediction of the risk of coal spillage.

[0138] To illustrate this technical solution more clearly, a specific example is used below. Assume that during coal loading, the system continuously collects local material accumulation height data inside the car. At a certain moment, the sensor detects a rapid increase in material accumulation height in a certain area within a short period, and the calculated local instantaneous material accumulation rate is significantly higher than the current local average material accumulation rate. Simultaneously, the system monitors that the opening of one or both discharge ports is continuously increasing. At this point, based on the deviation between the local instantaneous material accumulation rate and the local average material accumulation rate, and the increasing trend of the discharge port opening, the system determines that the current material accumulation rate tends to accelerate. Based on this, the system will correct the local average material accumulation rate upwards to obtain a higher target material accumulation rate. Subsequently, using this corrected target material accumulation rate and the actual available space distribution in the car, the system will predict that the risk of coal spillage will arrive earlier or at a higher risk level. This dynamic correction mechanism allows the system to respond promptly to sudden changes in material accumulation and adjustments to the discharge control strategy, thereby taking preventative measures before actual spillage occurs, such as timely adjustment of the discharge port opening or suspension of discharge, effectively avoiding coal spillage accidents.

[0139] Through the above technical solution, this embodiment can significantly improve the accuracy and real-time performance of spillage risk prediction during coal silo loading. This embodiment considers the instantaneous changes in material accumulation and the influence of the discharge port opening, enabling earlier and more accurate identification of potential spillage risks, thus providing a more reliable basis for subsequent material discharge control. This helps avoid overloading or underloading due to inaccurate predictions, improving loading efficiency and safety, and reducing material waste and environmental pollution.

[0140] In some embodiments, step S105, controlling the coal feeding from both feed ports according to the risk of coal spillage, may include, but is not limited to, the following steps:

[0141] Obtain material properties;

[0142] Based on the material characteristics, calculate the mapping relationship between the feed inlet opening and the coal feed flow rate;

[0143] Monitor the actual flow rate at the feed inlet;

[0144] Calculate the expected discharge flow rate based on the discharge port opening and mapping relationship;

[0145] Calculate the flow deviation based on the actual flow rate at the discharge port and the expected discharge flow rate;

[0146] The opening of the feed inlet is adjusted based on the risk of coal spillage and flow deviation.

[0147] After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

[0148] In some embodiments, relying solely on predicted coal spillage risk for feed control may not adequately address dynamic changes in material properties and deviations between actual and expected feed flow rates at the discharge port, thus affecting loading accuracy and efficiency, and potentially leading to localized overload or spillage. To address this, material properties can be acquired beforehand. For example, the physical and chemical properties of the coal to be loaded, such as particle size distribution, density, moisture content, and coefficient of friction, can be obtained. These properties directly influence the flow behavior and packing morphology of the coal during the feeding process. Material properties can be measured in real-time using sensors, such as online particle size analyzers and densitometers, or queried from a pre-set material database.

[0149] Then, based on the material characteristics, the mapping relationship between the discharge port opening and the coal discharge flow rate is calculated. A mathematical model can be established or a preset reference table can be consulted. This model or reference table can predict the actual coal discharge flow rate at a specific discharge port opening based on the current material characteristics. For example, for coal with different particle sizes or moisture contents, the same discharge port opening may produce different flow rates. Therefore, it is necessary to dynamically adjust or select this mapping relationship according to the material characteristics.

[0150] Next, the actual flow rate at the discharge port is monitored. This can be done in real time using flow sensors (e.g., belt scales, Coriolis mass flow meters, or vision-based flow estimation systems) to measure the actual flow rate of coal currently falling through each discharge port. The purpose is to obtain real-time flow feedback for comparison with the expected discharge flow rate. Based on the discharge port opening and mapping relationship, the expected discharge flow rate is calculated, thus predicting the coal flow rate that should be generated at each discharge port under the current operating conditions. The flow deviation is calculated based on the actual and expected discharge flow rates. This allows for comparison between the real-time monitored actual flow rate and the expected discharge flow rate calculated based on the mapping relationship, revealing the difference between the two. This flow deviation can be an absolute value or a percentage, used to quantify the degree of deviation between the actual and planned discharge.

