A 3D digital management system and method for grain depots based on digital twins

Abstract

The present invention relates to the technical field of grain depot management, and in particular, to a three-dimensional digital management system and method for a grain depot based on digital twins. The method includes the following steps: acquiring grain depot structure data and grain depot environment data; performing three-dimensional refined modeling based on the grain depot structure data to obtain a panoramic digital twin scene of the grain depot; adjusting the geospatial accuracy of the panoramic digital twin scene of the grain depot according to the grain depot environment data to obtain a bin 3D model; collecting grain depot information using IoT sensors and uploading the grain depot information to the bin 3D model for data integration and annotation to establish a data visualization interface; extracting the grain depot storage information of the data visualization interface to construct a virtual granary; and performing 3D particle simulation on the virtual granary. The present invention improves the accuracy and response speed of three-dimensional digital management through three-dimensional digital twin modeling, IoT data integration, particle simulation, and automated control.

Description

Technical Field

[0001] The present invention relates to the technical field of grain depot management, and in particular to a three-dimensional digital management system and method for grain depots based on digital twins. Background Art

[0002] In grain depot management, traditional methods usually rely on manual inspections and manual records, which are not only inefficient but also easily affected by human factors, making management difficult. With the development of big data and sensor technologies, grain depots have begun to monitor the grain storage environment through sensors, such as temperature, humidity, gas composition, etc., providing a basis for subsequent data analysis and decision-making. The introduction of digital twin technology has brought grain depot management into a new stage. By establishing a three-dimensional virtual model of the grain depot and combining real-time sensor data, a virtual twin that synchronizes with the actual storage environment is formed, which can reflect the internal state of the grain depot in real time. The three-dimensional digital model can not only provide accurate visualization of the spatial layout but also realize real-time monitoring and dynamic management of factors such as grain quality, inventory, and storage environment, helping decision-makers to discover problems in a timely manner and take measures. However, in the current traditional management methods, grain depot information and equipment are usually scattered in different systems, resulting in data islands and poor information circulation. At the same time, changes in the grain depot environment are often difficult to detect in a timely manner, and the warning mechanism is insufficient, leading to low management accuracy and response speed. Summary of the Invention

[0003] Based on this, it is necessary to provide a three-dimensional digital management system and method for grain depots based on digital twins to solve at least one of the above technical problems.

[0004] To achieve the above object, a three-dimensional digital management method for grain depots based on digital twins, the method includes the following steps:

[0005] Step S1: Obtain grain depot structure data and grain depot environment data; perform three-dimensional fine-grained modeling based on the grain depot structure data to obtain a panoramic digital twin scene of the grain depot; adjust the geographical space accuracy of the panoramic digital twin scene of the grain depot according to the grain depot environment data to obtain a bin 3D model;

[0006] Step S2: Use IoT sensors to collect grain depot information and upload the grain depot information to the bin 3D model for data integration and annotation to establish a data visualization interface;

[0007] Step S3: Extract the grain storage information of the data visualization interface to construct a virtual granary; perform 3D particle simulation on the virtual granary, and perform virtual-real contrast warning on the bin 3D model according to the particle simulation results to generate a dynamic grain condition warning signal;

[0008] Step S4: Associate the automated equipment of the grain depot through the dynamic grain condition warning signal and perform adaptive regulation to obtain the optimal management strategy, and execute the three-dimensional digital management operation of the grain depot.

[0009] Through three-dimensional refined modeling, the present invention establishes a digital twin scenario, making the spatial information of the grain depot more intuitive and comprehensive, which helps to improve the understanding of the grain depot layout, structure and environment by grain depot management personnel, so as to carry out daily management more efficiently. Through the real-time collection and data integration of IoT sensors, information such as environmental changes and equipment status of the grain depot can be dynamically presented on the data visualization interface, providing tools for management personnel for real-time monitoring and data analysis, and enhancing the monitoring ability of the grain depot operation. The 3D particle simulation of the virtual granary and the virtual-real contrast warning of the bin 3D model can monitor the changes in the grain condition in the warehouse in real time, including temperature and humidity, ventilation conditions, etc., discover potential problems in time, improve the warning ability, and reduce grain loss or quality problems. By associating with automated equipment, the dynamic grain condition warning signal can trigger an automatic regulation mechanism to achieve adaptive adjustment of the grain depot environment. Specifically, automatically adjust temperature and humidity, air circulation, etc., improve the storage conditions of the granary, and ensure the quality of grain storage. Through adaptive regulation and optimized management strategies, the resource use of the grain depot will be more efficient, reducing unnecessary energy consumption and operating costs, and enhancing the economic benefits of the grain depot. The entire process from data collection, modeling, warning to intelligent regulation constructs an intelligent grain depot management system, promoting the digital transformation of grain depot management, making the management more accurate, automated and efficient. Therefore, the present invention improves the accuracy and response speed of three-dimensional digital management through three-dimensional digital twin modeling, IoT data integration, particle simulation and automated regulation.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: Perform three-dimensional point cloud scanning on the grain depot through lidar to obtain grain depot structure data;

