Coal yard coal discarding method and device based on multi-source data fusion

Abstract

This invention discloses a coal yard inventory method and apparatus based on multi-source data fusion, relating to the field of digital management technology for industrial bulk material storage yards. The method includes: deploying a fixed scanning network in the coal yard to perform a three-dimensional scan of the entire yard periodically or under triggered conditions, acquiring macroscopic basic point cloud data of the coal yard; analyzing this data using a central server to identify areas of coal pile change and scanning blind spots, delineating fine scanning targets and generating tasks; subsequently, dispatching a mobile inventory device to the target area to conduct fine scanning, simultaneously acquiring pose information to form a supplementary point cloud with spatial coordinates; the server unifies the coordinate system of the two types of point clouds and fuses them, reconstructing a three-dimensional model based on the fused point cloud, calculating the volume, and converting the coal weight into storage weight by combining the coal pile density, forming a complete inventory result. This invention achieves full-coverage, high-precision, automated coal inventory through the collaboration of fixed wide-area scanning and mobile supplementary scanning, solving the problems of blind spots, inefficiency, and high risk inherent in traditional coal inventory methods.

Description

Technical Field

[0001] This invention relates to the field of digital management technology for industrial bulk material storage yards, and in particular to a coal yard inventory method and apparatus based on multi-source data fusion. Background Technology

[0002] Regular coal inventory checks at coal yards are an indispensable and crucial part of the daily production and operation of large coal-consuming enterprises such as thermal power plants, steel mills, and ports. They directly affect the company's cost accounting, inventory management, production scheduling, and capital turnover efficiency. Accurate and efficient inventory results provide reliable data support for enterprises to formulate procurement plans, control fuel costs, and optimize inventory structure. They also effectively avoid production interruptions and capital waste caused by inventory accounting errors. Currently, there are many traditional inventory methods used in the industry, mainly including manual tape measure measurement, total station measurement, and vehicle-mounted laser inventory. All these methods have significant technical limitations and cannot meet the inventory needs of modern large-scale coal yards.

[0003] Among these methods, manual measurement with a measuring tape is the most traditional way to inventory coal. This method requires workers to carry measuring tapes and other tools deep into the coal yard to manually measure parameters such as the length, width, and height of the coal piles. They then use empirical formulas to estimate the volume of the coal piles. This method is not only extremely labor-intensive and inefficient, but also poses safety hazards due to vehicle traffic and potential coal pile collapses, making it difficult to effectively guarantee the personal safety of workers. Furthermore, manual estimation is greatly affected by human error and experience level, and the accuracy of coal inventory cannot meet the requirements of refined management in enterprises. While total station measurement, as a relatively accurate measurement method, can improve the accuracy of coal inventory to some extent, it still requires workers to enter the coal yard to set up measuring points and collect data. The measurement process is cumbersome and slow. For large coal yards with vast areas and numerous coal piles, this often requires a significant amount of manpower, resources, and time, failing to meet the actual needs of rapid inventory. The vehicle-mounted laser coal inventory method achieves semi-automation of coal inventory operations by equipping the working vehicle with laser scanning equipment, which improves the efficiency of coal inventory to a certain extent. However, the measurement accuracy of this method is easily affected by the bumps and vibrations during the vehicle's movement, which can lead to deviations in the scanning data. In addition, there are obvious measurement blind spots in places that the vehicle cannot reach, such as the top of the coal pile, edge corners, and areas obstructed by equipment, making it impossible to achieve full coverage measurement of the coal yard.

[0004] In recent years, to address the limitations of traditional coal inventory methods, a technical solution using fixed scanners has gradually emerged in the industry. This solution achieves automated, non-contact scanning of the coal yard by deploying scanning equipment around the perimeter, effectively improving inventory efficiency and operational safety. However, this solution suffers from high initial investment costs and difficulties in equipment deployment. Furthermore, in practical applications, large equipment such as stacker-reclaimers and transport vehicles within the coal yard can temporarily obstruct the scanning light, creating localized scanning blind spots. This results in incomplete point cloud data, leading to distortion of the 3D model of the coal pile and affecting the accuracy of the inventory results, failing to meet the core requirements of enterprises for inventory precision.

[0005] In summary, all existing coal inventory methods have their own technical shortcomings, such as low efficiency and poor safety, insufficient accuracy and blind spots, or excessive cost and limited practicality. The industry urgently needs a coal yard inventory solution that can balance high efficiency, high accuracy, full coverage and controllable cost to address the shortcomings of existing technologies and meet the actual needs of enterprises for refined management. Summary of the Invention

[0006] The main objective of this invention is to provide a coal yard inventory method based on multi-source data fusion.

[0007] Another objective of this invention is to propose a coal yard inventory device based on multi-source data fusion.

[0008] The third objective of this invention is to provide an electronic device.

