Coal storage yard batch identification method, system and device based on identification

Abstract

The invention provides a coal storage yard batch identification method, system and device based on identification. The method comprises the following steps: distributing a unique industrial internet identification code for each batch of coal samples; spatial position data, stacking state data and environment data of each batch of coal sample storage yard are collected respectively; spatial position data, stacking state data and environmental data of the coal sample storage yard are used as input of a spatial segmentation algorithm, different batches of the coal sample storage yard are used as output of the spatial segmentation algorithm, and batch spatial boundaries of different batches of divided coal samples are associated with industrial internet identification codes. Establishing a batch identification information base; the spatial position data, the stacking state data and the environmental data of the to-be-identified coal storage yard are collected, and the batch to which the to-be-identified coal belongs is determined based on the batch boundaries of different batches of coal samples in the batch identification information base, so that the efficiency and the reliability of coal storage yard batch identification in a power plant are improved.

Description

Technical Field

[0001] This invention relates to the field of coal yard management, and in particular to a method, system and apparatus for batch identification of coal yards based on identification tags. Background Technology

[0002] Coal, as a vital energy resource, plays a crucial role in coal storage management throughout its production, transportation, and use. Traditional batch management methods in power plants primarily rely on manual recording or simple electronic tagging technology, which suffers from the following problems: Low batch identification accuracy: Traditional methods struggle to accurately distinguish coal batches within power plants, especially when coal is stacked or mixed in the storage yard, making batch information prone to loss. Low management efficiency: Manual recording or traditional barcode technology requires significant human intervention, resulting in low efficiency and a high risk of errors. For example, a bottleneck discovered during the actual implementation of an energy management center project at a thermal power company was that the power plant needed to dynamically blend multiple batches of raw coal based on boiler operating conditions, leading to continuous fluctuations in the lower heating value (LHV) of the coal entering the furnace. In recent years, Industrial Internet identification technology, as an emerging digital management tool, has enabled information tracking of physical objects throughout their entire lifecycle by assigning unique digital identifiers.

[0003] However, the application of industrial internet identification technology in coal yard management in power plants is still in its early stages. Existing technologies lack the ability to accurately and in real-time identify the physical boundaries of coal batches in power plant yards. They can only use a crude method of manually averaging LHV to calculate standard coal quantity, resulting in significant lag and batch errors. This error directly leads to inaccurate standard coal quantity calculations, the inability to achieve precise batch-level traceability and responsibility allocation, and seriously hinders the refined management and cost control of energy consumption in thermal power companies; thus, the efficiency and reliability of batch identification in power plant coal yards are low.

[0004] To address this problem, the present invention provides a method, system, and apparatus for batch identification of coal stockpiles based on identification marks, thereby solving the aforementioned issues. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, this invention innovatively proposes a batch identification method, system and device for coal stockpiles based on identification, which effectively solves the problem of low efficiency and reliability of batch identification in coal stockpiles in power plants caused by the prior art, and effectively improves the efficiency and reliability of batch identification in coal stockpiles in power plants.

[0006] The first aspect of this invention provides a batch identification method for coal stockpiles based on identifiers, comprising: Each batch of coal samples is assigned a unique Industrial Internet identification code, which includes coal batch information; Spatial location data, stacking status data, and environmental data of each batch of coal samples were collected from the stockpile. The spatial location data, stacking status data, and environmental data of the coal sample stockpile are used as inputs to the spatial segmentation algorithm, and different batches of the coal sample stockpile are used as outputs to delineate the batch spatial boundaries of different batches of coal samples. The batch spatial boundaries of different batches of coal samples after division are associated with industrial internet identification codes to establish a batch identification information database. Collect spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. Based on the batch boundaries of different batches of coal samples in the batch identification information database, determine the batch to which the coal stockpile to be identified belongs.

[0007] Optionally, the identification code may also include coal source information, coal production time information, and a check code, and the identification code is associated with and bound to the quality parameter information of the coal stockpile.

[0008] Optionally, the spatial location data is a weighted fusion of first spatial location data and second spatial location data, wherein the first spatial location data is spatial location data collected by lidar, and the second spatial location data is spatial location data obtained by the camera through depth reconstruction of image information.

[0009] Optionally, the stacking status data includes the average density of the coal pile, the particle size distribution of the coal pile, and the height of the coal pile.

[0010] Furthermore, the spatial location data, stacking status data, and environmental data of the coal sample stockpile are used as inputs to the spatial segmentation algorithm, and different batches of the coal sample stockpile are used as outputs. The specific steps for defining the batch spatial boundaries of different batches of coal samples include: Randomly initialize k clusters and k cluster center points, with each cluster corresponding to a batch of coal samples; Each sample data point is assigned to the nearest center by using the weighted distance between the sample data point and the center point within the cluster; Update the cluster center point to the mean of all data points in the cluster until a preset termination condition is met; the preset termination condition is that the preset number of iterations is met or the objective function value is less than a preset function threshold.

[0011] Furthermore, the specific method for calculating the weighted distance between sample data points and cluster center points is as follows:

[0012] in, x is the weighted distance between the sample data points and the cluster center. n Let y be the value of the nth feature in the sample data points.n Let be the value of the nth feature among the cluster center points. The weight is the weight corresponding to the nth feature; and the weights corresponding to different features are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategies.