[0012] Step S12: Use meteorological monitoring sensors to obtain grain depot environmental data;

[0013] Step S13: Perform point cloud fusion on the grain depot structure data to generate a three-dimensional structure framework of the grain depot; perform fine-grained hierarchical modeling on the three-dimensional structure framework of the grain depot to generate high-precision grain depot hierarchical information, where the hierarchical modeling includes structure layer modeling, function layer modeling and equipment layer modeling;

[0014] Step S14: Extract the material, light and color of the grain depot structure data, perform texture mapping on the high-precision grain depot hierarchical information, and generate a panoramic digital twin scenario of the grain depot;

[0015] Step S15: Perform geographic coordinate calibration based on the grain depot environmental data to obtain geographic calibration data; use the geographic calibration data to optimize the spatial layout of the panoramic digital twin scene of the grain depot to obtain the 3D model of the granary.

[0016] Through the combination of lidar and meteorological monitoring sensors, the present invention can accurately obtain the three-dimensional structure data and environmental data of the grain depot, providing comprehensive data support for subsequent three-dimensional modeling. Point cloud fusion and fine-grained hierarchical modeling techniques ensure the high precision of the grain depot model. Especially in the modeling of the structure layer, function layer, and equipment layer, it can accurately reflect all aspects of the grain depot. Fine-grained hierarchical modeling not only helps to establish a multi-dimensional three-dimensional structure framework but also provides customized data support for different management needs. The clear division of the structure layer, function layer, and equipment layer makes the management and maintenance work of each layer clearer and easier to operate. By performing texture mapping of features such as material, lighting, and color on the grain depot structure data, the generated panoramic digital twin scene of the grain depot provides a highly realistic visual effect. This high-quality virtual environment can help managers better understand and analyze the overall layout and specific structure of the grain depot. Using the environmental data collected by meteorological monitoring sensors, through geographic coordinate calibration, the geographical spatial accuracy of the panoramic digital twin scene of the grain depot is ensured. This step is crucial for subsequent geographical information analysis and spatial optimization, which can improve the spatial positioning accuracy of the scene and make the location of the grain depot in the actual environment clearer. Through the application of geographic calibration data, the spatial layout of the panoramic digital twin scene of the grain depot is optimized, and the generated 3D model of the granary can more accurately reflect the spatial structure and location of the grain depot, facilitating subsequent intelligent management, decision-making analysis, and automated operations.

[0017] Preferably, step S15 includes the following steps:

[0018] Step S151: Extract the center point, boundary points, and key structure points of the grain depot in the grain depot environmental data as the geographic coordinate reference benchmark data;

[0019] Step S152: Confirm the global coordinates and regional standard coordinates based on the geographic coordinate reference benchmark data, thereby obtaining the global coordinate data and regional standard coordinate data respectively;

[0020] Step S153: Calculate the coordinate offset vector of the global coordinate data and the regional standard coordinate data to obtain the geographic calibration data;

[0021] Step S154: Perform scene space matching on the panoramic digital twin scene of the grain depot through the geographic calibration data to generate the 3D model of the granary.