[0151] Finally, the opening of the discharge ports is adjusted based on the coal spillage risk and flow deviation. The predicted coal spillage risk and the real-time calculated flow deviation can be comprehensively considered to finely adjust the opening of the dual discharge ports. For example, when the spillage risk is high, the opening may need to be reduced; when the flow deviation shows that the actual flow is lower than expected, the opening may need to be increased to compensate. This adjustment can be linear or nonlinear based on PID control or other advanced control algorithms to achieve more precise flow control and stockpiling management. After adjusting the discharge port opening, coal is discharged from the dual discharge ports. The actuators of the discharge ports (e.g., electric gates, pneumatic valves, etc.) can be driven to change their opening according to the adjusted opening command, thereby achieving precise control of the coal discharge flow rate.

[0152] To illustrate this technical solution more clearly, a specific example is used below. Suppose a coal silo needs to load a batch of lignite with a high moisture content. First, the system uses sensors to acquire the material characteristics of the lignite, such as moisture content and particle size. Based on these characteristics, the system selects or dynamically calculates a mapping relationship between the discharge port opening and flow rate applicable to this lignite from a pre-set database. For example, for lignite with high moisture content, its flowability may be poor, and the flow rate at the same opening will be lower than that of dry bituminous coal. During loading, the system continuously monitors the actual flow rate of both discharge ports and calculates the expected discharge flow rate based on the current discharge port opening and the established mapping relationship. If the actual flow rate of a discharge port is detected to be lower than the expected discharge flow rate, the system calculates a positive flow deviation. Simultaneously, the overflow risk prediction module predicts the current coal overflow risk based on the actual available space distribution inside the silo and the local average material accumulation rate. For example, if a high risk of overflow is predicted on the left side of the truck bed, and the actual flow rate at the left discharge port is lower than expected, the system will combine these two pieces of information and fine-tune the opening of the left discharge port. For instance, it might slightly increase the opening to compensate for insufficient flow, without increasing the overflow risk; or, if the overflow risk is extremely high, it might prioritize decreasing the opening to avoid overflow, even if this means a further reduction in flow rate. In this way, the system can dynamically and precisely adjust the opening of the discharge port based on the real-time characteristics of the material and the actual discharge situation, thereby minimizing coal overflow while ensuring loading efficiency.

[0153] Through the above technical solution, this embodiment, by considering material characteristics and establishing a dynamic flow mapping relationship, enables the system to better adapt to different types or states of coal, improving the accuracy of feed flow prediction. Simultaneously, the introduction of actual flow monitoring and flow deviation calculation provides a real-time feedback loop for the control system, allowing it to promptly detect and correct deviations during the feeding process, effectively avoiding flow fluctuations caused by external disturbances or changes in internal parameters. Therefore, when the risk of coal spillage is predicted, the system can not only make macroscopic adjustments based on the risk level but also perform microscopic corrections based on real-time flow deviations, significantly improving the precision and response speed of the feed opening adjustment, effectively reducing the risk of coal spillage, and ensuring loading efficiency and quality.

[0154] In some embodiments, step S105, controlling the coal feeding from both feed ports according to the risk of coal spillage, may include, but is not limited to, the following steps:

[0155] Step S201: Determine the opening adjustment amount for each feed port based on the coal spillage risk;

[0156] Step S202: Analyze the mutual influence between each discharge port and the material accumulation inside the car, and obtain the mutual influence coefficient;

[0157] Step S203: Based on the mutual influence coefficient, correct the opening adjustment amount corresponding to each feed port;

[0158] Step S204: Adjust the opening of each feed port according to the corrected opening adjustment amount, and control the coal feeding through the double feed ports.