[0013] Optionally, it also includes: The collected spatial location data, stacking status data, and environmental data of the coal storage yard to be identified are associated with the batch information of the coal storage yard to be identified in the Industrial Internet Identifier, so as to realize real-time query and full-chain traceability of coal batch information. In addition, the batch identification information database supports dynamic updates.

[0014] A second aspect of the present invention provides an identifier-based batch identification system for coal stockpiles, comprising: The identification management module assigns a unique Industrial Internet identification code to each batch of coal samples, and the identification code includes coal batch information. The data acquisition module collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile. The batch segmentation module takes the spatial location data, stacking status data, and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to the spatial segmentation algorithm, thus dividing the batch spatial boundaries of different batches of coal samples. The data processing module associates the batch spatial boundaries of different batches of coal samples with industrial internet identification codes to establish a batch identification information database. The batch identification module collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. Based on the batch boundaries of different batches of coal samples in the batch identification information database, it determines the batch to which the coal stockpile to be identified belongs.

[0015] Optionally, it also includes: The query and traceability module supports querying and full-chain traceability of batch information of coal stockpiles to be identified.

[0016] A third aspect of the present invention provides a batch identification device for coal stockpiles based on identification marks, comprising: In the cloud, an identifier resolution system is set up. The identifier resolution system assigns a unique industrial internet identifier code to each batch of coal samples. The identifier code includes coal batch information and is stored and queried based on the batch information of the coal stockpile. The IoT data acquisition device collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile, and also collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. The edge computing device takes the spatial location data, stacking status data and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to the spatial segmentation algorithm to divide the batch spatial boundaries of different batches of coal samples. The batch spatial boundaries of different batches of coal samples after segmentation are associated with industrial internet identification codes to establish a batch identification information database; based on the batch boundaries of different batches of coal samples in the batch identification information database, the batch to which the coal stockpile to be identified belongs is determined.

[0017] The technical solution adopted in this invention has the following technical effects: 1. The technical solution of this invention assigns a unique Industrial Internet identifier code to each batch of coal samples; collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile; uses the spatial location data, stacking status data, and environmental data of the coal sample stockpile as input to a spatial segmentation algorithm, and uses different batches of the coal sample stockpile as output to divide the batch spatial boundaries of different batches of coal samples; associates the divided batch spatial boundaries of different batches of coal samples with the Industrial Internet identifier code to establish a batch identifier information database; collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified, and determines the batch to which the coal stockpile to be identified belongs based on the batch boundaries of different batches of coal samples in the batch identifier information database, effectively solving the problem of low efficiency and reliability of batch identification of coal stockpiles in power plants caused by existing technologies, and effectively improving the efficiency and reliability of batch identification of coal stockpiles in power plants.

[0018] 2. The identification code in the technical solution of this invention also includes coal source information, coal production time information, and check code. The identification code is associated with and bound to the quality parameter information of the coal stockpile. The quality parameters are not directly written into the Handle identifier, but are pointed to by the identifier in the database record. After the identifier is parsed, the batch data record can be accessed. This method conforms to the minimum coding principle of industrial Internet identifiers, avoids excessively long identifiers, and supports dynamic updates of quality parameters.

[0019] 3. The spatial location data in the technical solution of this invention is a weighted fusion of first spatial location data and second spatial location data. The first spatial location data is spatial location data collected by lidar, and the second spatial location data is spatial location data obtained by depth reconstruction of image information collected by camera. The stacking status data includes the average density of coal pile, the particle size distribution of coal pile, and the height of coal pile. This ensures both the continuity of the coal pile surface and maintains high-precision distance information. These parameters form a multi-dimensional feature vector, ensuring the reliability of batch identification of coal stockpiles in power plants.

[0020] 4. The weights corresponding to different features in the technical solution of this invention are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategy. This improves the adaptability of batch identification of coal stockpiles in power plants.

[0021] 5. In the technical solution of this invention, the spatial location data, stacking status data, and environmental data of the coal storage yard to be identified are associated with the batch information of the coal storage yard to be identified in the Industrial Internet identifier, so as to realize real-time query and full-chain traceability of coal batch information in power plants. Moreover, the batch identifier information database supports dynamic updates, realizing real-time query and full-chain traceability of batch information, meeting management needs. The traceability mechanism supports data sharing with upstream and downstream of the supply chain (such as transportation, sales, and user units), effectively solving the data silo problem, significantly improving the supply chain collaboration efficiency, and realizing full life cycle management from production to use.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart (I) of the method in Embodiment 1 of the present invention. Figure 2 This is a flowchart (II) of the method in Embodiment 1 of the present invention. Figure 3 This is a flowchart (III) of the method in Embodiment 1 of the present invention. Figure 4 This is a schematic diagram of the system architecture in Embodiment 2 of the present invention; Figure 5 This is a schematic diagram of the device architecture in Embodiment 3 of the present invention. Detailed Implementation

[0025] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. To simplify the disclosure of the invention, components and arrangements of specific examples are described below. Furthermore, reference numerals and / or letters may be repeated in different examples. This repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. Descriptions of well-known components, processing techniques, and processes are omitted in this invention to avoid unnecessarily limiting the invention.