[0009] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0010] To achieve the above objectives, a first aspect of the present invention proposes a coal yard inventory method based on multi-source data fusion, comprising:

[0011] By deploying a fixed scanning network at predetermined locations in the coal yard, periodic or triggered 3D scans are performed on the entire coal yard area to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard. The central processing unit receives and analyzes the basic point cloud data, identifies the coal pile morphology change area and the scanning blind spot of the fixed scanning network, determines the target area that needs to be scanned in detail, and generates the corresponding scanning task. Based on the scanning task, the scheduling is issued to control the mobile coal storage device to move to the target area, and the mobile scanning unit is used to perform fine scanning. At the same time, the positioning and orientation unit collects pose information in real time to obtain supplementary point cloud data with spatial location information. The basic point cloud data and the supplementary point cloud data are unified into a global coordinate system and fused. Based on the fused point cloud data, a three-dimensional model of the coal pile is reconstructed and the coal pile inventory is calculated. The coal pile weight is converted by combining the preset coal pile density data to form a complete coal yard inventory result.

[0012] Optionally, a fixed scanning network deployed at predetermined locations within the coal yard can be used to perform periodic or triggered 3D scans of the entire coal yard area to acquire basic point cloud data reflecting the macroscopic morphology of the coal yard, including: Multiple fixed scanning units are deployed at predetermined locations in the coal yard and on surrounding pillars and building structures to form a basic scanning network covering the entire coal yard area. All fixed scanning units are activated according to a preset time frequency or external trigger command to carry out periodic or trigger-based 3D scanning operations. Real-time spatial point information of the entire coal yard is collected using two-dimensional lidar, three-dimensional lidar, millimeter-wave radar or structured light three-dimensional scanner; The collected spatial point information is initially sorted out, obvious abnormal data is removed, and basic point cloud data that can clearly reflect the macroscopic shape of the coal yard, the overall outline of the coal pile and the site boundary is generated.

[0013] Optionally, the central processing unit receives and analyzes the basic point cloud data to identify areas of coal pile morphological change and blind spots of the fixed scanning network, determines the target area requiring refined scanning, and generates corresponding scanning tasks, including: The central processing server receives basic point cloud data and performs noise reduction and filtering preprocessing on the data to remove environmental interference and invalid point clouds. By comparing and analyzing the preprocessed current basic point cloud data with historical scan data from the same period, and by calculating the data difference, the morphological change areas of coal pile elevation change and contour offset are identified. Determine the scanning blind spots of the fixed scanning network caused by equipment deployment location limitations and coal pile obstruction, and determine the spatial range and coordinate information of the blind spots; Based on the morphological change area and the scanning blind zone, the fine scanning target area is delineated, the scanning range and accuracy requirements of each target area are determined, and the corresponding fine scanning task is generated.

[0014] Optionally, scheduling is performed based on the scanning task to control the mobile coal storage device to move to the target area, including: The central processing server distributes the detailed scanning task to the mobile coal storage device via a wireless communication link. Based on the issued detailed scanning task, the mobile coal storage device, which uses an unmanned vehicle, remote control vehicle or handheld device, is controlled to autonomously plan its path or accept remote control operation according to the task information and drive to the target area. The device provides real-time feedback on its location status during movement, ensuring that it accurately reaches the designated work location.

[0015] Optionally, a fine scanning process is performed using a moving scanning unit, while pose information is acquired in real time using a positioning and orientation unit to obtain supplementary point cloud data with spatial location information, including: Once the mobile coal stacking device reaches the target area, a lightweight laser scanner or depth camera is activated to conduct a high-density fine scan and obtain detailed structural information of the local coal pile. The mobile coal storage device uses a global navigation satellite system receiver and inertial measurement unit to collect the device's position, attitude, and heading information in real time. The fine scan data is bound with the corresponding pose information to generate supplementary point cloud data with accurate spatial location information, which is then uploaded to the central processing server.

[0016] Optionally, the basic point cloud data and the supplementary point cloud data are fused together in a global coordinate system. Based on the fused point cloud data, a three-dimensional model of the coal pile is reconstructed, and the coal pile inventory is calculated. Combined with preset coal pile density data, the coal pile weight is converted to form a complete coal yard inventory result, including: Using the coordinate system of the fixed scanning network as the global reference, the central processing server converts the supplementary point cloud data into a unified global coordinate system based on the pose information uploaded by the mobile coal storage device. A point cloud registration algorithm is used to match and stitch together the basic point cloud data and the supplementary point cloud data, eliminating duplicate and redundant data, and achieving seamless fusion of multi-source point clouds. A three-dimensional model of the coal pile is constructed based on the fused complete point cloud data, and the volume of the coal pile is calculated by triangulation algorithm or integral algorithm. By combining the pre-entered coal pile density data, the coal pile volume is converted into coal pile weight, and the three-dimensional model, volume and weight information are integrated to finally form a complete coal yard inventory result.

