Smart warehouse management method and system based on Internet of Things technology

Abstract

The invention discloses a smart warehouse management method and system based on the Internet of Things technology, and relates to the technical field of monitoring analysis, and the method comprises the steps: collecting global sensing data corresponding to a warehouse, and generating a warehouse data set; carrying out three-dimensional space modeling based on the warehouse data set, constructing a warehouse dynamic digital twinborn model, carrying out warehouse material distribution feature extraction, and generating a material layout topological graph; obtaining a risk incidence matrix based on the material layout topological graph and the global perception data; carrying out abnormal event probability prediction on the storehouse dynamic digital twinborn model, generating a risk event prediction result, carrying out propagation path analysis on the risk event prediction result, obtaining a risk diffusion gradient map based on an analysis result, and outputting a local risk thermodynamic diagram. And configuring an adaptive management strategy for the warehouse dynamic digital twinborn model based on the local risk thermodynamic diagram, deploying the adaptive management strategy to a warehouse Internet of Things execution terminal, and generating a real-time control instruction set. The method has the effect of improving the warehouse management efficiency.

Description

Technical Field

[0001] This application relates to the field of monitoring and analysis technology, and in particular to a smart warehouse management method and system based on Internet of Things (IoT) technology. Background Technology

[0002] Modern enterprise warehouses have become logistics centers, making warehouse security management paramount. Smart warehouse management is an intelligent warehouse management system built upon next-generation information technologies such as the Internet of Things, big data, artificial intelligence, and 5G. Its core lies in deploying terminal devices such as smart sensors, RFID tags, visual recognition equipment, and intelligent sorting robots to collect multi-dimensional data in real time, including the location of goods, inventory quantity, environmental temperature and humidity, and equipment operating status within the warehouse. This data is then integrated across systems and deeply analyzed through edge computing and cloud platforms.

[0003] In related technologies, current warehouse management relies heavily on isolated sensing devices for data acquisition, which leads to problems such as limited data collection dimensions, difficulty in integrating multi-source information, and lagging dynamic risk assessment, thereby reducing the efficiency of warehouse management and leaving room for improvement. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application provides a smart warehouse management method and system based on Internet of Things (IoT) technology.

[0005] Firstly, this application provides a smart warehouse management method based on Internet of Things (IoT) technology, comprising the following steps: Step S1: Collect real-time omni-channel perception data of the warehouse through the Internet of Things sensor network. The omni-channel perception data includes environmental parameter data, warehouse material distribution data and equipment operation status data. Perform heterogeneous data source alignment processing on the omni-channel perception data to generate a warehouse dataset. Step S2: Based on the warehouse dataset, perform 3D spatial modeling to construct a dynamic digital twin model of the warehouse. Extract the distribution features of stored materials from the dynamic digital twin model of the warehouse to generate a material layout topology map. Based on the material layout topology map and the full-domain perception data, perform warehouse safety risk coupling analysis to obtain a risk correlation matrix. Step S3: Based on the risk correlation matrix, perform abnormal event probability prediction on the warehouse dynamic digital twin model, generate risk event prediction results, perform propagation path analysis on the risk event prediction results, obtain a risk diffusion gradient map based on the analysis results, and then dynamically classify the risk level of the warehouse area based on the risk diffusion gradient map, and output a local risk heat map. Step S4: Configure an adaptive management strategy for the warehouse dynamic digital twin model based on the local risk heat map, and deploy the adaptive management strategy to the warehouse IoT execution terminal to generate a real-time control instruction set.

[0006] Preferably, step S1 specifically includes the following steps: Step S11: Collect warehouse environmental parameter data using temperature and humidity sensors, gas concentration sensors, and infrared imaging equipment. The environmental parameter data includes temperature field distribution data, humidity change gradient data, smoke concentration data, and heat source anomaly signals. Collect warehouse material distribution data using RFID tag readers, weight sensors, and machine vision equipment. The warehouse material distribution data includes material ID data, material stacking coordinate data, material weight load data, and material stacking tilt angle data. Collect equipment operating status data corresponding to the equipment in the warehouse using equipment status monitoring sensors. The equipment operating status data includes motor vibration spectrum data, energy consumption curve data, and fault alarm signals. Step S12: Align the environmental parameter data, warehouse material distribution data, and equipment operation status data, and generate a warehouse dataset based on the knowledge graph multimodal data fusion method after aligning the overall perception data.