[0159] In some embodiments, simple opening adjustments are made based solely on the overall or local overflow risk, without fully considering the potential mutual influence between the coal discharged from the two feed ports during simultaneous discharge, as the coal accumulates inside the car. This mutual influence may lead to uneven material accumulation, or even unexpected rapid accumulation in localized areas, thereby increasing the actual overflow risk or resulting in low loading efficiency, potentially preventing accurate, efficient, and safe coal loading.

[0160] To address this, the opening adjustment amount for each feed port can be determined based on the coal spill risk. For example, the required opening change for each feed port can be initially calculated based on the system's predicted coal spill risk, aiming to reduce or eliminate potential spill risks. For instance, if a high coal spill risk is predicted for the left side of the car, it might be initially determined that the opening of the left feed port needs to be reduced.

[0161] Then, the mutual influence between each discharge port and the material accumulation inside the car is analyzed to obtain the mutual influence coefficient. When coal is discharged simultaneously through two discharge ports, the accumulation pattern and diffusion range of coal falling from one discharge port may affect the accumulation area of ​​coal falling from the other discharge port, and vice versa. This interaction may cause the actual accumulation pattern of material inside the car to deviate from the predicted accumulation pattern when considering the discharge situation of each discharge port individually. The mutual influence coefficient aims to quantify the degree and direction of this interaction, for example, the degree of influence of discharge from one discharge port on the accumulation height of the area near the other discharge port.

[0162] Then, based on the mutual influence coefficient, the opening adjustment amount corresponding to each discharge port is corrected. After initially determining the opening adjustment amount for each discharge port, these opening adjustment amounts can be further optimized using the obtained mutual influence coefficient. For example, if the analysis finds that reducing the opening of the left discharge port will lead to a faster material accumulation rate in the right discharge port area, then the opening adjustment amount of the right discharge port may be appropriately adjusted during the correction to avoid new local overflow risks or maintain the overall accumulation balance.

[0163] Finally, based on the corrected opening adjustment amount, the opening of each feed port is adjusted, and the coal is fed through both feed ports. The more precise opening adjustment amount, corrected by the mutual influence coefficient, can be applied to the actual feed port control, thereby achieving coordinated operation of the two feed ports and ensuring uniform and stable coal accumulation inside the car.

[0164] To illustrate this technical solution more clearly, a specific example is used below. Suppose that during loading, the system predicts a high risk of coal spillage in the middle area of ​​the car. Initial analysis might indicate that both the left and right feed ports need to have their openings reduced. However, further analysis reveals that if the left feed port opening is reduced too much, the diffusion range of its feed stream will shrink, leading to a slower accumulation rate in the area it originally covered. Meanwhile, coal from the right feed port may more easily diffuse into the left area, accelerating accumulation there and potentially causing new localized spillage risks. In this case, the calculated mutual influence coefficient indicates that reducing the left feed port opening promotes accumulation in the right feed port area, and vice versa. Therefore, when correcting the opening adjustment, the reduction in the opening of the left discharge port may be fine-tuned, and the reduction in the opening of the right discharge port may also be adjusted accordingly. For example, the reduction in the right discharge port may need to be larger than initially calculated to offset the effect of the adjustment of the left discharge port, thereby ensuring that the material accumulation in the middle of the carriage can be more stable and uniform under the coordinated action of the two discharge ports, ultimately achieving precise control and avoiding overflow.

[0165] Through the above technical solution, this embodiment, by considering the mutual influence between the discharge ports, can effectively avoid uneven local material accumulation or accidental rapid accumulation caused by the interaction between the discharge ports, thereby significantly reducing the risk of coal spillage. Furthermore, this refined control also helps improve the uniformity and efficiency of loading, ensuring that the car body can be fully and safely loaded, thus enhancing the intelligence and automation level of the entire coal silo loading process.