[0017] To achieve the above objectives, a second aspect of the present invention provides a coal yard inventory device based on multi-source data fusion, comprising: The global scanning module is used to perform periodic or triggered 3D scanning of the entire coal yard area through a fixed scanning network deployed at predetermined locations in the coal yard, and to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard. The blind spot identification module is used to receive and analyze the basic point cloud data through the central processor, identify the coal pile morphology change area and the scanning blind spot of the fixed scanning network, determine the target area that needs to be scanned in detail and generate the corresponding scanning task. The fine scanning module is used to issue and schedule based on scanning tasks, control the mobile coal storage device to move to the target area, use the mobile scanning unit to perform fine scanning, and at the same time collect pose information in real time through the positioning and orientation unit to obtain supplementary point cloud data with spatial location information. The data fusion module is used to unify the basic point cloud data and the supplementary point cloud data into a global coordinate system for fusion processing. Based on the fused point cloud data, a three-dimensional model of the coal pile is reconstructed and the coal pile inventory is calculated. Combined with the preset coal pile density data, the coal pile weight is converted to form a complete coal yard inventory result.

[0018] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0019] To achieve the above objectives, a third aspect of this application provides an electronic device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a coal yard coal inventory method based on multi-source data fusion as described in the first aspect embodiment.

[0020] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a coal yard inventory method based on multi-source data fusion as described in the first aspect embodiment.

[0021] The embodiments of the present invention have the following beneficial effects: 1. Full coverage and high precision: Fixed networks enable wide-area coverage and continuous monitoring, while mobile devices solve blind spots and fine measurement problems. The combination of the two ensures the integrity and accuracy of the 3D model.

[0022] 2. High efficiency and automation: Fixed scanning can operate automatically without human intervention. The mobile device automatically executes tasks after receiving instructions, significantly reducing manual operation time and intensity, and realizing the automation of the coal inventory process.

[0023] 3. Safety and Economy: Reduces the frequency of personnel entering high-risk work areas, improving safety management. The system is flexible in its construction, allowing for the configuration of fixed scanners based on the size and needs of the coal yard, thus reducing initial investment and operating costs.

[0024] 4. Maximizing data value: Continuous fixed scanning data is not only used for coal inventory, but also for monitoring the stacking and reclaiming process, analyzing the natural loss of coal piles, etc., providing comprehensive data support for the digital management of coal yards. Attached Figure Description

[0025] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart of a coal yard inventory method based on multi-source data fusion provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the overall architecture of the coal yard inventory system provided in an embodiment of the present invention; Figure 3 A schematic diagram of the on-site deployment of fixed and mobile coal storage systems in a closed coal yard, provided in an embodiment of the present invention; Figure 4 This is a structural diagram of a coal yard coal inventory device based on multi-source data fusion, provided as an embodiment of the present invention. Detailed Implementation

[0026] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] The following description, with reference to the accompanying drawings, describes a coal yard inventory method and apparatus based on multi-source data fusion according to an embodiment of the present invention.

[0029] Example 1 This invention provides a method for coal inventory in a coal yard based on multi-source data fusion. Figure 1 This is a flowchart illustrating a coal yard inventory method based on multi-source data fusion, provided as an embodiment of the present invention. Figure 1 As shown, the method includes the following steps: Step S1: By deploying a fixed scanning network at a predetermined location in the coal yard, periodic or triggered three-dimensional scans are performed on the entire coal yard area to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard.

[0030] S1, through a fixed scanning network deployed at predetermined locations in the coal yard, performs periodic or triggered 3D scanning of the entire coal yard area to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard.

[0031] In the embodiments of this application, such as Figure 2As shown, the overall system architecture of this application is divided into a perception layer, a network layer, and an application layer. The perception layer is responsible for raw data acquisition, and its core component is a fixed scanning network covering the entire coal yard area. This network consists of multiple sets of fixed scanning units deployed in the south and north coal yards, specifically including fixed scanning stations #1 and #2 in the south coal yard, and fixed scanning stations #1 and #2 in the north coal yard. Each fixed scanning unit is installed on pillars or building structures around the coal yard. Through reasonable visual arrangement and location planning, a basic scanning network covering the entire coal yard area without blind spots is formed, ensuring that the spatial morphological information of the entire coal yard area can be comprehensively collected.

[0032] In this embodiment, the fixed scanning unit can be one or more combinations of two-dimensional LiDAR, three-dimensional LiDAR, millimeter-wave radar, or structured light 3D scanner to adapt to the scanning needs of different coal yard environments. During actual operation, each fixed scanning unit initiates 3D scanning according to a preset time frequency (e.g., once per hour) or an external trigger command, performing periodic or triggered non-contact scanning of the entire coal yard area. This process collects real-time information such as spatial point coordinates, distances, and reflection intensity within the entire coal yard area, forming an original point cloud dataset.

[0033] The acquired raw spatial point information undergoes preliminary preprocessing, specifically including removing obviously abnormal noise points, invalid points, and outliers caused by environmental dust or equipment interference. The data is then smoothed and regularized to generate basic point cloud data that clearly reflects the macroscopic morphology of the coal yard, the overall outline of the coal pile, and the site boundaries. This generated basic point cloud data will be uploaded to a central processing server for storage and subsequent analysis via wireless or wired transmission at the network layer, providing reliable raw data support for morphological change recognition, blind zone determination, and scanning task generation in step S2.

[0034] Step S2: The central processing unit receives and analyzes the basic point cloud data, identifies the coal pile morphology change area and the scanning blind zone of the fixed scanning network, determines the target area that needs to be scanned in detail, and generates the corresponding scanning task.