[0007] Preferably, the process of constructing a dynamic digital twin model of the warehouse includes: The warehouse structure is scanned using a 3D LiDAR scanner to generate a warehouse point cloud model. Combined with the distribution data of stored materials, the coordinate data of the stacked materials and the bounding box information are marked in the warehouse point cloud model. The bounding box information includes material ID data, material weight load data and material stacking tilt angle data. Extract historical equipment operation status data corresponding to the equipment in the warehouse from the cloud database, and restore the motion path data and motion speed curve data corresponding to the equipment in the warehouse in the warehouse dynamic digital twin model based on the historical equipment operation status data corresponding to the equipment in the warehouse. The dynamic digital twin model of the warehouse is optimized using a reinforcement learning algorithm, and the parameters corresponding to the dynamic digital twin model of the warehouse are dynamically adjusted.

[0008] Preferably, a risk correlation matrix is ​​obtained by performing a coupled analysis of warehouse security risks based on the material layout topology map and the overall perception data, specifically including: Obtain the material layout topology map corresponding to the warehouse, and identify key risk nodes based on the material layout topology map. The key risk nodes include the hazardous chemical storage area and the heavy material stacking area. Based on the corresponding environmental parameter data in the warehouse, the corresponding abnormal environmental areas in the warehouse are extracted, and the abnormal environmental areas are compared with the areas corresponding to the hazardous chemical storage areas to confirm the spatial overlap. Based on the environmental changes corresponding to the abnormal environmental areas in the warehouse, the corrosion risk coefficient of the hazardous chemical storage areas is calculated. Stability analysis was conducted on the heavy material storage area to determine the corresponding stability coefficient, and a correlation was established between equipment operating status data and the stability coefficient of the heavy material storage area. The correlation between the spatial overlap, the corrosion risk coefficient, and the equipment operating status data and the stability coefficient of the heavy material storage area is modeled to generate a risk correlation matrix.

[0009] Preferably, the propagation path analysis is performed on the risk event prediction results, and a risk diffusion gradient map is obtained based on the analysis results, specifically including: Based on the risk event prediction results, the risk events and their corresponding initial risk sources are identified; The spatiotemporal propagation path of risk events is simulated by a preset risk propagation model. The warehouse is divided into multiple sub-regions. Based on the spatiotemporal propagation path, the risk status data of each sub-region is identified. The risk transmission weight factor between adjacent sub-regions is introduced to calculate the diffusion speed and probability distribution of risk events from the initial risk source to the surrounding sub-regions. A risk diffusion gradient map is generated based on the diffusion speed and probability distribution of risk events from the initial risk source to the surrounding sub-regions.

[0010] Preferably, the adaptive management strategy includes optimizing the temperature and humidity control parameters of the air conditioning system, dynamically planning the storage and handling routes, and formulating emergency control rules for multi-device collaborative linkage.

[0011] Preferably, the method also includes evaluating the effectiveness of the strategy through the real-time control instruction set fed back by the warehouse IoT execution terminal, and adjusting and managing the dynamic digital twin model of the warehouse based on the results of the effectiveness evaluation.

[0012] Secondly, this application provides a smart warehouse management system based on Internet of Things (IoT) technology, comprising: The data acquisition module is used to collect real-time omni-channel perception data of the warehouse through the Internet of Things sensor network. The omni-channel perception data includes environmental parameter data, warehouse material distribution data and equipment operation status data. The omni-channel perception data is processed to align heterogeneous data sources and generate a warehouse dataset. The modeling module is used to perform three-dimensional spatial modeling based on the warehouse dataset, construct a dynamic digital twin model of the warehouse, extract the distribution features of stored materials from the dynamic digital twin model of the warehouse, and generate a material layout topology map; based on the material layout topology map and the full-domain perception data, a coupled analysis of warehouse safety risks is performed to obtain a risk correlation matrix; The analysis module is used to predict the probability of abnormal events in the dynamic digital twin model of the warehouse based on the risk correlation matrix, generate risk event prediction results, perform propagation path analysis on the risk event prediction results, obtain a risk diffusion gradient map based on the analysis results, and then dynamically classify the risk level of the warehouse area based on the risk diffusion gradient map, and output a local risk heat map. The execution module is used to configure adaptive management strategies for the dynamic digital twin model of the warehouse based on the local risk heat map, and to deploy the adaptive management strategies to the warehouse IoT execution terminal to generate a real-time control instruction set.