[0026] Example 1 like Figures 1-2 As shown, the present invention provides a batch identification method for coal stockpiles based on identifiers, including: S1, assign a unique Industrial Internet identification code to each batch of coal samples, the identification code including coal batch information; S2 collects spatial location data, stacking status data, and environmental data of each batch of coal sample yards; S3 takes the spatial location data, stacking status data and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to divide the batch spatial boundaries of different batches of coal samples. S4. Associate the batch spatial boundaries of different batches of coal samples after division with the industrial internet identification code to establish a batch identification information database. S5 collects spatial location data, stacking status data, and environmental data of the coal storage yard to be identified. Based on the batch boundaries of different batches of coal samples in the batch identification information database, it determines the batch to which the coal storage yard to be identified belongs.

[0027] In step S1, which is the identification allocation stage, the identification code also includes coal source information, coal production time information, and a check code. The identification code is associated and bound to the quality parameter information of the coal stockpile.

[0028] Specifically, each batch of coal is assigned a unique Industrial Internet Identifier code. This code contains information such as coal origin, production time, and quality parameters, and is stored in the identifier resolution system. The identifier coding structure adopts the Handle coding system (which defines coding rules, a resolution system independent of the Internet domain name system, and a globally distributed management architecture), with the format "name authority (prefix) / coal batch identifier (suffix)," for example, "86.1000.123 / 20250701120000-B001." Here, the prefix is ​​the name authority registered by the enterprise (e.g., "86.1000.123" represents a Chinese enterprise), and the suffix includes the production time (YYYYMMDDHHMMSS, 14 digits), the batch number (4 digits, such as "B001"), and a checksum (4 digits). The 4-digit checksum is added to the identifier suffix to verify the validity of the batch identifier and avoid character errors during scanning, RFID reading, or manual input. The checksum is generated using either MOD-11 or CRC16 checksum algorithms. The verification code does not participate in business field parsing during the identifier resolution phase; it is only used to determine whether the identifier has been tampered with or read incorrectly. When verification fails, a rescanning or rereading mechanism for the identifier is triggered, thereby improving identifier reliability. The allocation rule is based on the global uniqueness of the Handle system, and the resolution process uses the Handle protocol (based on TCP / IP) for distributed resolution: first, the prefix is ​​resolved to the enterprise node through the global Handle registry, and then the suffix is ​​resolved to batch details through the local Handle service. This solution optimizes the embedding of time and quality parameters for coal batches, reducing redundant fields and improving resolution efficiency compared to general Handle systems (such as object identifiers OID and item codes Ecode). Quality parameters are not directly written to the Handle identifier but are instead linked to database records. A one-to-one binding method is used between the identifier and the quality parameters: after identifier resolution, the batch data record in the system can be accessed, which stores quality parameter fields (such as ash content A%, volatile matter V%, total sulfur S%, and lower heating value Q_net). This method conforms to the minimum coding principle of industrial internet identifiers, avoiding excessively long identifiers, while also supporting dynamic updates of quality parameters.

[0029] In step S2, which is the data acquisition stage, IoT devices (such as RFID tags, sensors, cameras, etc.) deployed in the coal sample storage yard of the power plant are used to collect spatial location data, stacking status data, and environmental data of each batch of coal sample storage yard in the power plant.

[0030] Specifically, the stacking status data includes the average density of the coal pile, the particle size distribution of the coal pile, and the height of the coal pile.

[0031] Stacking status (stack height and / or slope) and environmental data (humidity, temperature, rainfall) are mainly used for: calibrating the credibility of 3D point clouds; triggering the re-identification and division of spatial boundaries when the shape of the coal pile changes due to rainfall or collapse; and adding it to the spatial segmentation algorithm model as an extended feature of the spatial segmentation algorithm (clustering algorithm).

[0032] In addition to incorporating spatial location (x, y, z), average coal density, and coal particle size distribution into the feature vector of the coal sample stockpile, this invention also incorporates stockpile height h (m) and surface humidity w (%) (which can also be surface temperature, surface rainfall, etc.) into the feature vector. Therefore, the complete feature vector of the coal sample stockpile is: F=[x, y, z, density, particle, h, w]; Among them, h and w are obtained by scanning with a laser rangefinder and environmental sensors (humidity sensors), making the batch boundary division more accurate.

[0033] This invention, based on collected spatial data and identification information, utilizes spatial segmentation and batch matching algorithms to accurately delineate batch boundaries in coal stockpiles. The spatial segmentation algorithm employs K-means clustering, combined with coal physical properties (such as density 1.4 g / cm³, particle size 0.5-50 mm) for parameter setting and optimization to achieve accurate batch division.

[0034] Specifically, the spatial location data acquisition method can be: weighted fusion of first spatial location data and second spatial location data, wherein the first spatial location data is spatial location data collected by lidar, and the second spatial location data is spatial location data obtained by the camera through depth reconstruction of image information.

[0035] Spatial coordinates (x, y, z, 3D position data, acquired from a laser rangefinder and a camera respectively, unit: meters), the spatial coordinates are fused using a fusion method of laser rangefinder (LiDAR) and binocular camera depth reconstruction: (1) LiDAR provides high-precision distance measurement (±3–5cm); (2) The binocular camera provides a dense depth map; (3) The laser rangefinder and visual depth are fused using the Kalman filter concept, as shown in the following formula:

[0036] Among them, P fused The fused high-precision spatial coordinates; P LiDAR P represents the coordinates of the point measured by the lidar. cameraK1 and K2 are the coordinate points reconstructed by the binocular camera; K1 and K2 are the fusion weight coefficients, satisfying K1 + K2 = 1 (recommended values ​​K1 = 0.7, K2 = 0.3), which can be dynamically adjusted according to the dust concentration on site. This ensures both the continuity of the coal pile surface and maintains high-precision distance information.