[0035] In the embodiments of this application, such as Figure 2As shown, the network layer, serving as the core hub for system data transmission, includes network switches and servers that receive basic point cloud data uploaded from the fixed scanning network of the perception layer in real time, achieving efficient transmission and initial storage of the basic point cloud data. After data transmission is completed, the data processing module built into the central processing server performs further noise reduction and filtering preprocessing on the received basic point cloud data. This focuses on removing invalid point clouds, noise points, and outliers caused by factors such as dust obstruction, equipment electromagnetic interference, and light reflection deviations in the coal yard operating environment. This data purification process significantly improves the purity and effectiveness of the basic point cloud data, laying a solid foundation for subsequent data analysis and identification.

[0036] After preprocessing, the data processing module of the central processing server performs a precise comparative analysis of the current preprocessed basic point cloud data with the historical scan data stored in the system. By calculating the difference, elevation deviation, and contour offset between the two sets of data, it automatically identifies areas of morphological change in the coal pile during storage, such as elevation changes and contour offsets, accurately capturing the dynamic changes in the coal pile's shape. Simultaneously, the data processing module also identifies scanning blind spots in the fixed scanning network. Considering factors such as the deployment location limitations of the fixed scanning units, the coal pile's own obstruction, and temporary obstruction from large operating equipment like stacker-reclaimers, it precisely defines the spatial range, coordinate information, and size of the scanning blind spots, ensuring that no area requiring detailed scanning is missed.

[0037] After identifying areas of coal pile morphological change and determining blind spots, this embodiment of the application comprehensively delineates target areas requiring refined scanning based on the identified two types of areas. Combining the actual conditions of each area, it clarifies the specific scanning range, scanning accuracy requirements, scanning duration, and priority of each target area. Subsequently, the central processing server generates corresponding refined scanning tasks based on these parameters, specifying task instructions, target coordinates, and operational standards. This scanning task serves as the core basis for the subsequent scheduling and execution of the mobile coal inventory unit, ensuring that the mobile coal inventory unit can accurately and efficiently complete refined supplementary scanning operations, thus compensating for the deficiencies of the fixed scanning network.

[0038] Step S3: Based on the scanning task, a scheduling is issued to control the mobile coal storage device to move to the target area, and a fine scan is performed using the mobile scanning unit. At the same time, the pose information is collected in real time through the positioning and orientation unit to obtain supplementary point cloud data with spatial location information.

[0039] In the embodiments of this application, such as Figure 2 As shown, after the central processing server generates a detailed scanning task, it will accurately send the scanning task to the mobile coal storage device through the wireless communication link of the network layer, ensuring that the task instructions are transmitted efficiently and without omission.

[0040] Mobile coal palletizing devices serve as the core carrier for performing fine scanning operations. Their types can be flexibly selected according to actual operational needs. They can be unmanned vehicles with autonomous mobility, portable devices that are easy to operate by hand, or remote-controlled devices adapted to small-scale operational scenarios. The core purpose is to complete the fine scanning of the target area while accurately collecting relevant position and posture information.

[0041] After receiving a scanning task, the device will first parse the task instructions to clarify the coordinate range of the target scanning area and the scanning accuracy requirements. Then, it will autonomously plan its movement path or smoothly drive towards the target area according to manual remote control instructions. During the movement, it will provide real-time feedback on its own position status to avoid deviating from the working route or missing the target area, ensuring that it accurately arrives at the designated scanning position.

[0042] Upon reaching the target area, the device immediately activates its integrated mobile scanning unit to conduct a high-density, high-precision fine scan of the target area, comprehensively capturing detailed structural information of the coal pile to ensure the integrity and accuracy of the scan data. Simultaneously, the positioning and orientation unit is activated, acquiring the device's real-time position via a Global Navigation Satellite System (GNSS) receiver, and combining this with the attitude and heading information captured by the Inertial Measurement Unit (IMU), achieving synchronous acquisition of scan data and attitude information.

[0043] After scanning is completed, the device binds the collected fine scan data with the corresponding pose information to generate supplementary point cloud data with precise spatial location identifiers. Then, through a wireless communication link, the supplementary point cloud data is uploaded to the central processing server, providing reliable supplementary data support for subsequent point cloud fusion, model building and data processing, ensuring the accuracy and completeness of subsequent multi-source data fusion, and also laying the foundation for subsequent coal pile volume calculation and inventory statistics.

[0044] Step S4: Unify the basic point cloud data and the supplementary point cloud data into the global coordinate system for fusion processing. Reconstruct the three-dimensional model of the coal pile based on the fused point cloud data and calculate the coal pile inventory. Combine the preset coal pile density data to convert the coal pile weight and form a complete coal yard inventory result.

[0045] In the embodiments of this application, such as Figure 2As shown, after receiving the basic point cloud data uploaded by the fixed scanning network and the supplementary point cloud data uploaded by the mobile coal storage device, the central processing server performs unified processing by the application layer's data fusion and 3D modeling module. To ensure that the multi-source point cloud data are processed under the same spatial reference, this application uses the coordinate system adopted by the fixed scanning network as the global reference coordinate system. Based on the position, attitude, and heading pose information uploaded by the mobile coal storage device, the supplementary point cloud data is transformed from its own local coordinate system to the global coordinate system, thus completing the coordinate unification and spatial alignment of the two types of point cloud data.