[0013] Thirdly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform any of the above-described smart warehouse management system based on Internet of Things technology.

[0014] In summary, this application includes the following beneficial technical effects: This application provides a smart warehouse management system based on Internet of Things (IoT) technology. It constructs a dynamic digital twin model of the warehouse by performing 3D spatial modeling based on warehouse datasets, thereby supporting remote monitoring and early warning. This effectively reduces the efficiency and cost of manual inspections. The system extracts the distribution characteristics of stored materials from the dynamic digital twin model, generating a material layout topology map. Based on the material layout topology map and global perception data, it performs warehouse safety risk coupling analysis to obtain a risk correlation matrix. This effectively helps managers focus on highly coupled risk combinations, prioritize the allocation of control resources, and predict the probability of abnormal events based on the risk correlation matrix. It then generates risk event prediction results, analyzes the propagation path of these predictions, and obtains a risk diffusion gradient map. Based on this gradient map, it dynamically classifies the risk levels of warehouse areas and outputs a local risk heat map. Based on the local risk heat map, it configures an adaptive management strategy for the dynamic digital twin model and deploys this strategy to the warehouse IoT execution terminal, generating a real-time control instruction set. This effectively reduces the decrease in warehouse management efficiency caused by problems such as single data collection dimensions, difficulties in multi-source information fusion, and delayed dynamic risk assessment. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1This is a schematic diagram of a smart warehouse management system based on Internet of Things (IoT) technology, according to an embodiment of this application.

[0017] Figure 2 This is a flowchart of a smart warehouse management method based on Internet of Things (IoT) technology, according to an embodiment of this application. Detailed Implementation

[0018] The following is in conjunction with the appendix Figure 1-2 This application will be described in further detail.

[0019] Example 1 This application discloses a smart warehouse management method based on Internet of Things (IoT) technology.

[0020] Reference Figure 1 A smart warehouse management method based on Internet of Things (IoT) technology includes the following steps: Step S1: Collect real-time omni-channel perception data of the warehouse through the Internet of Things sensor network. The omni-channel perception data includes environmental parameter data, warehouse material distribution data and equipment operation status data. Perform heterogeneous data source alignment processing on the omni-channel perception data to generate a warehouse dataset. Step S2: Based on the warehouse dataset, perform 3D spatial modeling to construct a dynamic digital twin model of the warehouse. Extract the distribution features of stored materials from the dynamic digital twin model of the warehouse to generate a material layout topology map. Based on the material layout topology map and the full-domain perception data, perform warehouse safety risk coupling analysis to obtain a risk correlation matrix. Step S3: Based on the risk correlation matrix, perform abnormal event probability prediction on the warehouse dynamic digital twin model, generate risk event prediction results, perform propagation path analysis on the risk event prediction results, obtain a risk diffusion gradient map based on the analysis results, and then dynamically classify the risk level of the warehouse area based on the risk diffusion gradient map, and output a local risk heat map. Step S4: Configure an adaptive management strategy for the warehouse dynamic digital twin model based on the local risk heat map, and deploy the adaptive management strategy to the warehouse IoT execution terminal to generate a real-time control instruction set.

[0021] It should be noted that step S1 specifically includes the following steps: Step S11: Collect warehouse environmental parameter data using temperature and humidity sensors, gas concentration sensors, and infrared imaging equipment. The environmental parameter data includes temperature field distribution data, humidity change gradient data, smoke concentration data, and heat source anomaly signals. Collect warehouse material distribution data using RFID tag readers, weight sensors, and machine vision equipment. The warehouse material distribution data includes material ID data, material stacking coordinate data, material weight load data, and material stacking tilt angle data. Collect equipment operating status data corresponding to the equipment in the warehouse using equipment status monitoring sensors. The equipment operating status data includes motor vibration spectrum data, energy consumption curve data, and fault alarm signals. Step S12: Align the environmental parameter data, warehouse material distribution data, and equipment operation status data, and generate a warehouse dataset based on the knowledge graph multimodal data fusion method after aligning the overall perception data.

[0022] The alignment process includes using a dynamic time warping algorithm to eliminate the differences in the acquisition delay of the aforementioned sensor data, and transforming the multi-sensor coordinate system to the global reference coordinate system of the warehouse through spatial coordinate transformation. The warehouse multi-source fusion dataset includes timestamps, spatial coordinates, and data credibility weights.