[0022] By extracting the central points, boundary points, and key structural points from the environmental data of the grain depot and using these points as the reference benchmark for geographical coordinates, the spatial data of the grain depot can be ensured to have high accuracy, which lays the foundation for subsequent geographical calibration and helps with precise spatial positioning at the global and regional levels. By confirming the global coordinates and regional standard coordinates, the coordinate system conversion between different geographical regions can be achieved. This conversion not only helps integrate global and regional data but also provides a reliable basis for the collaborative work of multi-regional data. The acquisition of global coordinate and regional standard coordinate data enhances the cross-regional compatibility and adaptability of the system. By calculating the coordinate offset vector between the global coordinate data and the regional standard coordinate data, geographical calibration data is generated to accurately dock the global coordinates with the regional coordinates. This process helps eliminate the errors between different coordinate systems, improve the spatial accuracy of the digital twin scenario of the grain depot, and ensure its consistency in the actual geographical environment. Through spatial matching of the panoramic digital twin scenario of the grain depot with the geographical calibration data, its high consistency with the real-world geographical space is ensured. This matching process enables the 3D model of the granary to truly reflect the actual structure of the grain depot in terms of spatial positioning, scale, and layout, improving the practicality and accuracy of the model. The data processing of geographical calibration is not limited to a single grain depot but can be extended to the integration of grain depot data in multiple regions and fields. This standardized and precise spatial calibration makes the management of grain depots more collaborative and consistent in large-scale and multi-regional application scenarios, facilitating cross-regional intelligent monitoring and management. Through precise geographical calibration, the digital twin scenario of the grain depot can more accurately reflect the geographical location and structure, helping decision-makers make more scientific and reasonable decisions in aspects such as site selection, equipment layout, and environmental optimization.

[0023] Preferably, step S2 includes the following steps:

[0024] Step S21: Collect grain depot information using IoT sensors;

[0025] Step S22: Calculate the data capacity of the grain depot information and set the data transmission nodes to obtain the data transmission nodes;

[0026] Step S23: Distributively upload the grain depot information to the 3D model of the granary for data integration according to the data transmission nodes to generate integrated grain depot information;

[0027] Step S24: Bind and visualize the module data of the integrated grain depot information and the 3D model of the granary to generate a data visualization interface.

[0028] Through the application of IoT sensors, the present invention can collect key information such as the environmental data, equipment status, and storage conditions of the granary in real time. This real-time data collection provides accurate real-time information for granary management, supporting more efficient dynamic monitoring and management decisions. By calculating the data capacity of the granary information and setting data transmission nodes, the efficiency and stability of data transmission can be ensured. Through reasonable node settings, the data traffic can be effectively shared, the data transmission efficiency can be improved, network bottlenecks or data loss problems can be avoided, and the timeliness and integrity of data can be guaranteed. Through distributed uploading according to the data transmission nodes, the granary information can be effectively distributed and stored in the 3D model of the bin. Such distributed uploading not only improves the data transmission efficiency but also enhances the stability and reliability of the data, ensuring the smooth progress of the data integration process. By generating integrated granary information, the integration and unified management of data from different sources can be achieved, enabling data from various sensors, devices, and systems to be seamlessly docked into the 3D model of the bin, providing unified data support for subsequent comprehensive analysis and management, and promoting data collaboration and information sharing among systems. Binding the integrated granary information with the 3D model of the bin for module data and presenting it through visualization means can present complex granary management information to the management personnel in an intuitive graphical interface. This visual interface facilitates decision-makers to quickly view, analyze, and process data while improving management efficiency and response speed.

[0029] Preferably, step S22 includes the following steps:

[0030] Calculate the data capacity of the granary information;

[0031] Set the data transmission node parameters according to the data capacity: the data transmission rate is 10 kbps to 100 Gbps; the bandwidth requirement is 50 kbps to 100 Gbps; the transmission frequency is 1 time per second to 1 time per hour; the data volume is 10 KB / day to 50 GB / day; the latency requirement ≤ 200 ms; the maximum number of node connections is 1000 nodes / base station;

[0032] Deploy the data transmission nodes for the granary information based on the data transmission node parameters to obtain the data transmission nodes.