[0046] Based on the unified coordinate system, the ICP point cloud registration algorithm is used to match, stitch and optimize the basic point cloud data and the supplementary point cloud data. This process eliminates duplicate and redundant point clouds generated during the scanning process, corrects data offsets caused by equipment errors and environmental interference, and achieves seamless integration of fixed scanning data and mobile scanning data. Finally, a complete point cloud dataset covering the entire coal yard area with continuous and reliable data is formed, providing a high-quality data foundation for subsequent 3D modeling and volume calculation.

[0047] Based on the fused global point cloud data, this embodiment further constructs a three-dimensional surface model of the coal pile, realistically restoring the external contour, elevation distribution, and spatial morphology of the coal pile. The volume of the coal pile is then accurately calculated using either a triangular meshing algorithm or an integral algorithm. When using an integral algorithm, a preset benchmark ground is used as the reference plane. The elevation distribution on the surface of the coal pile is denoted as The volume V of the coal pile can then be expressed as a double integral:

[0048] Where V is the total volume of the coal pile. To preset the elevation value of the reference ground, Let be the actual elevation of the coal pile surface at the horizontal coordinate (x, y), D be the horizontal projection area of ​​the coal pile on the reference ground, and x and y be the two-dimensional coordinate variables in the horizontal direction.

[0049] When using the triangular meshing algorithm, the system discretizes the three-dimensional surface model of the coal pile into N spatial triangular patches, each triangular patch corresponding to a micro-element volume. The total volume of the coal pile is the sum of the volumes of its individual infinitesimal elements:

[0050] Where N is the total number of triangular facets obtained from discretizing the 3D model. Let i be the infinitesimal volume corresponding to the i-th triangular facet, where i is the index of the triangular facet.

[0051] For a single triangular facet element, if its projected area on the reference plane is S...i The average elevation of the coal piles in the corresponding area is Then the volume of the infinitesimal element can be expressed as:

[0052] in, Let be the projected area of ​​the i-th triangular facet on the reference ground. Let be the average elevation of the coal pile within the i-th micro-element region.

[0053] After obtaining the coal pile volume, and combining it with the pre-entered coal pile density parameter ρ, the volume is converted into the coal pile weight using the mass calculation formula, which represents the actual inventory in the coal yard. The calculation formula is as follows:

[0054] Where m is the total mass of the coal pile, i.e. the coal yard inventory, and ρ is the average bulk density of the coal pile.

[0055] If a coal yard contains multiple types of coal with varying densities, the coal pile can be divided into K characteristic zones, each with a volume of [missing information]. The corresponding density is The total coal stockpile inventory can then be expressed as:

[0056] Where K is the total number of zones in the coal pile. Let be the bulk density of the coal body in the j-th partition. Let j be the volume corresponding to the j-th partition, where j is the partition number.

[0057] Finally, the system integrates the 3D model of the coal pile, volume calculation results, weight calculation results, and related operational information to form a complete and standardized coal yard inventory result. For example... Figure 2 As shown, the coal inventory results are transmitted to the coal inventory monitoring center at the application layer through the network layer, and then visualized on user terminals such as computers, mobile terminals, and monitoring screens. This enables intuitive presentation, remote viewing, and centralized management of coal inventory data, providing reliable data support for coal yard inventory control, production scheduling, and cost accounting.

[0058] In the application of one embodiment of the present invention, the implementation process is as follows: like Figure 3 As shown, this application embodiment takes a closed arched coal yard as a typical application scenario, and fully deploys a fixed coal inventory system and a mobile coal inventory system to work together to achieve efficient, accurate, and comprehensive coal inventory operations. The specific implementation process is as follows: In this embodiment of the application, as shown in 3, the enclosed coal yard body is an arched steel structure canopy. A fixed scanning unit is installed on each of the four corner columns of the coal yard, which are respectively labeled as fixed scanning unit 101, fixed scanning unit 102, fixed scanning unit 103, and fixed scanning unit 104. Each scanning unit uses a two-dimensional lidar or a high-performance millimeter-wave radar. The installation position has a wide field of view and no obstruction. Its scanning surface covers the entire coal yard area by horizontal rotation, realizing normalized and automated full-area inventory. Moreover, the technical characteristics of actively emitting detection waves make it unaffected by environmental factors such as light and dust inside the coal yard.

[0059] A mobile coal inventory device is deployed in the center of the coal yard. This device can be an unmanned vehicle integrating a laser scanner 201 and a GNSS / IMU antenna 202, or a track-mounted coal inventory system within the coal shed. It stands by within the coal yard to supplement blind spots of fixed equipment, accurately verify abnormal data, and flexibly conduct localized inventory checks. The central processing server 300 is located in the control room. Users can access the system via computer (user terminal 400) and is responsible for data reception, fusion processing, modeling calculations, and result display.