[0037] The stack height (h) can be defined as the vertical height difference between the Z-axis coordinates of the point cloud scanned by LiDAR and the ground reference plane. The abrupt change in stack height is a key boundary criterion for physically distinguishing different coal piles or between a coal pile and the ground.

[0038] The humidity (w) value can be taken as a percentage of the measured values ​​from the ambient humidity sensors and contact moisture meters deployed in the coal yard. This parameter is introduced because changes in humidity will alter the loose density reading of the coal, and it needs to be included in the calculation as a correction factor to prevent density data fluctuations caused by rainfall from being misinterpreted as batch changes.

[0039] Particle size distribution (obtained from image analysis, average particle size, unit: mm) can be achieved by measuring each coal block in the image along a certain direction (e.g., horizontal, vertical, or other directions) using an image analysis system to obtain parameters such as the size, shape, and diameter of the coal blocks. Common image processing and analysis methods include edge detection, grayscale calculation, and morphological calculation. The particle size distribution of the coal stockpile can also be obtained through other methods (laser particle size analysis, microscopic observation). This embodiment of the invention does not impose any limitations on this.

[0040] The particle size distribution (p) can also be represented by real-time particle size analysis results from visual analysis systems (based on image recognition) deployed upstream of coal mines or in coal storage yards. To quantify the "distribution" as a single feature value for clustering, the value can be expressed using, in addition to the average particle size, the standard deviation of particle size or the uniformity index. A smaller standard deviation indicates more uniform particle size in a batch of coal, serving as an important auxiliary basis for determining whether they belong to the same batch.

[0041] Average density of coal pile Calculated using the following formula:

[0042] in, The net weight of coal entering the site (unit: tons or kg) is obtained by weighing on a truck scale (weighbridge). The volume of the stack calculated from the 3D reconstruction of the lidar point cloud (unit: m). 3 ).

[0043] Step S3 corresponds to the batch segmentation stage. It uses the spatial location data, stacking status data, and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm (clustering algorithm), and the different batches of the coal sample stockpile as output. Specifically, the batch spatial boundaries for different batches of coal samples include: Randomly initialize k clusters and k cluster center points, with each cluster corresponding to a batch of coal samples; Specifically, the number of clusters k can be determined using the Elbow Method or the Davies-Bouldin Index (DBI). For example, in the pilot test, k=6 to minimize the DBI, ensuring cluster segregation and compactness. There is a one-to-one correspondence between coal yard batch information and the number of central clusters, meaning one coal yard batch corresponds to one cluster, and one coal yard batch corresponds to a unique Industrial Internet identifier.

[0044] Each sample data point is assigned to the nearest center by using the weighted distance between the sample data point and the center point within the cluster; Specifically, the formula for the weighted distance d (weighted Euclidean distance) between sample point A and cluster center B is: in: x is the weighted distance between the sample data points and the cluster center. n Let y be the value of the nth (which can be 5) feature in the sample data points. n Let be the value of the nth feature among the cluster center points. The weight is the weight corresponding to the nth feature; and the weights corresponding to different features are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategies.

[0045] In the application of feature weights in coal storage yards of power plants, an example is as follows: Initial weights of spatial location coordinates: w pos = (Emphasizing location characteristics, corresponding to spatial coordinates x, y, z); Initial weight of coal pile height: w h = (Reflects the stacking state and changes in the geometric shape of the pile, corresponding to the coal pile height h); Initial weight of average coal density: w ρ = (A core physical property reflecting coal quality and stacking density, corresponding to the average density of the coal pile). Initial weight of surface humidity of coal pile: w w = (As an environmental correction factor, it is mainly used to eliminate environmental disturbances, corresponding to the surface humidity w of the coal pile). Initial weights for coal particle size distribution: wp = (Reflects fluidity and degree of mixing, corresponding to coal particle size distribution) ); Subsequent weights can be dynamically adjusted based on coal type, stockpiling conditions, and management strategies. Weights can be empirically adjusted based on coal type, stockpiling conditions, and management strategies, or an adaptive adjustment formula for feature weights can be constructed, for example: ; in: The weights for the nth feature are adaptively adjusted but not normalized. T represents the type of coal (e.g., bituminous coal = 1, lignite = 0.8, lean coal = 0.6). L is the coal pile location factor (the closer to the boundary of the stockpile, the larger L is, and the value ranges from 0 to 1). C is the enterprise management rule factor (determined by the enterprise weighting strategy, with a value range of 0-1). , , These correspond to the adjustment coefficients of the nth feature; For example, the weights after adaptive adjustment of spatial location coordinates but without normalization can be: W pos =0.6T + 0.2L + 0.2C; The adaptively adjusted but unnormalized weights for the average coal density can be: W ρ =0.3T + 0.6L + 0.1C; Based on the calculated but unnormalized weights of each feature, normalization is performed to obtain the subsequent adaptively adjusted weights. This allows for dynamic adjustment of the contribution of clustering features based on the actual scenario, improving the accuracy of batch segmentation.