[0166] In some embodiments, step S202 involves analyzing the mutual influence between each discharge port and the material accumulation inside the carriage to obtain a mutual influence coefficient, which may include, but is not limited to, the following steps:

[0167] Acquire surface morphology data of materials already piled up inside the carriage;

[0168] Based on surface morphology data, identify local accumulation features inside the carriage, including slope and pits;

[0169] Based on the local accumulation characteristics, calculate the diffusion path and landing point distribution of the feed stream;

[0170] Based on the diffusion path and landing point distribution, the mutual influence between each discharge port and the material accumulation inside the carriage is analyzed, and the mutual influence coefficient is obtained.

[0171] In some embodiments, surface morphology data of the materials already piled up inside the car is first acquired. For example, this can be collected using LiDAR, 3D vision sensors, or other suitable depth sensing devices. This data can accurately reflect the actual three-dimensional shape of the coal pile currently inside the car.

[0172] Then, based on the surface morphology data, local accumulation features inside the carriage are identified. These local accumulation features refer to areas on the material accumulation surface inside the carriage that have significant geometric characteristics, including slope and depressions. Slope refers to the degree of inclination of the material accumulation surface relative to the horizontal plane, and depressions refer to local low-lying areas on the material accumulation surface that are lower than the surrounding area. These features can be identified through geometric analysis and feature extraction algorithms on the surface morphology data. For example, gradient calculation and curvature analysis can be used to quantify the slope, and region growing and connected component analysis can be used to identify depressions.

[0173] Then, based on the local accumulation characteristics, the diffusion path and landing point distribution of the feed bundle are calculated. The diffusion path refers to the trajectory of the coal particles in space after they fall from the feed inlet, under the influence of gravity, airflow, and other factors. The landing point distribution refers to the area and density distribution of the coal particles that ultimately land on the material accumulation surface inside the car. These parameters can be calculated based on the physical properties of the material (such as particle size, density, and coefficient of friction), the height and opening of the feed inlet, and the airflow conditions inside the car, and can be predicted using physical simulation models (such as Discrete Element Method (DEM)) or empirical formulas.

[0174] Finally, based on the diffusion path and landing point distribution, the mutual influence between each discharge port and the material accumulation inside the car was analyzed, and the mutual influence coefficient was obtained. The mutual influence coefficient is an indicator that quantifies the impact of different discharge ports on the material accumulation pattern inside the car during the discharge process. This coefficient can reflect the influence of the material flow from one discharge port on the accumulation height, slope, or degree of pit filling in the area near another discharge port.

[0175] Through the above technical solution, this embodiment can more accurately quantify the mutual influence of material accumulation during the dual-feeding-port process. By identifying local accumulation characteristics and calculating the diffusion path and landing point distribution of the feed stream, the accumulation behavior of materials inside the car can be predicted more accurately, especially the complex accumulation patterns that may occur when multiple feed ports are operating simultaneously. This helps avoid problems such as excessive local accumulation or underutilization of space due to inaccurate estimation of mutual influence, thereby improving loading efficiency and uniformity, and effectively reducing the risk of coal spillage.

[0176] In some embodiments, step S105, controlling the coal feeding from both feed ports according to the risk of coal spillage, may include, but is not limited to, the following steps:

[0177] Obtain information on overflow risk at the end of loading, including remaining available space and local stacking height;

[0178] If the remaining available space is less than the preset space threshold and the local stacking height is greater than the preset critical value, then the material stacking trend is calculated based on the local instantaneous material stacking rate.

[0179] Based on the risk of coal spillage and the trend of material accumulation, the opening of the discharge port is adjusted to reduce the step size of the discharge port opening adjustment and increase the frequency of the discharge port opening adjustment.