[0060] The complete daily coal inventory process is as follows: 1. Task triggering and basic scanning.

[0061] The system supports multiple trigger modes for inventory tasks: First, timed triggering, automatically starting fixed scanning units 101, 102, 103, and 104 according to a preset cycle (e.g., once per hour); second, manual triggering, allowing administrators to initiate a full or partial inventory with a single click on user terminal 400; and third, event triggering, automatically starting scanning before or after important operations such as large-scale material handling and stockpiling. After a task is triggered, the four fixed scanning units start simultaneously, performing a non-contact 3D scan of the entire coal yard, collecting spatial point coordinates, distances, and other information to form basic point cloud data reflecting the macroscopic shape, overall outline, and site boundaries of the coal yard. This data is then uploaded to the central processing server 300 via a wired network.

[0062] 2. Blind spot identification and refined task generation.

[0063] After receiving the basic point cloud data, the central processing server 300 first performs noise reduction and filtering preprocessing to remove invalid point clouds, noise points, and outliers caused by environmental dust, equipment interference, etc., thereby improving data purity and effectiveness. Then, it compares and analyzes the preprocessed current scan data with historical scan data from the same period. By calculating the data difference, it automatically identifies areas of change in the coal pile morphology. It discovers that a new depression has formed on the top of coal pile No. 3 due to the operation of the stacker-reclaimer. This area is a relative blind spot for fixed scanning units 101, 102, 103, and 104, making it impossible to obtain accurate data through fixed scanning. Based on the identification results, the server automatically generates a task for "refined scanning of the top of coal pile No. 3," specifying the target area coordinates, scanning range, and accuracy requirements, and sends this task to the unmanned vehicle (mobile coal stacking device) 200 via the 5G network.

[0064] 3. Mobile fine scanning and pose acquisition.

[0065] After receiving the detailed scanning task, the unmanned vehicle 200 first analyzes the task instructions, clarifying the coordinates of the target area on top of coal pile No. 3 and the required scanning accuracy. It then autonomously plans the optimal movement path (such as a zigzag path), avoiding blind spots and obstacles, smoothly driving towards the bottom of coal pile No. 3, and then ascending along the zigzag path to the target area. During the movement, it provides real-time feedback on its position, attitude, and driving status, facilitating real-time monitoring and correction of path deviations by the central processing server 300, ensuring accurate arrival at the designated work location.

[0066] Upon arrival at the target area, the integrated laser scanner 201 is immediately activated to conduct a high-density, high-precision fine scan of the recessed area on top of coal pile No. 3 in high-frequency mode, comprehensively capturing detailed structural information of the local coal pile. Simultaneously, the GNSS / IMU antenna 202 is activated to collect the device's position, attitude, and heading information in real time. The base station deployed within the coal yard can also provide real-time differential GNSS signals, further improving positioning accuracy. After the scan is completed, the fine scan data is synchronously bound with the corresponding pose information to generate supplementary point cloud data with precise spatial location markers. This data is then uploaded to the central processing server 300 via the wireless communication unit, providing supplementary data for subsequent multi-source data fusion.

[0067] 4. Multi-source data fusion and 3D modeling calculation.

[0068] The data processing module of the central processing server 300 uses the coordinate system adopted by the fixed scanning units 101, 102, 103, and 104 as the global reference coordinate system. Based on the position, attitude, and heading information uploaded by the unmanned vehicle 200, it transforms the supplementary point cloud data from its own local coordinate system to the global coordinate system, completing the coordinate unification and spatial alignment of the two types of point cloud data. Subsequently, the ICP (Iterative Closest Point) point cloud registration algorithm is used to match, stitch, and optimize the basic point cloud data and the supplementary point cloud data, eliminating duplicate and redundant point clouds generated during the scanning process, correcting data offsets caused by equipment errors and environmental interference, and achieving seamless integration of fixed scanning data and mobile scanning data. Finally, a complete point cloud dataset covering the entire coal yard area, with continuous data and reliable accuracy is formed, laying a high-quality data foundation for subsequent 3D modeling and volume calculation.

[0069] The model calculation module constructs a 3D surface model of the coal pile based on the fused global point cloud data, realistically reproducing the external contour, elevation distribution, and spatial morphology of the coal pile, and accurately calculates the volume of the coal pile through triangular meshing or integral algorithms. When using an integral algorithm, a preset reference ground surface is used as the reference plane. The elevation distribution on the surface of the coal pile is denoted as The volume V of the coal pile can then be expressed as a double integral:

[0070] Where V is the total volume of the coal pile. To preset the elevation value of the reference ground, Let be the actual elevation of the coal pile surface at the horizontal coordinate (x, y), D be the horizontal projection area of ​​the coal pile on the reference ground, and x and y be the two-dimensional coordinate variables in the horizontal direction.

[0071] When using the triangular meshing algorithm, the system discretizes the three-dimensional surface model of the coal pile into N spatial triangular patches, each triangular patch corresponding to a micro-element volume. The total volume of the coal pile is the sum of the volumes of its individual infinitesimal elements:

[0072] Where N is the total number of triangular facets obtained from discretizing the 3D model. Let i be the infinitesimal volume corresponding to the i-th triangular facet, where i is the index of the triangular facet.

