Three-dimensional intelligent coal level monitoring method for coal feeder silos
By using millimeter-wave radar sensors and a three-dimensional digital twin model inside the coal bunker to segment the coal surface into micro-graphics and calculate the coal volume, the inaccuracy problem of detecting the undulations of the coal body inside the bunker was solved, achieving high-precision and low-cost real-time monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- YANKUANG ENERGY GRP CO LTD
- Filing Date
- 2026-05-12
- Publication Date
- 2026-06-30
AI Technical Summary
Existing technologies cannot accurately quantify the volume of coal with varying surface heights within a coal bunker in real time. Especially in the environment of coal blocks and coal dust within the bunker, the detection device is prone to damage or failure, resulting in poor detection accuracy and high maintenance costs.
A three-dimensional digital twin model is used in conjunction with a millimeter-wave radar sensor. A three-dimensional rectangular coordinate system is established in the coal bunker using calculus. The surface of the coal body is divided into micro-figures, the coordinates of each point are recorded, and the volume of the coal body is calculated. Real-time monitoring is performed using a multi-channel FMCW millimeter-wave radar probe.
It improves the accuracy and real-time performance of coal level detection, adapts to complex working conditions in coal bunkers, reduces maintenance costs, and is suitable for coal feeder silos in major coal mines and thermal power plants.
Smart Images
Figure CN122306189A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a three-dimensional intelligent coal level monitoring method for a coal feeder hopper. Background Technology
[0002] Coal bunkers are the central distribution facilities for coal products in mines, and their coal level status is crucial to the operation of multiple systems, including production, transportation, sales, and safety. Real-time monitoring of the coal level is therefore of paramount importance. Current methods for detecting coal levels in coal bunkers vary. For example, ultrasonic testing, laser ranging, and level radar are used, with testing devices installed inside the bunker to detect the coal level at the bottom. However, due to the presence of coal chunks, coal dust, and water mist within the bunker, these devices are easily damaged by mechanical impacts or rendered ineffective by dust accumulation and blockage, making them unsuitable for the harsh conditions of coal bunkers in mines. Using a plumb bob detection method is prone to mechanical failure due to coal impacts, resulting in high maintenance costs and unreliable real-time performance. Detecting coal level changes based on the frequency of the sound produced by falling coal has poor accuracy. Therefore, it is essential to develop a cost-effective coal level detection method and system suitable for the harsh and complex conditions of coal bunkers in mines.
[0003] Chinese invention patent application CN117073792A, authorized on November 17, 2023, discloses a coal bunker coal level detection system and method. The system includes two belt conveyors installed at the coal bunker inlet and outlet, and an information processing module. The belt conveyors are equipped with an image acquisition module, a thickness measurement module, and a speed measuring device. The method involves obtaining the thickness and area information of the coal flow and the conveyor speed through the aforementioned modules to obtain the amount of coal entering and leaving the bunker. The difference between these two quantities is used to calculate the amount of coal in the bunker, and then the coal level height is calculated based on the shape of the bunker's interior. This invention calculates the change in coal level within the bunker based on the difference between the coal inflow (Vin) and the coal outflow (Vout) in conjunction with the shape of the bunker's interior, which is prone to cumulative errors, leading to inaccurate coal level detection.
[0004] Chinese invention patent CN114581619B, authorized on July 29, 2025, discloses a coal bunker modeling method based on three-dimensional positioning and two-dimensional mapping. To accelerate the efficiency of cleaning the coal bunker, the method utilizes sensors to acquire information about the interior of the coal bunker, and uses three-dimensional lidar information and an inertial measurement unit to perform real-time positioning of the pusher machine performing the cleaning operation. A fixed rotation sensor composed of a single-line radar and a DC servo motor is used to perceive the coal bunker environment, and the front-end positioning and serial port rotation angle are used as the basis for single-frame matching to obtain a complete point cloud of the interior of the coal bunker. The superimposed point cloud is semantically fused and segmented to classify the coal material point cloud, complete the outer boundary of the point cloud, fill in the internal empty point cloud, and calculate the coal volume based on the three-dimensional grid point cloud of the coal material.
[0005] Chinese invention patent application CN120891505A, published on November 4, 2025, discloses a method, apparatus, equipment, and detection system for detecting the volume of materials in a quantitative silo. The method involves acquiring at least two first point cloud information collected by at least two lidar sensors; fusing the at least two first point cloud information to obtain fused point cloud information of the material in the quantitative silo; and determining the volume information of the material in the quantitative silo based on the fused point cloud information.