[0180] After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

[0181] In some embodiments, as the available space inside the car body decreases significantly towards the end of loading, and the material accumulation height approaches its upper limit, using conventional material feeding control strategies at this stage may lead to coal spillage due to insufficient control precision or untimely response, affecting loading efficiency and safety. Therefore, spillage risk information at the end of loading can be obtained beforehand. This information refers to a specific set of data used to assess the risk of coal spillage near the end of the loading process. The spillage risk information includes remaining available space and local accumulation height. Remaining available space refers to the volume of space inside the car body that has not yet been filled with coal, which can be calculated by scanning the inside of the car body using sensors (e.g., lidar, ultrasonic sensors) and combining this data with a car body model. Local accumulation height refers to the coal accumulation height in a specific area inside the car body, which can also be obtained through real-time monitoring using sensors.

[0182] If the remaining available space is less than a preset space threshold, and the local stacking height is greater than a preset critical value, the material stacking trend is calculated based on the local instantaneous material stacking rate. The preset space threshold is a pre-set space quantity; when the remaining available space is below this threshold, it indicates that loading has entered its final stage. The preset critical value is a pre-set height value; when the local stacking height exceeds this value, it indicates that the stacking in that area is approaching full capacity. The local instantaneous material stacking rate refers to the rate of change of the coal stacking height in a specific area inside the car at a given moment; it can be obtained by continuously monitoring the local stacking height and performing time-difference calculations. The material stacking trend refers to the direction and magnitude of the coal stacking height change over a future period predicted based on the local instantaneous material stacking rate, such as whether it is accelerating, decelerating, or stable stacking.

[0183] Based on the risk of coal spillage and the trend of material accumulation, the opening of the discharge port is adjusted to reduce the step size of the adjustment and increase the frequency of adjustment. After adjusting the discharge port opening, coal is discharged through both discharge ports. Reducing the step size of the discharge port opening adjustment means that the change in the opening size is smaller each time it is adjusted, thus achieving more precise control. Increasing the frequency of discharge port opening adjustment means that the system will detect and adjust the discharge port opening more frequently to respond more quickly to changes in material accumulation.

[0184] To illustrate this technical solution more clearly, a specific example is used below. Assume a coal silo is being loaded into a car. When the loading process is approximately 95% complete, the system uses sensors to determine that the remaining usable space inside the car is 0.5 cubic meters, which is less than a preset space threshold (e.g., 1 cubic meter). Simultaneously, the coal accumulation height in a certain local area of ​​the car is 3.8 meters, which is greater than a preset critical value (e.g., 3.5 meters). At this point, the system determines that loading is nearing completion and the risk of overflow is extremely high. To avoid overflow, the system immediately activates a refined control mode. Specifically, the system continuously monitors the coal accumulation height in that local area and calculates the local instantaneous material accumulation rate, for example, 0.01 meters per second. Based on this, the system predicts that the material accumulation trend will continue to rise. According to the current predicted coal overflow risk and material accumulation trend, the system no longer uses the conventional discharge port opening adjustment strategy. Instead, it reduces the adjustment step size from the usual 5% to 1%, while increasing the adjustment frequency from once every 5 seconds to once every 1 second. This means the system will fine-tune the opening of the dual discharge ports more frequently and with smaller amplitudes. For example, when a slight increase in local accumulation height is detected, the opening of the discharge ports will be immediately reduced by 1%, and then reassessed and adjusted again in the next second. In this way, even when the car is about to be fully loaded, the system can accurately control the amount of coal discharged, effectively preventing coal spillage and ensuring a smooth and efficient loading process.

[0185] Through the above technical solution, this embodiment enables more precise and real-time control of the dual discharge ports during the final stage of coal silo loading, a critical phase characterized by extremely limited usable space inside the car and a high risk of overflow. This embodiment effectively avoids coal overflow caused by control lag or excessive adjustments in critical conditions, significantly improving the safety of the loading process. Simultaneously, by optimizing the step size and frequency of discharge port opening adjustments, it ensures that the remaining space in the car is utilized to the maximum extent possible without overflow, thereby improving overall loading efficiency and material utilization.