[0073] For a single triangular facet element, if its projected area on the reference plane is S... i The average elevation of the coal piles in the corresponding area is Then the volume of the infinitesimal element can be expressed as:

[0074] in, Let be the projected area of ​​the i-th triangular facet on the reference ground. Let be the average elevation of the coal pile within the i-th micro-element region.

[0075] After obtaining the coal pile volume, and combining it with the pre-entered coal pile density parameter ρ, the volume is converted into the coal pile weight using the mass calculation formula, which represents the actual inventory in the coal yard. The calculation formula is as follows:

[0076] Where m is the total mass of the coal pile, i.e. the coal yard inventory, and ρ is the average bulk density of the coal pile.

[0077] If a coal yard contains multiple types of coal with varying densities, the coal pile can be divided into K characteristic zones, each with a volume of [missing information]. The corresponding density is The total coal stockpile inventory can then be expressed as:

[0078] Where K is the total number of zones in the coal pile. Let be the bulk density of the coal body in the j-th partition. Let j be the volume corresponding to the j-th partition, where j is the partition number.

[0079] 5. Results output and visualization management.

[0080] The system integrates information such as the 3D model of coal pile No. 3, volume calculation results, inventory weight calculation results, inventory time, and operation logs to generate a standardized coal inventory report. The report can be exported as PDF or Excel for easy archiving and submission. The inventory results are transmitted via the network layer to the coal inventory monitoring center at the application layer, where they are visualized on user terminals (such as computers, mobile terminals, and large monitoring screens). This dynamically presents the 3D coal pile model, real-time inventory data, historical change curves, and other content, enabling intuitive presentation, remote viewing, and centralized management of the inventory data. All inventory data is automatically stored in the database for subsequent historical queries, comparative analysis, and auxiliary decision-making in production scheduling and cost accounting.

[0081] Example 2 This invention provides a coal yard inventory device based on multi-source data fusion. Figure 4 This is a schematic diagram of a coal yard inventory device based on multi-source data fusion, provided as an embodiment of the present invention. Figure 4 As shown, the device includes: The global scanning module 100 is used to perform periodic or triggered three-dimensional scanning of the entire coal yard area through a fixed scanning network deployed at predetermined locations in the coal yard, and to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard. The blind spot identification module 200 is used to receive and analyze the basic point cloud data through the central processing unit, identify the coal pile morphology change area and the scanning blind spot of the fixed scanning network, determine the target area that needs to be scanned in detail and generate the corresponding scanning task. The fine scanning module 300 is used to issue and schedule based on scanning tasks, control the mobile coal storage device to move to the target area, use the mobile scanning unit to perform fine scanning, and at the same time collect pose information in real time through the positioning and orientation unit to obtain supplementary point cloud data with spatial location information. The data fusion module 400 is used to unify the basic point cloud data and the supplementary point cloud data into a global coordinate system for fusion processing, reconstruct a three-dimensional model of the coal pile based on the fused point cloud data and calculate the coal pile inventory, and convert the coal pile weight by combining the preset coal pile density data to form a complete coal yard inventory result.

[0082] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0083] Example 3 To implement the methods of the above embodiments, the present invention also provides an electronic device, which includes a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the various steps of the methods described above.

[0084] Example 4 To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in the foregoing embodiments.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0086] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0087] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A coal yard inventory method based on multi-source data fusion, characterized in that, include: By deploying a fixed scanning network at predetermined locations in the coal yard, periodic or triggered 3D scans are performed on the entire coal yard area to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard. The central processing unit receives and analyzes the basic point cloud data, identifies the coal pile morphology change area and the scanning blind spot of the fixed scanning network, determines the target area that needs to be scanned in detail, and generates the corresponding scanning task. Based on the scanning task, the scheduling is issued to control the mobile coal storage device to move to the target area, and the mobile scanning unit is used to perform fine scanning. At the same time, the positioning and orientation unit collects pose information in real time to obtain supplementary point cloud data with spatial location information. The basic point cloud data and the supplementary point cloud data are unified into a global coordinate system and fused. Based on the fused point cloud data, a three-dimensional model of the coal pile is reconstructed and the coal pile inventory is calculated. The coal pile weight is converted by combining the preset coal pile density data to form a complete coal yard inventory result.

2. The method according to claim 1, characterized in that, By deploying a fixed scanning network at predetermined locations within the coal yard, periodic or triggered 3D scans are performed on the entire coal yard area to acquire basic point cloud data reflecting the macroscopic morphology of the coal yard, including: Multiple fixed scanning units are deployed at predetermined locations in the coal yard and on surrounding pillars and building structures to form a basic scanning network covering the entire coal yard area. All fixed scanning units are activated according to a preset time frequency or external trigger command to carry out periodic or trigger-based 3D scanning operations. Real-time spatial point information of the entire coal yard is collected using two-dimensional lidar, three-dimensional lidar, millimeter-wave radar or structured light three-dimensional scanner; The collected spatial point information is initially sorted out, obvious abnormal data is removed, and basic point cloud data that can clearly reflect the macroscopic shape of the coal yard, the overall outline of the coal pile and the site boundary is generated.