[0006] Both of the above patented technologies use 3D modeling to establish a solid model of the coal inside the coal bunker, and use this solid model to determine the total amount of coal in the bunker. However, the coal inside the bunker has limited fluidity, and its surface is uneven. As coal is continuously fed into the bunker and discharged from the outlet, the surface of the coal changes constantly, making it difficult to quantify and detect the volume of coal with uneven surface in real time. Summary of the Invention
[0007] The technical problem to be solved by the present invention is how to overcome the above-mentioned defects of the prior art and provide a three-dimensional intelligent coal level monitoring method for coal feeder hoppers that is suitable for real-time quantitative detection of the volume of coal with surface undulations in coal bunkers.
[0008] To solve the above technical problems, this coal feeder hopper uses a three-dimensional intelligent coal level monitoring method, which includes the following steps: (1) Establish a coal feeder silo that is easy to monitor in real time. The coal feeder silo includes a coal silo, multiple millimeter-wave radar sensors, a PLC controller, a host computer, and an alarm. The coal silo includes a silo top, a cylinder, and a cone bucket connected sequentially from top to bottom. The silo top is provided with a feed inlet, and the bottom of the cone bucket is provided with a discharge outlet with a screw feeder. The millimeter-wave radar sensors are installed inside the coal silo and fixed on the silo top. The millimeter-wave radar sensors, PLC controller, host computer, and alarm are electrically connected. (2) Based on the design shape of the coal feeder silo for easy real-time monitoring, a three-dimensional digital twin model of the coal feeder silo for easy real-time monitoring is made using geometric methods; (3) On the three-dimensional digital twin model described in step (2), find the cross-sectional view of the cylinder corresponding to the reference horizontal plane of the millimeter-wave radar sensor. With the center point of the cross-sectional view as the zero point, establish a three-dimensional rectangular coordinate system. Divide the inner wall region of the cylinder in the cross-sectional view into N micro-figures with the same area S. Record the coordinates of the center point of each micro-figure: (X1, Y1, 0), (X2, Y2, 0), (X3, Y3, 0)...(Xn, Yn, 0). In this step, N is a positive integer, and the subscript numbers of X and Y in each parenthesis represent their serial numbers. (4) On the three-dimensional digital twin model, find the intersection of the vertical line extending downward from the center point of each micro-graphic and the plane of the inner surface of the cone or the lower edge of the discharge port, and record their coordinates respectively: (X1, Y1, L1), (X2, Y2, L2), (X3, Y3, L3)……(Xn, Yn, Ln). In this step, the subscript numbers of X, Y, and L in each parenthesis represent their serial numbers. (5) Put the coal feeder hopper, which is easy to monitor in real time, into use. Using the millimeter-wave radar sensor installed in step (1), collect the reflection point cloud dataset of the upper surface of the coal body and the exposed inner surface of the cone hopper below the millimeter-wave radar sensor. Filter out the outliers in the reflection point cloud dataset to obtain the point cloud data with the outliers removed. Then, on the three-dimensional digital twin model described in step (2), establish the same three-dimensional rectangular coordinate system according to the zero point, orientation, and unit described in step (3) to build a three-dimensional digital twin model of the upper surface of the coal body. Find and record the points on the three-dimensional digital twin model of the upper surface of the coal body that correspond to the center points of each micro-graphic described in step (3): (X1, Y1, Z1), (X2, Y2, Z2), (X3, Y3, Z3)...(X... n Y n Z n In this step, the subscript numbers of X, Y, and Z in each parenthesis represent their respective serial numbers; ⑹. Calculate the real-time volume of coal in the feeder hopper according to [(Z1-L1)+(Z2-L2)+(Z3-L3)……+(Zn-Ln)]×S, which is convenient for real-time monitoring. This design belongs to the calculus method. The smaller the area of the micro-figure, the higher the accuracy of the real-time detection of the coal volume.
[0009] As an optimization, the millimeter-wave radar sensor is a multi-channel FMCW millimeter-wave radar probe with an IP54-rated explosion-proof housing and a built-in self-cleaning probe, operating in an ambient temperature range of -40°C to +80°C. This design ensures high accuracy.
[0010] As an optimization, the micro-graphics are regular hexagonal shapes. This design facilitates calculation.