[0186] In some embodiments, step S105, controlling the coal feeding from both feed ports according to the risk of coal spillage, may include, but is not limited to, the following steps:

[0187] Obtain material stacking characteristics;

[0188] The changing trends of material packing characteristics are analyzed to obtain the changing trends of material packing characteristics;

[0189] Determine the material diffusion parameters based on the changing trends of material accumulation characteristics;

[0190] Based on material diffusion parameters and material accumulation characteristics, the diffusion range is predicted using a material diffusion model;

[0191] Adjust the opening of the discharge port according to the diffusion range;

[0192] After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

[0193] In some embodiments, controlling spillover risks based solely on macroscopic factors may not adequately account for the dynamic physical characteristics of coal accumulation and diffusion within the car, potentially leading to uneven local accumulation, low space utilization, or unpredictable local spillover risks at the end of loading. To address this, material accumulation characteristics can be obtained beforehand. For example, physical properties of coal during the accumulation process can be acquired, such as the angle of repose, particle size distribution, density, moisture content, and coefficient of friction. These characteristics can be estimated through prior laboratory testing, real-time monitoring by online sensors, or by combining historical data.

[0194] Then, the changing trends of the material's stacking characteristics are analyzed to identify dynamic changes in material properties during loading, such as fluctuations caused by vibration, humidity variations, or mixing of different batches of material. This analysis can be achieved through time series analysis or statistical analysis of continuously collected material property data.

[0195] Next, based on the changing trends of material accumulation characteristics, material diffusion parameters are determined. These parameters, such as the diffusion coefficient and the rate of change of accumulation slope, can be updated in real-time or periodically based on the dynamic changes in material properties. These parameters can be predicted using empirical formulas, lookup tables, or machine learning models. Simultaneously, based on the material diffusion parameters and material accumulation characteristics, the diffusion range is predicted using a material diffusion model. Numerical simulations or physical models can be used to predict the accumulation morphology and diffusion boundary of coal inside the car after it falls from the feed port. Material diffusion models can include discrete element method (DEM), cellular automata models, or models based on continuum mechanics. Their purpose is to accurately simulate the coal accumulation process, thereby obtaining the expected accumulation height and horizontal diffusion area of ​​coal inside the car.

[0196] Finally, the opening of the discharge port is adjusted according to the diffusion range, and coal is discharged through both discharge ports after adjustment. The opening of each discharge port can be dynamically adjusted based on the matching between the coal accumulation range predicted by the material diffusion model and the actual available space inside the car. For example, if the predicted diffusion range is too narrow, it may lead to excessive local accumulation, so the discharge port opening can be appropriately increased or the discharge port position adjusted to promote more uniform coal diffusion; if the predicted diffusion range is too wide, it may lead to coal overflowing at the edge of the car, so the discharge port opening can be appropriately decreased. After adjusting the discharge port opening, coal is discharged through both discharge ports to ensure that the adjusted discharge strategy is implemented.

[0197] To illustrate this technical solution more clearly, a specific example is used below. Suppose that during coal loading, the system detects in real-time through sensors that the moisture content of the loaded coal has increased. This leads to an increase in the angle of repose in the material's packing characteristics, meaning the coal is more likely to form a steep packing slope and a reduced diffusion range. The system first acquires this material packing characteristic of increased moisture content and analyzes its trend, identifying the increasing angle of repose. Based on this trend, the system determines new material diffusion parameters and uses a material diffusion model to predict that, at the current discharge port opening, the coal will form a higher and more concentrated packing area in the middle of the car, thus increasing the risk of local overflow. To address this prediction, the system proactively adjusts the opening of the dual discharge ports according to the predicted diffusion range. For example, it can appropriately increase the discharge port opening or adjust the discharge direction to promote more uniform coal diffusion within the car and avoid excessive local packing. After adjusting the opening of the feed inlet, the system continues to control the dual feed inlets to feed coal, thereby maintaining the uniformity of coal accumulation throughout the loading process and effectively preventing local overflow.

[0198] Through the above technical solution, this embodiment can accurately predict the diffusion range of coal, and the system can effectively avoid local overload and overflow, thereby reducing material loss and improving loading safety. Furthermore, by optimizing the coal accumulation pattern inside the car, the effective volume utilization rate of the car can be maximized, ensuring optimal results for each loading, thereby reducing operating costs.