[0046] Update the cluster center point to the mean of all data points in the cluster until a preset termination condition is met; the preset termination condition is that the preset number of iterations is met or the objective function value is less than a preset function threshold.

[0047] Specifically, the iteration can be repeated 50-100 times, or until the objective function (center movement distance) is less than a preset function threshold (e.g., 0.01m). If the density distribution is uneven, the DBSCAN variant (ε=0.5m, MinPts=5) can be switched to handle noise and outliers. The algorithm is integrated with the Handle identifier: each cluster is associated with a corresponding Handle suffix to achieve dynamic boundary adjustment and avoid the duplication of general tracing methods.

[0048] Step S4 involves information association, linking the batch spatial boundaries of different batches of coal samples after segmentation with industrial internet identification codes to establish a batch identification information database. Specifically, it links the spatial location data, stacking status data, environmental data, and batch spatial boundaries of different batches of coal samples with industrial internet identifiers to form a full lifecycle information database for each batch, supporting dynamic updates. Data flow can utilize RESTful API interfaces to transmit JSON format data. For example, raw data can be pushed via the MQTT protocol, parsed, associated with a Handle identifier, and stored.

[0049] In step S5, spatial location data, stacking status data, and environmental data of the coal stockpile to be identified are collected. Based on the batch boundaries of different batches of coal samples in the batch identification information database, the batch to which the coal stockpile belongs is determined. Specifically, based on the spatial location data, stacking status data, and environmental data of the coal stockpile to be identified, the distance between it and the batch boundaries of different batches of coal samples in the batch identification information database is calculated to determine the batch to which the coal stockpile belongs. The calculation method can be to minimize the sum of squared errors (SSE) within the cluster, defined as: Where k is the number of clusters, C i For the i-th cluster, μ i For clusters The center point, x, is a data point. The number of clusters, k, can be determined using the Elbow Method or the Davies-Bouldin index (DBI), for example, k=6 in the pilot to minimize the DBI and ensure cluster segregation and compactness.

[0050] like Figure 3 As shown, the present invention also provides a batch identification method for coal stockpiles based on identifiers, which further includes: S6 associates the collected spatial location data, stacking status data, and environmental data of the coal storage yard to be identified with the batch information of the coal storage yard to be identified in the Industrial Internet Identifier to form a full life cycle information database of the batch, realizing real-time query and full-chain traceability of coal batch information. Moreover, the batch identification information database supports dynamic updates.

[0051] The Industrial Internet Identifier Resolution System provides real-time batch information query and full-chain traceability to meet management needs. The traceability mechanism supports data sharing with upstream and downstream supply chain units (such as transportation, sales, and user units), and enables cross-system queries through Handle resolution nodes, with a response time of less than 1 second.

[0052] By combining industrial Internet identification technology, Internet of Things devices, and advanced algorithms, the present invention has significantly improved the efficiency and accuracy of coal yard management in power plants. In the scenario of thermal power plants, the life cycle of coal is "entering the factory -> storage -> feeding into the furnace for combustion". For the closed and controllable scenario of the coal yard in thermal power plants, it provides high-precision batch boundary recognition (pilot accuracy of 98.5%) and full-life-cycle traceability based on industrial Internet identification (from entering the yard -> storage -> reclaiming -> feeding into the furnace). The traceability scope covers the upstream of the supply chain (suppliers, transportation units) and the whole process within the factory. Without extending to the terminal decentralized sales link, significant management benefits can be generated within the thermal power plant. The present invention combines Internet of Things devices and clustering algorithms to solve the pain points in the dynamic blending combustion scenario of the coal yard in thermal power plants, and can provide an accurate digital foundation for the subsequent Energy Management System (EMS). It can real-time inform the EMS which batch the coal currently being grabbed by the reclaiming machine belongs to and what its calorific value is, so as to guide accurate coal blending combustion, improve combustion efficiency and reduce carbon emissions, with obvious positive effects. Specifically, the following effects can be achieved: High-precision batch identification: By adopting the K-means clustering algorithm and batch matching algorithm, combined with the physical properties of coal (such as density of 1.4 g / cm³, particle size distribution of 0.5 - 50 mm), in the 5000-square-meter yard pilot, the batch identification accuracy rate reaches 98.5%, which is 13.5 percentage points higher than that of traditional manual identification (accuracy rate of about 85%), and the batch confusion rate is reduced by about 80%. Efficient full-chain traceability: Based on the industrial Internet identification and resolution system, the query response time of batch information is shortened to 0.3 seconds, and the traceability accuracy rate reaches 99.8%. Compared with traditional traceability (2 - 3 hours per batch), the efficiency is increased by more than 99%, realizing full-life-cycle management from production to use. Automated management: By deploying 10 RFID readers, 5 laser rangefinders, and 8 cameras, 1000 groups of spatial data (accuracy of ±5 cm) are collected per minute, about 1 GB of data is generated daily, the system availability reaches 99.9%, and the manual verification workload is reduced from 10 people per day to 2 people, reducing the labor cost by 80%. Data interconnection: Through the industrial Internet identification system, the data sharing response time between the upstream and downstream supply chains is less than 1 second, and the data consistency reaches 99.8%, effectively solving the data island problem and significantly improving the supply chain collaboration efficiency.