[0011] This invention relates to a three-dimensional intelligent coal level monitoring method for coal feeder silos. Utilizing calculus, a large number of cylinders with micro-graphic cross-sections of varying lengths are pieced together within a three-dimensional digital twin model of the coal silo to create a geometric body that matches the real-time coal surface within the silo. The total length of all cylinders is calculated, and multiplied by the area of the micro-graphics, yielding a more accurate estimate of the real-time coal volume. This method overcomes the adverse effects of uneven coal surface on coal level detection, significantly improving the accuracy of the coal level monitoring system. It is suitable for use in coal feeder silos in major coal mines and thermal power plants. Attached Figure Description
[0012] The following description, in conjunction with the accompanying drawings, further illustrates the three-dimensional intelligent coal level monitoring method for the coal feeder hopper of the present invention: Figure 1 This is a schematic diagram of the structure of the coal feeder silo, which uses a three-dimensional intelligent coal level monitoring method for easy real-time monitoring. Figure 2 It is a top view schematic diagram of dividing the inner wall region of the cylinder 6 of the three-dimensional digital twin model into N micro-figures with the same area S. Figure 3 It is a front view schematic diagram of the three-dimensional digital twin model and the coordinates of the center points of each micro-graphic and the intersection points of the center points of each micro-graphic and the vertical line in step (3); Figure 4 It is based on the point cloud data after filtering out outliers. Figure 3 The three-dimensional digital twin model shown is supplemented by a three-dimensional digital twin model 11 of the upper surface of the coal body, and labeled. Figure 3 A schematic diagram of the intersection of the vertical line shown above with the three-dimensional digital twin model of the upper surface of the coal body.
[0013] To keep the drawing simple, Figure 3 , Figure 4 The drawing only shows four perpendicular lines and four sets of center points and intersection coordinates. In reality, the number of perpendicular lines and sets of center points and intersections should correspond to the actual number of micro-figures that the cross-section of the cylinder can accommodate. Figure 3 , Figure 4 For illustrative purposes only.
[0014] In the figure: 1 is the millimeter-wave radar sensor, 2 is the PLC controller, 3 is the host computer, 4 is the alarm, 5 is the top of the silo, 6 is the cylinder, 7 is the cone bucket, 8 is the feed inlet, 9 is the discharge outlet, 10 is the zero point, 11 is the three-dimensional digital twin model of the upper surface of the coal body, P is the cross-section of the cylinder on the three-dimensional digital twin model corresponding to the reference horizontal plane of the millimeter-wave radar sensor 1, and 12 is the micro-graphics. Detailed Implementation
[0015] Implementation method one: such as Figure 1-4 As shown, the three-dimensional intelligent coal level monitoring method used in this coal feeder hopper includes the following steps: (1) Establish a coal feeder hopper for easy real-time monitoring. This hopper includes a coal bunker, multiple millimeter-wave radar sensors 1, a PLC controller 2, a host computer 3, and an alarm 4. The coal bunker comprises a bunker top 5, a cylinder 6, and a conical hopper 7 connected sequentially from top to bottom. The bunker top 5 has a feed inlet 8, and the bottom of the conical hopper 7 has a discharge outlet 9 with a screw feeder (not shown in the figure). The millimeter-wave radar sensors 1 are installed inside the coal bunker and fixed to the bunker top 5. The millimeter-wave radar sensors 1, PLC controller 2, host computer 3, and alarm 4 are electrically connected. Figure 1 As shown, this step prepares the necessary hardware equipment.
[0016] (2) Based on the design shape of the coal feeder hopper for easy real-time monitoring, a three-dimensional digital twin model of the coal feeder hopper for easy real-time monitoring is created using geometric methods, such as... Figure 3 As shown.
[0017] (3) On the three-dimensional digital twin model described in step (2), find the cross-sectional view P of the cylinder corresponding to the reference horizontal plane of the millimeter-wave radar sensor 1. Establish a three-dimensional rectangular coordinate system with the center point of the cross-sectional view P as the zero point 10. Divide the inner wall region of the cylinder in the cross-sectional view into a lattice of N micro-figures with the same area S. The micro-figures are preferably regular hexagonal figures. The top view of the lattice of the micro-figures is shown below. Figure 2 As shown.
[0018] Then, locate and record the coordinates of the center points of each micro-graphic on the 3D digital twin model: (X1, Y1, 0), (X2, Y2, 0), (X3, Y3, 0)...(Xn, Yn, 0). In this step, N is a positive integer, and the subscripts of X and Y within each parenthesis represent their ordinal numbers, such as... Figure 3 As shown.
[0019] (4) On the 3D digital twin model, find the intersection points of the vertical lines extending downwards from the center points of each micro-graphic and the inner surface of the cone hopper or the lower edge of the discharge port, and record their coordinates respectively: (X1, Y1, L1), (X2, Y2, L2), (X3, Y3, L3)...(Xn, Yn, Ln). In this step, the subscript numbers of X, Y, and L in each parenthesis represent their serial numbers, such as... Figure 3 As shown.