[0023] It should be noted that the process of building a dynamic digital twin model of the warehouse specifically includes: The warehouse structure is scanned using a 3D LiDAR scanner to generate a warehouse point cloud model. Combined with the distribution data of stored materials, the coordinate data of the stacked materials and the bounding box information are marked in the warehouse point cloud model. The bounding box information includes material ID data, material weight load data and material stacking tilt angle data. Extract historical equipment operation status data corresponding to the equipment in the warehouse from the cloud database, and restore the motion path data and motion speed curve data corresponding to the equipment in the warehouse in the warehouse dynamic digital twin model based on the historical equipment operation status data corresponding to the equipment in the warehouse. The dynamic digital twin model of the warehouse is optimized by using a reinforcement learning algorithm, and the parameters corresponding to the dynamic digital twin model of the warehouse are dynamically adjusted to match the changes in the spatial structure of the actual warehouse.

[0024] Specifically, a 3D LiDAR is used to perform a full-range scan of the warehouse. The distance to the target is measured by laser pulses, and a massive number of 3D coordinate points of the warehouse walls, shelves, aisles, and other structures are obtained to generate high-density point cloud data. Then, point cloud filtering and noise reduction are performed to remove outliers. Mesh modeling technology is then used to transform the point cloud into an intuitive dynamic digital twin model of the warehouse. Combined with material distribution data, precise coordinates are marked for each material in the dynamic digital twin model of the warehouse, and a bounding box is drawn. The bounding box embeds attribute data such as material ID, weight load, and stacking tilt angle to form a digital stacking location with semantic information. The above steps realize the accurate mapping from the physical space of the warehouse to the digital space, providing a visual basis for subsequent analysis. Managers can quickly locate high-load areas or stacks of materials that are tilted beyond the limit and provide early warning of safety hazards. A reinforcement learning framework is constructed, with warehouse space utilization, equipment energy consumption, and operational efficiency as optimization objectives. Adjustments to shelf layout, aisle width, and equipment scheduling rules are used as adjustable parameters. By simulating warehouse operation scenarios under different parameter combinations, the merits of each solution are evaluated using a reward function, gradually converging to the optimal parameter configuration. When the actual warehouse spatial structure changes due to the addition of shelves or adjustments to aisles, LiDAR scans in real time to acquire new point cloud data, triggering dynamic updates to the parameters of the warehouse's dynamic digital twin model.

[0025] It should be noted that a risk correlation matrix was obtained by performing a coupled analysis of warehouse security risks based on the material layout topology map and the overall perception data, which specifically includes: Obtain the material layout topology map corresponding to the warehouse, and identify key risk nodes based on the material layout topology map. The key risk nodes include the hazardous chemical storage area and the heavy material stacking area. Based on the corresponding environmental parameter data in the warehouse, the corresponding abnormal environmental areas in the warehouse are extracted, and the abnormal environmental areas are compared with the areas corresponding to the hazardous chemical storage areas to confirm the spatial overlap. Based on the environmental changes corresponding to the abnormal environmental areas in the warehouse, the corrosion risk coefficient of the hazardous chemical storage areas is calculated. Stability analysis was conducted on the heavy material storage area to determine the corresponding stability coefficient, and a correlation was established between equipment operating status data and the stability coefficient of the heavy material storage area. The correlation between the spatial overlap, the corrosion risk coefficient, and the equipment operating status data and the stability coefficient of the heavy material storage area is modeled to generate a risk correlation matrix.