[0053] In this invention, a unique Industrial Internet identifier code is assigned to each batch of coal samples. Spatial location data, stacking status data, and environmental data of each batch of coal samples are collected. These data are used as input to a spatial segmentation algorithm, with different batches of coal samples serving as the algorithm's output, thus defining the batch spatial boundaries of different batches. The defined batch spatial boundaries are then associated with the Industrial Internet identifier code to establish a batch identifier information database. Finally, spatial location data, stacking status data, and environmental data of the coal sample storage area to be identified are collected. Based on the batch boundaries of different batches of coal samples in the batch identifier information database, the batch to which the coal sample storage area belongs is determined. This effectively solves the problem of low efficiency and reliability in batch identification of coal storage areas in power plants caused by existing technologies, and significantly improves the efficiency and reliability of batch identification in power plants.

[0054] The identification code in the technical solution of this invention also includes coal source information, coal production time information, and a check code. The identification code is associated with and bound to the quality parameter information of the coal stockpile. The quality parameters are not directly written into the Handle identifier, but are pointed to by the identifier in the database record. After the identifier is parsed, the batch data record can be accessed. This method conforms to the minimum coding principle of industrial Internet identifiers, avoids excessively long identifiers, and supports dynamic updates of quality parameters.

[0055] In this invention, the spatial location data is a weighted fusion of first spatial location data and second spatial location data. The first spatial location data is spatial location data collected by lidar, and the second spatial location data is spatial location data obtained by depth reconstruction of image information collected by a camera. The stacking status data includes the average density of the coal pile, the particle size distribution of the coal pile, and the height of the coal pile. This ensures both the continuity of the coal pile surface and maintains high-precision distance information. These parameters form a multi-dimensional feature vector, ensuring the reliability of batch identification of coal stockpiles in power plants.

[0056] In the technical solution of this invention, the weights corresponding to different features are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategy. This improves the adaptability of batch identification of coal stockpiles in power plants.

[0057] In this invention, the spatial location data, stacking status data, and environmental data of the coal storage yard to be identified are associated with the batch information of the coal storage yard in the Industrial Internet identifier. This enables real-time querying and full-chain traceability of coal batch information. Furthermore, the batch identifier information database supports dynamic updates, achieving real-time querying and full-chain traceability of batch information to meet management needs. The traceability mechanism supports data sharing with upstream and downstream supply chain units (such as transportation, sales, and user units), effectively solving the data silo problem, significantly improving supply chain collaboration efficiency, and realizing full lifecycle management from production to use.

[0058] Example 2 like Figure 4 As shown, the present invention also provides a batch identification system for coal stockpiles based on identification, comprising: The identification management module assigns a unique Industrial Internet identification code to each batch of coal samples, and the identification code includes coal batch information. The data acquisition module collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile. The batch segmentation module (i.e., a sub-module of the batch identification module) takes the spatial location data, stacking status data, and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to the spatial segmentation algorithm, thus dividing the batch spatial boundaries of different batches of coal samples. The data processing module associates the batch spatial boundaries of different batches of coal samples with industrial internet identification codes to establish a batch identification information database. The batch identification module collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. Based on the batch boundaries of different batches of coal samples in the batch identification information database, it determines the batch to which the coal stockpile to be identified belongs.

[0059] Preferably, the coal stockpile batch identification system based on identifiers provided in this embodiment of the invention further includes: The query and traceability module supports querying and full-chain traceability of batch information of coal stockpiles to be identified.

[0060] The identifier management module and the data processing module exchange JSON data via a RESTful API (POST / associate); the data acquisition module and the batch identification module use MQTT topic subscription (topic: / coal / data) to transmit JSON-formatted environmental data. Anomaly handling includes: reallocating handle identifiers via a backup suffix when conflicts occur; triggering a re-acquisition mechanism (5-second interval, 3 retries) when data acquisition fails; and locally caching data during network interruptions and synchronizing it upon recovery. In boundary scenarios such as overlapping stockpile spaces, a variant of the DBSCAN algorithm is used to handle noisy batches.

[0061] It should be noted that the implementation process of each module in this embodiment corresponds to the method steps in Embodiment 1, and this embodiment does not impose any limitations on it.

[0062] In this invention, a unique Industrial Internet identifier code is assigned to each batch of coal samples. Spatial location data, stacking status data, and environmental data of each batch of coal samples are collected. These data are used as input to a spatial segmentation algorithm, with different batches of coal samples serving as the algorithm's output, thus defining the batch spatial boundaries of different batches. The defined batch spatial boundaries are then associated with the Industrial Internet identifier code to establish a batch identifier information database. Finally, spatial location data, stacking status data, and environmental data of the coal sample storage area to be identified are collected. Based on the batch boundaries of different batches of coal samples in the batch identifier information database, the batch to which the coal sample storage area belongs is determined. This effectively solves the problem of low efficiency and reliability in batch identification of coal storage areas in power plants caused by existing technologies, and significantly improves the efficiency and reliability of batch identification in power plants.

[0063] The identification code in the technical solution of this invention also includes coal source information, coal production time information, and a check code. The identification code is associated with and bound to the quality parameter information of the coal stockpile. The quality parameters are not directly written into the Handle identifier, but are pointed to by the identifier in the database record. After the identifier is parsed, the batch data record can be accessed. This method conforms to the minimum coding principle of industrial Internet identifiers, avoids excessively long identifiers, and supports dynamic updates of quality parameters.