3. The method according to claim 2, characterized in that, The central processing unit receives and analyzes the basic point cloud data, identifies areas of coal pile morphological change and blind spots of the fixed scanning network, determines target areas requiring refined scanning, and generates corresponding scanning tasks, including: The central processing server receives basic point cloud data and performs noise reduction and filtering preprocessing on the data to remove environmental interference and invalid point clouds. By comparing and analyzing the preprocessed current basic point cloud data with historical scan data from the same period, and by calculating the data difference, the morphological change areas of coal pile elevation change and contour offset are identified. Determine the scanning blind spots of the fixed scanning network caused by equipment deployment location limitations and coal pile obstruction, and determine the spatial range and coordinate information of the blind spots; Based on the morphological change area and the scanning blind zone, the fine scanning target area is delineated, the scanning range and accuracy requirements of each target area are determined, and the corresponding fine scanning task is generated.

4. The method according to claim 3, characterized in that, Based on the scanning task, a scheduling mechanism is issued to control the mobile coal storage device to move to the target area, including: The central processing server distributes the detailed scanning task to the mobile coal storage device via a wireless communication link. Based on the issued detailed scanning task, the mobile coal storage device, which uses an unmanned vehicle, remote control vehicle or handheld device, is controlled to autonomously plan its path or accept remote control according to the task information and drive to the target area. The device provides real-time feedback on its location status during movement, ensuring that it accurately reaches the designated work location.

5. The method according to claim 4, characterized in that, Fine scanning is performed using a mobile scanning unit, while pose information is acquired in real time using a positioning and orientation unit to obtain supplementary point cloud data with spatial location information, including: Once the mobile coal stacking device reaches the target area, a lightweight laser scanner or depth camera is activated to conduct a high-density fine scan and obtain detailed structural information of the local coal pile. The mobile coal storage device uses a global navigation satellite system receiver and inertial measurement unit to collect the device's position, attitude, and heading information in real time. The fine scan data is bound with the corresponding pose information to generate supplementary point cloud data with accurate spatial location information, which is then uploaded to the central processing server.

6. The method according to claim 5, characterized in that, The basic point cloud data and the supplementary point cloud data are fused together in a global coordinate system. A three-dimensional model of the coal pile is reconstructed based on the fused point cloud data, and the coal pile inventory is calculated. The weight of the coal pile is then converted using preset coal pile density data to form a complete coal yard inventory result, including: Using the coordinate system of the fixed scanning network as the global reference, the central processing server converts the supplementary point cloud data into a unified global coordinate system based on the pose information uploaded by the mobile coal storage device. A point cloud registration algorithm is used to match and stitch together the basic point cloud data and the supplementary point cloud data, eliminating duplicate and redundant data, and achieving seamless fusion of multi-source point clouds. A three-dimensional model of the coal pile is constructed based on the fused complete point cloud data, and the volume of the coal pile is calculated by triangulation algorithm or integral algorithm. By combining the pre-entered coal pile density data, the coal pile volume is converted into coal pile weight, and the three-dimensional model, volume and weight information are integrated to finally form a complete coal yard inventory result.

7. A coal yard inventory device based on multi-source data fusion, characterized in that, include: The global scanning module is used to perform periodic or triggered 3D scanning of the entire coal yard area through a fixed scanning network deployed at predetermined locations in the coal yard, and to obtain basic point cloud data reflecting the macroscopic morphology of the coal yard. The blind spot identification module is used to receive and analyze the basic point cloud data through the central processor, identify the coal pile morphology change area and the scanning blind spot of the fixed scanning network, determine the target area that needs to be scanned in detail and generate the corresponding scanning task. The fine scanning module is used to issue and schedule based on scanning tasks, control the mobile coal storage device to move to the target area, use the mobile scanning unit to perform fine scanning, and at the same time collect pose information in real time through the positioning and orientation unit to obtain supplementary point cloud data with spatial location information. The data fusion module is used to unify the basic point cloud data and the supplementary point cloud data into a global coordinate system for fusion processing. Based on the fused point cloud data, a three-dimensional model of the coal pile is reconstructed and the coal pile inventory is calculated. Combined with the preset coal pile density data, the coal pile weight is converted to form a complete coal yard inventory result.

8. The apparatus according to claim 7, characterized in that, The global scanning module is also used for: Multiple fixed scanning units are deployed at predetermined locations in the coal yard and on surrounding pillars and building structures to form a basic scanning network covering the entire coal yard area. All fixed scanning units are activated according to a preset time frequency or external trigger command to carry out periodic or trigger-based 3D scanning operations. Real-time spatial point information of the entire coal yard is collected using two-dimensional lidar, three-dimensional lidar, millimeter-wave radar or structured light three-dimensional scanner; The collected spatial point information is initially sorted out, obvious abnormal data is removed, and basic point cloud data that can clearly reflect the macroscopic shape of the coal yard, the overall outline of the coal pile and the site boundary is generated.

9. An electronic device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method as described in any one of claims 1-6.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images