[0026] Specifically, by using 3D LiDAR point cloud data and warehouse material distribution labeling information, a warehouse material layout topology map is constructed using graph theory algorithms. For example, in a warehouse topology map, node A is labeled as "hazardous chemical storage area" and node B is labeled as "heavy material stacking area". By using preset rules (such as hazardous chemical storage exceeding a threshold, or heavy material single stack weight > 1.5 tons), key risk nodes are automatically identified. The above steps transform the complex spatial structure of the warehouse into a computable graph model, which facilitates the rapid location of high-risk areas. For example, the topology map can intuitively reveal that the hazardous chemical storage area is too close to the electrical distribution box node, providing early warning of the risk caused by electrical sparks. Environmental parameter data within the warehouse is collected, and normal threshold ranges are set. A continuous environmental parameter field is generated from discrete sensor data using a spatial interpolation algorithm. Abnormal environmental areas exceeding the threshold are identified (e.g., humidity reaching 90% RH in the northwest corner of the warehouse, marked as a damp abnormality area). Spatial overlay analysis is performed between these abnormal environmental areas and the hazardous chemical storage area to calculate the spatial overlap (e.g., the overlap between an acid storage area and a damp area reaches 40%). A corrosion risk coefficient is calculated based on a corrosion kinetic model. For example, for every 5% increase in humidity (RH), the corrosion rate of metal containers increases by 8%, and the corrosion risk coefficient K=0.6 (threshold 0.5 is the danger level), triggering an early warning for anti-corrosion coating inspection. This process achieves spatial quantification of environmental risk, preventing leaks of hazardous chemicals due to environmental corrosion. For heavy material storage areas, finite element analysis (FEA) is used to simulate the load-bearing state of the racks or floor, calculating structural stress distribution, deformation displacement, and other indicators to generate a stability coefficient S (e.g., S=0.9 indicates structural safety, S<0.8 requires reinforcement). For example, a heavy rack storing three layers of machine tool components (total weight 6 tons) showed that the stress on the bottom rack beams reached 200 MPa (close to the material yield strength of 235 MPa), with a stability coefficient S=0.78, indicating an overload risk. Simultaneously, operational data from equipment such as stacker cranes and hoists in the heavy-duty area are extracted, and correlation analysis is used to establish the relationship between equipment operating parameters and the stability coefficient. For instance, it was found that when the stacker crane starts and stops more than 50 times per day, the rack stability coefficient S decreases by an average of 0.05, showing a negative correlation (r=-0.72), indicating that frequent operation exacerbates structural fatigue. A risk coupling model is constructed by using a Bayesian network with spatial overlap, corrosion risk coefficient, and stability correlation as input variables. The model parameters are trained using historical accident data to simulate the interaction of multiple factors and finally generate a risk correlation matrix. For example, in the risk correlation matrix of a warehouse, the cell where "overlap between damp abnormal area and hazardous chemical area is 50%" and "stacking crane starts and stops 60 times a day" shows a risk level of III (the highest level is V), indicating that ventilation and dehumidification and equipment vibration reduction modification need to be carried out simultaneously.

[0027] It should be noted that a propagation path analysis is performed on the risk event prediction results, and a risk diffusion gradient map is obtained based on the analysis results, specifically including: Based on the risk event prediction results, risk events and their corresponding initial risk sources are identified. The initial risk sources include, but are not limited to, the fire initiation point, the water leakage spread area, and the impact range of equipment failure. The spatiotemporal propagation path of risk events is simulated by a preset risk propagation model. The warehouse is divided into multiple sub-regions. Based on the spatiotemporal propagation path, the risk status data of each sub-region is identified. The risk transmission weight factor between adjacent sub-regions is introduced to calculate the diffusion speed and probability distribution of risk events from the initial risk source to the surrounding sub-regions. A risk diffusion gradient map is generated based on the diffusion speed and probability distribution of risk events from the initial risk source to the surrounding sub-regions.

[0028] In the risk diffusion gradient map, the risk level is visually displayed through the intensity of color.

[0029] Furthermore, the configuration adaptive management strategy includes optimizing the temperature and humidity control parameters of the air conditioning system, dynamically planning warehouse handling routes, and formulating emergency control rules for multi-device collaborative linkage.

[0030] Specifically, the system optimizes temperature and humidity control parameters of the air conditioning system based on fuzzy control algorithms. Combining the temperature field distribution and humidity gradient within the warehouse, it dynamically adjusts cooling power, air supply speed, and dehumidification frequency to ensure that stored goods are in suitable environmental conditions. The A* algorithm is used for dynamic planning of storage and handling paths, prioritizing avoidance of high-risk areas while comprehensively considering handling efficiency and energy consumption costs to generate the optimal path solution. Emergency control rules for multi-device collaboration are formulated, including the coordinated shutdown of power in relevant areas, activation of fire extinguishing devices, and guidance of stacker cranes to evacuate hazardous materials when a fire hazard is detected. The above adaptive management strategies are deployed to IoT execution terminals in the warehouse to generate a real-time control command set containing a priority command queue, and achieve millisecond-level response through edge computing nodes to ensure the timeliness and accuracy of control commands.

[0031] Furthermore, it also includes evaluating the effectiveness of strategies through real-time control command sets fed back from warehouse IoT execution terminals, and adjusting and managing the dynamic digital twin model of the warehouse based on the results of the effectiveness evaluation.