[0064] In this invention, the spatial location data is a weighted fusion of first spatial location data and second spatial location data. The first spatial location data is spatial location data collected by lidar, and the second spatial location data is spatial location data obtained by depth reconstruction of image information collected by a camera. The stacking status data includes the average density of the coal pile, the particle size distribution of the coal pile, and the height of the coal pile. This ensures both the continuity of the coal pile surface and maintains high-precision distance information. These parameters form a multi-dimensional feature vector, ensuring the reliability of batch identification of coal stockpiles in power plants.

[0065] In the technical solution of this invention, the weights corresponding to different features are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategy. This improves the adaptability of batch identification of coal stockpiles in power plants.

[0066] In this invention, the spatial location data, stacking status data, and environmental data of the coal storage yard to be identified are associated with the batch information of the coal storage yard in the Industrial Internet identifier. This enables real-time querying and full-chain traceability of coal batch information. Furthermore, the batch identifier information database supports dynamic updates, achieving real-time querying and full-chain traceability of batch information to meet management needs. The traceability mechanism supports data sharing with upstream and downstream supply chain units (such as transportation, sales, and user units), effectively solving the data silo problem, significantly improving supply chain collaboration efficiency, and realizing full lifecycle management from production to use.

[0067] Example 3 like Figure 5 As shown, the present invention also provides a batch identification device for coal stockpiles based on identification, comprising: In the cloud, an identifier resolution system is set up. The identifier resolution system assigns a unique industrial internet identifier code to each batch of coal samples. The identifier code includes coal batch information and is stored and queried based on the batch information of the coal stockpile. The IoT data acquisition device collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile, and also collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. The edge computing device takes the spatial location data, stacking status data and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to the spatial segmentation algorithm to divide the batch spatial boundaries of different batches of coal samples. The batch spatial boundaries of different batches of coal samples after segmentation are associated with industrial internet identification codes to establish a batch identification information database; based on the batch boundaries of different batches of coal samples in the batch identification information database, the batch to which the coal stockpile to be identified belongs is determined.

[0068] The cloud-based identifier resolution system generates a unique Industrial Internet identifier code for each batch of coal at a coal storage yard. The format is "Enterprise Code + Production Time + Batch Number", for example, "CN123-20250701-B001". The identifier is stored in the cloud-based identifier resolution system. IoT data collection devices, namely RFID readers (for coal yard asset ID, production time, coal source and other attribute information), laser rangefinders and cameras, are deployed in the coal yard to collect spatial coordinates, stacking height and surface images in real time. Edge computing devices perform batch segmentation and batch identification: K-means clustering is used to segment spatial data, and combined with identification information, the spatial boundaries of each batch of coal are determined. The algorithm considers the physical properties of the coal (such as density and particle size) to improve segmentation accuracy. Input parameters include coordinates and characteristic values, and the distance metric is weighted Euclidean distance. For overlapping boundary scenarios, DBSCAN is used for optimization.

[0069] The cloud-based identifier resolution system communicates with edge computing devices to associate spatial data, environmental data, and identifier codes, forming a batch information database stored in a distributed database. Data flows are transmitted via a RESTful API and support shared queries between upstream and downstream systems, with local caching in case of anomalies such as network outages.

[0070] In this invention, a unique Industrial Internet identifier code is assigned to each batch of coal samples. Spatial location data, stacking status data, and environmental data of each batch of coal samples are collected. These data are used as input to a spatial segmentation algorithm, with different batches of coal samples serving as the algorithm's output, thus defining the batch spatial boundaries of different batches. The defined batch spatial boundaries are then associated with the Industrial Internet identifier code to establish a batch identifier information database. Finally, spatial location data, stacking status data, and environmental data of the coal sample storage area to be identified are collected. Based on the batch boundaries of different batches of coal samples in the batch identifier information database, the batch to which the coal sample storage area belongs is determined. This effectively solves the problem of low efficiency and reliability in batch identification of coal storage areas in power plants caused by existing technologies, and significantly improves the efficiency and reliability of batch identification in power plants.

[0071] The identification code in the technical solution of this invention also includes coal source information, coal production time information, and a check code. The identification code is associated with and bound to the quality parameter information of the coal stockpile. The quality parameters are not directly written into the Handle identifier, but are pointed to by the identifier in the database record. After the identifier is parsed, the batch data record can be accessed. This method conforms to the minimum coding principle of industrial Internet identifiers, avoids excessively long identifiers, and supports dynamic updates of quality parameters.

[0072] In this invention, the spatial location data is a weighted fusion of first spatial location data and second spatial location data. The first spatial location data is spatial location data collected by lidar, and the second spatial location data is spatial location data obtained by depth reconstruction of image information collected by a camera. The stacking status data includes the average density of the coal pile, the particle size distribution of the coal pile, and the height of the coal pile. This ensures both the continuity of the coal pile surface and maintains high-precision distance information. These parameters form a multi-dimensional feature vector, ensuring the reliability of batch identification of coal stockpiles in power plants.

[0073] In the technical solution of this invention, the weights corresponding to different features are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategy. This improves the adaptability of batch identification of coal stockpiles in power plants.