[0020] (5) Put the coal feeder hopper, which is easy to monitor in real time, into use. Using the millimeter-wave radar sensor installed in step (1), collect the reflection point cloud dataset of the upper surface of the coal body and the exposed inner surface of the cone hopper below the millimeter-wave radar sensor 1. Filter out the abnormal points in the reflection point cloud dataset to obtain the point cloud data with the filtered out abnormal points. Then, on the three-dimensional digital twin model described in step (2), establish the same three-dimensional rectangular coordinate system according to the zero point, orientation, and unit described in step (3), and establish a three-dimensional digital twin model 11 of the upper surface of the coal body. Find and record the points on the three-dimensional digital twin model of the upper surface of the coal body that correspond to the center points of each micro-graphic described in step (3): (X1, Y1, Z1), (X2, Y2, Z2), (X3, Y3, Z3)...(X... n Y n Z n In this step, the subscripts of X, Y, and Z within each parenthesis represent their sequence numbers; for example... Figure 4 As shown.
[0021] ⑹. Calculate the real-time volume of coal in the feeder bin according to [(Z1-L1)+(Z2-L2)+(Z3-L3)……+(Zn-Ln)]×S, which is convenient for real-time monitoring.
[0022] For example: S=1cm 2 =0.0001m 2
(Z1- L1)+(Z2- L2)+(Z3- L3)……+(Zn- Ln)
[0023] The millimeter-wave radar sensor 1 is a multi-channel FMCW millimeter-wave radar probe with an explosion-proof housing with IP54 protection rating and a built-in self-cleaning probe. The operating ambient temperature is -40°C to +80°C.
Claims
1. A three-dimensional intelligent coal level monitoring method for a coal feeder hopper, comprising the following steps: (1) Establish a coal feeder silo that is easy to monitor in real time. The coal feeder silo includes a coal silo, multiple millimeter-wave radar sensors, a PLC controller, a host computer, and an alarm. The coal silo includes a silo top, a cylinder, and a cone bucket connected sequentially from top to bottom. The silo top is provided with a feed inlet, and the bottom of the cone bucket is provided with a discharge outlet with a screw feeder. The millimeter-wave radar sensors are installed inside the coal silo and fixed on the silo top. The millimeter-wave radar sensors, PLC controller, host computer, and alarm are electrically connected. (2) Based on the design shape of the coal feeder silo for easy real-time monitoring, a three-dimensional digital twin model of the coal feeder silo for easy real-time monitoring is made using geometric methods; (3) On the three-dimensional digital twin model described in step (2), find the cross-sectional view of the cylinder corresponding to the reference horizontal plane of the millimeter-wave radar sensor. Establish a three-dimensional rectangular coordinate system with the center point of the cross-sectional view as the zero point. Divide the inner wall region of the cylinder in the cross-sectional view into N micro-figures with the same area S, and record the coordinates of the center points of each micro-figure: (X1, Y1, 0), (X2, Y2, 0), (X3, Y3, 0)...(X... n Y n ,0), where N is a positive integer, and the subscript numbers of X and Y in each parenthesis represent their ordinal numbers; (4) On the three-dimensional digital twin model, find the intersection points of the perpendicular lines extending downward from the center points of each micro-graphic and the inner surface of the cone hopper or the lower edge of the discharge port, and record their coordinates respectively: (X1, Y1, L1), (X2, Y2, L2), (X3, Y3, L3)...(X... n Y n L n In this step, the subscript numbers of X, Y, and L in each parenthesis represent their sequence numbers; (5) Put the coal feeder hopper, which is easy to monitor in real time, into use. Using the millimeter-wave radar sensor installed in step (1), collect the reflection point cloud dataset of the upper surface of the coal body and the exposed inner surface of the cone hopper below the millimeter-wave radar sensor. Filter out the outliers in the reflection point cloud dataset to obtain the point cloud data with the outliers removed. Then, on the three-dimensional digital twin model described in step (2), establish the same three-dimensional rectangular coordinate system according to the zero point, orientation, and unit described in step (3) to build a three-dimensional digital twin model of the upper surface of the coal body. Find and record the points on the three-dimensional digital twin model of the upper surface of the coal body that correspond to the center points of each micro-graphic described in step (3): (X1, Y1, Z1), (X2, Y2, Z2), (X3, Y3, Z3)...(X... n Y n Z n In this step, the subscript numbers of X, Y, and Z in each parenthesis represent their sequence numbers; ⑹. According to [(Z1-L1)+(Z2-L2)+(Z3-L3)……+(Z n - L n )】×S, to obtain the real-time volume of coal in the coal feeder hopper for easy real-time monitoring.
2. The three-dimensional intelligent coal level monitoring method for the coal feeder silo according to claim 1, characterized in that: The millimeter-wave radar sensor is a multi-channel FMCW millimeter-wave radar probe with an IP54-rated explosion-proof housing and a built-in self-cleaning probe. The operating ambient temperature is -40°C to +80°C.
3. The three-dimensional intelligent coal level monitoring method for the coal feeder silo according to claim 1 or 2, characterized in that: The micro-graphic is a regular hexagonal shape.