[0032] Example 2 This application also discloses a smart warehouse management system based on Internet of Things (IoT) technology.

[0033] Reference Figure 2 A smart warehouse management system based on Internet of Things (IoT) technology includes: The data acquisition module is used to collect real-time omni-channel perception data of the warehouse through the Internet of Things sensor network. The omni-channel perception data includes environmental parameter data, warehouse material distribution data and equipment operation status data. The omni-channel perception data is processed to align heterogeneous data sources and generate a warehouse dataset. The modeling module is used to perform three-dimensional spatial modeling based on the warehouse dataset, construct a dynamic digital twin model of the warehouse, extract the distribution features of stored materials from the dynamic digital twin model of the warehouse, and generate a material layout topology map; based on the material layout topology map and the full-domain perception data, a coupled analysis of warehouse safety risks is performed to obtain a risk correlation matrix; The analysis module is used to predict the probability of abnormal events in the dynamic digital twin model of the warehouse based on the risk correlation matrix, generate risk event prediction results, perform propagation path analysis on the risk event prediction results, obtain a risk diffusion gradient map based on the analysis results, and then dynamically classify the risk level of the warehouse area based on the risk diffusion gradient map, and output a local risk heat map. The execution module is used to configure adaptive management strategies for the dynamic digital twin model of the warehouse based on the local risk heat map, and to deploy the adaptive management strategies to the warehouse IoT execution terminal to generate a real-time control instruction set.

[0034] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention, they should all fall within the protection scope of the present invention.

[0035] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," 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 invention. In this specification, 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.

[0036] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A smart warehouse management method based on Internet of Things (IoT) technology, characterized in that, Includes the following steps: Step S1: Collect real-time omni-channel perception data of the warehouse through the Internet of Things sensor network. The omni-channel perception data includes environmental parameter data, warehouse material distribution data and equipment operation status data. Perform heterogeneous data source alignment processing on the omni-channel perception data to generate a warehouse dataset. Step S2: Based on the warehouse dataset, perform three-dimensional spatial modeling, construct a dynamic digital twin model of the warehouse, extract the distribution features of warehouse materials from the dynamic digital twin model of the warehouse, and generate a material layout topology map; Based on the material layout topology map and the whole-domain perception data, a coupled analysis of warehouse safety risks was conducted to obtain a risk correlation matrix; Step S3: Based on the risk correlation matrix, perform abnormal event probability prediction on the warehouse dynamic digital twin model, generate risk event prediction results, perform propagation path analysis on the risk event prediction results, obtain a risk diffusion gradient map based on the analysis results, and then dynamically classify the risk level of the warehouse area based on the risk diffusion gradient map, and output a local risk heat map. Step S4: Configure an adaptive management strategy for the warehouse dynamic digital twin model based on the local risk heat map, and deploy the adaptive management strategy to the warehouse IoT execution terminal to generate a real-time control instruction set.

2. The smart warehouse management method based on Internet of Things technology according to claim 1, characterized in that, Step S1 specifically includes the following steps: Step S11: Collect warehouse environmental parameter data using temperature and humidity sensors, gas concentration sensors, and infrared imaging equipment. The environmental parameter data includes temperature field distribution data, humidity change gradient data, smoke concentration data, and heat source anomaly signals. Collect warehouse material distribution data using RFID tag readers, weight sensors, and machine vision equipment. The warehouse material distribution data includes material ID data, material stacking coordinate data, material weight load data, and material stacking tilt angle data. Collect equipment operating status data corresponding to the equipment in the warehouse using equipment status monitoring sensors. The equipment operating status data includes motor vibration spectrum data, energy consumption curve data, and fault alarm signals. Step S12: Align the environmental parameter data, warehouse material distribution data, and equipment operation status data, and generate a warehouse dataset based on the knowledge graph multimodal data fusion method after aligning the overall perception data.

3. The smart warehouse management method based on Internet of Things technology according to claim 1, characterized in that, The process of constructing a dynamic digital twin model of the warehouse specifically includes: The warehouse structure is scanned using a 3D LiDAR scanner to generate a warehouse point cloud model. Combined with the distribution data of stored materials, the coordinate data of the stacked materials and the bounding box information are marked in the warehouse point cloud model. The bounding box information includes material ID data, material weight load data and material stacking tilt angle data. Extract historical equipment operation status data corresponding to the equipment in the warehouse from the cloud database, and restore the motion path data and motion speed curve data corresponding to the equipment in the warehouse in the warehouse dynamic digital twin model based on the historical equipment operation status data corresponding to the equipment in the warehouse. The dynamic digital twin model of the warehouse is optimized using a reinforcement learning algorithm, and the parameters corresponding to the dynamic digital twin model of the warehouse are dynamically adjusted.