[0074] In this invention, the spatial location data, stacking status data, and environmental data of the coal storage yard to be identified are associated with the batch information of the coal storage yard in the Industrial Internet identifier. This enables real-time querying and full-chain traceability of coal batch information. Furthermore, the batch identifier information database supports dynamic updates, achieving real-time querying and full-chain traceability of batch information to meet management needs. The traceability mechanism supports data sharing with upstream and downstream supply chain units (such as transportation, sales, and user units), effectively solving the data silo problem, significantly improving supply chain collaboration efficiency, and realizing full lifecycle management from production to use.

[0075] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A batch identification method for coal stockpiles based on identifiers, characterized in that, include: Each batch of coal samples is assigned a unique Industrial Internet identification code, which includes coal batch information; Spatial location data, stacking status data, and environmental data of each batch of coal samples were collected from the stockpile. The spatial location data, stacking status data, and environmental data of the coal sample stockpile are used as inputs to the spatial segmentation algorithm, and different batches of the coal sample stockpile are used as outputs to delineate the batch spatial boundaries of different batches of coal samples. The batch spatial boundaries of different batches of coal samples after division are associated with industrial internet identification codes to establish a batch identification information database. Collect spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. Based on the batch boundaries of different batches of coal samples in the batch identification information database, determine the batch to which the coal stockpile to be identified belongs.

2. The method for batch identification of coal stockpiles based on identification according to claim 1, characterized in that, The identification code also includes coal source information, coal production time information, and a check code. The identification code is associated with and bound to the quality parameter information of the coal storage yard.

3. The method for batch identification of coal stockpiles based on identification according to claim 1, characterized in that, The spatial location data is a weighted fusion of the first spatial location data and the second spatial location data. The first spatial location data is the spatial location data collected by the LiDAR, and the second spatial location data is the spatial location data obtained by the camera through depth reconstruction of the image information collected.

4. The method for batch identification of coal stockpiles based on identification according to claim 1, characterized in that, Stacking status data includes average coal density, coal particle size distribution, and coal height.

5. The method for batch identification of coal stockpiles based on identification according to claim 4, characterized in that, The spatial location data, stacking status data, and environmental data of the coal sample stockpile are used as inputs to the spatial segmentation algorithm, while different batches of the coal sample stockpile are used as outputs. The specific steps for defining the batch spatial boundaries of different batches of coal samples include: Randomly initialize k clusters and k cluster center points, with each cluster corresponding to a batch of coal samples; Each sample data point is assigned to the nearest center by using the weighted distance between the sample data point and the center point within the cluster; Update the cluster center point to the mean of all data points in the cluster until a preset termination condition is met; the preset termination condition is that the preset number of iterations is met or the objective function value is less than a preset function threshold.

6. The method for batch identification of coal stockpiles based on identification according to claim 5, characterized in that, The specific method for calculating the weighted distance between sample data points and cluster centroids is as follows: in, x is the weighted distance between the sample data points and the cluster center. n Let y be the value of the nth feature in the sample data points. n Let be the value of the nth feature among the cluster center points. The weight is the weight corresponding to the nth feature; and the weights corresponding to different features are dynamically adjusted according to the coal type, stacking conditions and enterprise management strategies.

7. A method for batch identification of coal stockpiles based on identification according to any one of claims 1-6, characterized in that, Also includes: The collected spatial location data, stacking status data, and environmental data of the coal storage yard to be identified are associated with the batch information of the coal storage yard to be identified in the Industrial Internet Identifier, so as to realize real-time query and full-chain traceability of coal storage yard batch information. Moreover, the batch identification information database supports dynamic updates.

8. A batch identification system for coal stockpiles based on identifiers, characterized in that, include: The identification management module assigns a unique Industrial Internet identification code to each batch of coal samples, and the identification code includes coal batch information. The data acquisition module collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile. The batch segmentation module takes the spatial location data, stacking status data, and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to the spatial segmentation algorithm, thus dividing the batch spatial boundaries of different batches of coal samples. The data processing module associates the batch spatial boundaries of different batches of coal samples with industrial internet identification codes to establish a batch identification information database. The batch identification module collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. Based on the batch boundaries of different batches of coal samples in the batch identification information database, it determines the batch to which the coal stockpile to be identified belongs.

9. A batch identification system for coal stockpiles based on identification as described in claim 8, characterized in that, Also includes: The query and traceability module supports querying and full-chain traceability of batch information of coal stockpiles to be identified.

10. A batch identification device for coal stockpiles based on identification tags, characterized in that, include: In the cloud, an identifier resolution system is set up. The identifier resolution system assigns a unique industrial internet identifier code to each batch of coal samples. The identifier code includes coal batch information and is stored and queried based on the batch information of the coal stockpile. The IoT data acquisition device collects spatial location data, stacking status data, and environmental data of each batch of coal sample stockpile, and also collects spatial location data, stacking status data, and environmental data of the coal stockpile to be identified. The edge computing device takes the spatial location data, stacking status data and environmental data of the coal sample stockpile as input to the spatial segmentation algorithm, and takes different batches of the coal sample stockpile as output to the spatial segmentation algorithm to divide the batch spatial boundaries of different batches of coal samples. The batch spatial boundaries of different batches of coal samples after segmentation are associated with industrial internet identification codes to establish a batch identification information database; based on the batch boundaries of different batches of coal samples in the batch identification information database, the batch to which the coal stockpile to be identified belongs is determined.

Images