[0199] The beneficial effects of implementing the embodiments of this application include: the embodiments of this application first obtain local spatial distribution data and material accumulation surface data inside the car, then calculate the actual usable space distribution inside the car based on the local spatial distribution data, and calculate the local average material accumulation rate based on the material accumulation surface data, then predict the coal overflow risk based on the actual usable space distribution and the local average material accumulation rate, and finally control the coal discharge from the dual discharge ports based on the coal overflow risk. Thus, it is possible to combine spatial distribution and accumulation data to predict overflow risk, thereby achieving loading control and improving control accuracy and safety.

[0200] like Figure 2 As shown in the figure, this application also provides a coal silo loading control system, including:

[0201] The data acquisition module 301 is used to acquire local spatial distribution data and material accumulation surface data inside the carriage;

[0202] The spatial distribution calculation module 302 is used to calculate the actual usable space distribution inside the carriage based on local spatial distribution data.

[0203] The stacking rate calculation module 303 is used to calculate the local average material stacking rate based on the material stacking surface data.

[0204] The spillover risk prediction module 304 is used to predict coal spillover risk based on the actual available space distribution and local average material accumulation rate.

[0205] The feeding control module 305 is used to control the feeding of coal from the dual feeding ports based on the risk of coal spillage.

[0206] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0207] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

Claims

1. A method for controlling loading coal silos, characterized in that, Includes the following steps: Acquire local spatial distribution data and material accumulation surface data inside the carriage; Based on the local spatial distribution data, the actual usable space distribution inside the carriage is calculated; Calculate the local average material accumulation rate based on the material accumulation surface data; Based on the actual available space distribution and the local average material accumulation rate, the risk of coal spillage is predicted. Based on the aforementioned coal spillage risk, coal feeding is controlled through both feed ports.

2. The coal silo loading control method according to claim 1, characterized in that, The calculation of the local average material accumulation rate based on the material accumulation surface data includes: Based on the material accumulation surface data, identify local suspended coal dust layer areas inside the carriage; Radar scanning was performed on the local suspended coal dust layer area to obtain radar echo data before airflow disturbance. When injecting controlled airflow into the local suspended coal dust layer region, radar echo data of airflow disturbance is collected; After injecting controlled airflow into the local suspended coal dust layer area, radar echo data after airflow disturbance is collected; Based on the radar echo data before the airflow disturbance, the radar echo data during the airflow disturbance, and the radar echo data after the airflow disturbance, differential analysis is performed to calculate the radar sensing height difference. If the difference in radar sensing height is greater than a preset stability threshold, the thickness of the suspended coal dust layer is calculated based on the radar echo data before the airflow disturbance and the radar echo data after the airflow disturbance. The height of the solid coal accumulation surface is calculated based on the thickness of the suspended coal dust layer and the current radar sensing height. The local average material accumulation rate is calculated based on the surface height of the solid coal accumulation.

3. The coal silo loading control method according to claim 1, characterized in that, The method of predicting coal spill risk based on the actual available space distribution and the local average material accumulation rate includes: Acquire vehicle attitude data and vehicle position data; Calculate the carriage posture change based on the vehicle attitude data and the vehicle position data; Based on the change in the carriage's posture, the coordinate transformation of the material accumulation surface data is performed to correct the position of the material accumulation surface data in the carriage coordinate system; Calculate the relative positional relationships based on the actual available spatial distribution and the material accumulation surface data after coordinate transformation; If the relative positional relationship is greater than the preset positional deviation threshold, then the material accumulation surface data after coordinate transformation is locally re-matched according to the actual available space distribution. Based on the actual available space distribution, the local average material accumulation rate, and the material accumulation surface data after local spatial rematching, the risk of coal spillage is predicted.