4. The smart warehouse management method based on Internet of Things technology according to claim 1, characterized in that, Based on the material layout topology map and the overall perception data, a coupled analysis of warehouse security risks was conducted to obtain a risk correlation matrix, which specifically includes: Obtain the material layout topology map corresponding to the warehouse, and identify key risk nodes based on the material layout topology map. The key risk nodes include the hazardous chemical storage area and the heavy material stacking area. Based on the corresponding environmental parameter data in the warehouse, the corresponding abnormal environmental areas in the warehouse are extracted, and the abnormal environmental areas are compared with the areas corresponding to the hazardous chemical storage areas to confirm the spatial overlap. Based on the environmental changes corresponding to the abnormal environmental areas in the warehouse, the corrosion risk coefficient of the hazardous chemical storage areas is calculated. Stability analysis was conducted on the heavy material storage area to determine the corresponding stability coefficient, and a correlation was established between equipment operating status data and the stability coefficient of the heavy material storage area. The correlation between the spatial overlap, the corrosion risk coefficient, and the equipment operating status data and the stability coefficient of the heavy material storage area is modeled to generate a risk correlation matrix.

5. The smart warehouse management method based on Internet of Things technology according to claim 1, characterized in that, The propagation path of risk event prediction results is analyzed, and a risk diffusion gradient map is obtained based on the analysis results, specifically including: Based on the risk event prediction results, the risk events and their corresponding initial risk sources are identified; The spatiotemporal propagation path of risk events is simulated by a preset risk propagation model. The warehouse is divided into multiple sub-regions. Based on the spatiotemporal propagation path, the risk status data of each sub-region is identified. The risk transmission weight factor between adjacent sub-regions is introduced to calculate the diffusion speed and probability distribution of risk events from the initial risk source to the surrounding sub-regions. A risk diffusion gradient map is generated based on the diffusion speed and probability distribution of risk events from the initial risk source to the surrounding sub-regions.

6. The smart warehouse management method based on Internet of Things technology according to claim 1, characterized in that, The adaptive management strategy includes optimizing the temperature and humidity control parameters of the air conditioning system, dynamically planning warehouse handling routes, and formulating emergency control rules for multi-device collaboration.

7. The smart warehouse management method based on Internet of Things technology according to claim 1, characterized in that, It also includes evaluating the effectiveness of strategies through real-time control command sets fed back from warehouse IoT execution terminals, and adjusting and managing the dynamic digital twin model of the warehouse based on the results of the effectiveness evaluation.

8. A smart warehouse management system based on Internet of Things (IoT) technology, applied to the smart warehouse management method based on IoT technology described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to collect real-time omni-channel perception data of the warehouse through the Internet of Things sensor network. The omni-channel perception data includes environmental parameter data, warehouse material distribution data and equipment operation status data. The omni-channel perception data is processed to align heterogeneous data sources and generate a warehouse dataset. The modeling module is used to perform three-dimensional spatial modeling based on the warehouse dataset, construct a dynamic digital twin model of the warehouse, extract the distribution features of warehouse materials from the dynamic digital twin model of the warehouse, and generate a material layout topology map. Based on the material layout topology map and the whole-domain perception data, a coupled analysis of warehouse safety risks was conducted to obtain a risk correlation matrix; The analysis module is used to predict the probability of abnormal events in the dynamic digital twin model of the warehouse based on the risk correlation matrix, generate risk event prediction results, perform propagation path analysis on the risk event prediction results, obtain a risk diffusion gradient map based on the analysis results, and then dynamically classify the risk level of the warehouse area based on the risk diffusion gradient map, and output a local risk heat map. The execution module is used to configure adaptive management strategies for the dynamic digital twin model of the warehouse based on the local risk heat map, and to deploy the adaptive management strategies to the warehouse IoT execution terminal to generate a real-time control instruction set.

9. A computer-readable storage medium, characterized in that: The system stores instructions that, when executed on a computer, cause the computer to perform a smart warehouse management system based on Internet of Things (IoT) technology as described in any one of claims 1 to 7.

Images