4. The coal silo loading control method according to claim 1, characterized in that, The method of predicting coal spill risk based on the actual available space distribution and the local average material accumulation rate includes: Collect the local material accumulation height; Calculate the local instantaneous material accumulation rate based on the local material accumulation height; The accumulation rate deviation is calculated based on the local instantaneous material accumulation rate and the local average material accumulation rate. Based on the stacking rate deviation and the changing trend of the discharge port opening, the local average material stacking rate is corrected to obtain the target material stacking rate. Based on the target material accumulation rate and the actual available space distribution, the risk of coal spillage is predicted.

5. The coal silo loading control method according to claim 1, characterized in that, The method of controlling coal feeding through both feed ports based on the coal spillage risk includes: Obtain material properties; Based on the material characteristics, calculate the mapping relationship between the discharge port opening and the coal discharge flow rate; Monitor the actual flow rate at the feed inlet; Calculate the expected discharge flow rate based on the discharge port opening and the mapping relationship; Calculate the flow deviation based on the actual flow rate at the discharge port and the expected discharge flow rate; The opening of the feed inlet is adjusted based on the coal spillage risk and the flow deviation. After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

6. The coal silo loading control method according to claim 1, characterized in that, The method of controlling coal feeding through both feed ports based on the coal spillage risk includes: Based on the coal spillage risk, determine the opening adjustment amount corresponding to each feed port; The mutual influence between each discharge port and the material accumulation inside the car is analyzed to obtain the mutual influence coefficient. Based on the mutual influence coefficient, the opening adjustment amount corresponding to each feed port is corrected; Based on the corrected opening adjustment amount, adjust the opening of each feed port and control the coal feeding through both feed ports.

7. The coal silo loading control method according to claim 6, characterized in that, The analysis of the mutual influence between each discharge port and the material accumulation inside the carriage yields a mutual influence coefficient, including: Acquire surface morphology data of materials already piled up inside the carriage; Based on the surface morphology data, local accumulation features inside the carriage are identified, including slopes and pits; Based on the local accumulation characteristics, calculate the diffusion path and landing point distribution of the feed bundle; Based on the diffusion path and the distribution of landing points, the mutual influence between each discharge port and the material accumulation inside the carriage is analyzed to obtain the mutual influence coefficient.

8. The coal silo loading control method according to claim 1, characterized in that, The method of controlling coal feeding through both feed ports based on the coal spillage risk includes: Obtain end-of-loading overflow risk information, which includes remaining available space and local stacking height; If the remaining available space is less than a preset space threshold and the local stacking height is greater than a preset critical value, then the material stacking trend is calculated based on the local instantaneous material stacking rate. Based on the coal spillage risk and the material accumulation trend, the discharge port opening is adjusted to reduce the step size of the discharge port opening adjustment and increase the frequency of the discharge port opening adjustment. After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

9. The coal silo loading control method according to claim 1, characterized in that, The method of controlling coal feeding through both feed ports based on the coal spillage risk includes: Obtain material stacking characteristics; The changing trend of the material's packing characteristics is analyzed to obtain the changing trend of the material's packing characteristics; Based on the changing trend of the material's stacking characteristics, determine the material diffusion parameters; Based on the material diffusion parameters and the material accumulation characteristics, the diffusion range is predicted using a material diffusion model. The opening of the discharge port is adjusted according to the diffusion range. After adjusting the opening of the feed inlet, control the coal feeding through both feed inlets.

10. A coal silo loading control system, characterized in that, include: The data acquisition module is used to acquire local spatial distribution data and material accumulation surface data inside the carriage; The spatial distribution calculation module is used to calculate the actual usable space distribution inside the carriage based on the local spatial distribution data. The accumulation rate calculation module is used to calculate the local average material accumulation rate based on the material accumulation surface data. The spillover risk prediction module is used to predict the coal spillover risk based on the actual available space distribution and the local average material accumulation rate. The material feeding control module is used to control the coal feeding from the dual feeding ports based on the coal overflow risk.

Images