[0033] By calculating the data capacity of the grain depot information, the present invention can ensure that the system can anticipate the required bandwidth, storage, and transmission capabilities, thereby reasonably planning the resource requirements for data transmission. This forward-looking data capacity calculation helps to avoid overloading the system and improves the stability and efficiency of data processing. Setting the data transmission node parameters (such as transmission rate, bandwidth requirement, transmission frequency, etc.) according to the data capacity can ensure that data is not affected by network bottlenecks or data loss during transmission. This enables the system to adjust the transmission mode according to different data requirements during actual operation, thus ensuring the efficient transmission and timely update of data. The set data transmission rate (10 kbps to 100 Gbps) and transmission frequency (1 time per second to 1 time per hour) provide flexible configuration options that can be customized according to different grain depot management requirements. The high transmission rate can meet the rapid transmission of large amounts of data, while the low-frequency transmission is suitable for situations where data updates are infrequent, ensuring that the system can be dynamically adjusted according to actual needs. The bandwidth requirement (50 kbps to 100 Gbps) can be flexibly configured according to the specific environment and information flow of the grain depot, ensuring that both small-scale data uploads and large-scale data integrations can be fully supported in terms of network resources, and avoiding data delays or losses caused by insufficient bandwidth. Setting the delay requirement ≤ 200 ms ensures that data can be transmitted to the target node within the specified time, greatly improving the real-time requirement, which is crucial for real-time data monitoring, emergency regulation, and dynamic management in the grain depot environment, and can ensure the timely feedback of information and improve the overall management response efficiency. By setting the maximum number of node connections to 1000 nodes / base station, it can support the wide deployment of grain depot information and meet the needs of multi-node and multi-base station data collection. This flexible expansion ability enables the system to process a large amount of sensor information and adapt to information interaction and integration between large grain depots or multiple grain depots. By deploying data transmission nodes based on the above parameters, the network layout can be optimized to ensure reliable data transmission. The node deployment takes into account various factors such as transmission rate, bandwidth requirement, and delay requirement, and can maximize the efficiency and stability of data transmission, ensuring smooth data transmission in the grain depot management system.

[0034] Preferably, the construction of the virtual granary by extracting the grain depot storage information of the data visualization interface in step S3 includes:

[0035] Extract the grain depot storage information of the data visualization interface;

[0036] Perform time series analysis on the grain depot storage information to generate grain depot storage time series data;

[0037] Based on the grain depot storage time series data, perform a storage structure mirror copy on the bin 3D model to generate a virtual storage framework;

[0038] Simulate node deployment for the virtual storage framework to generate a virtual granary.

[0039] By extracting the granary storage information from the data visualization interface, the present invention can accurately capture various storage data of the granary (such as storage quantity, storage status, equipment usage, etc.), which provides a complete and accurate data basis for subsequent virtual granary construction, data analysis, and management decision-making. Conducting time-series analysis on the granary storage information can reveal the dynamic changes in granary storage at different time periods. This time-series data analysis helps managers deeply understand the changing trends, periodic fluctuations, and existing problems of storage, thus providing strong support for optimizing storage management, adjusting storage plans, and implementing early warning management. Mirror-copying the storage structure of the granary 3D model based on the granary storage time-series data can efficiently construct a virtual storage framework that is consistent with the actual storage structure. Through this process, the virtual warehouse can accurately simulate the spatial layout and storage methods of the real warehouse, providing a powerful virtual platform for subsequent storage management, resource allocation, and space optimization. By simulating node deployment for the virtual storage framework, more detailed control can be provided for the operation and management of the virtual granary. This node deployment can be flexibly adjusted according to the requirements of the actual warehouse, optimizing the allocation of storage resources, adjusting the storage structure, and enhancing the response speed and processing capacity of the system. Conducting node deployment on the basis of the virtual granary enables the virtual granary to have flexible expansion capabilities. When the scale of the granary or storage requirements change, the virtual granary can quickly adjust the node configuration to easily handle the management requirements of large-scale or multi-scenario granaries.

[0040] Preferably, the 3D particle simulation of the virtual granary in step S3 includes:

[0041] Extract the temperature and humidity, in-warehouse location layout, and grain particle data inside the virtual granary;

[0042] Add initial simulation attributes to the grain particle data to generate particle physical attributes, where the attribute addition includes particle size distribution, density, and elasticity;

[0043] Calculate the air flow direction and velocity of the in-warehouse location layout according to the temperature and humidity, and simulate the flow path of the grain particles in combination with the grain particle data to generate grain flow simulation data;

[0044] Simulate the interaction and collision between grain particles in the grain flow simulation data based on the particle physical attributes to obtain grain movement simulation data;

[0045] Simulate the accumulation and movement of particles in the warehouse for the grain movement simulation data through the in-warehouse location layout to obtain grain accumulation simulation data;

[0046] Integrate the grain flow simulation data, grain movement simulation data, and grain accumulation simulation data into particle simulation results.

[0047] In the present invention, by adding initial attributes for simulation to the grain particle data (such as particle size distribution, density, and elasticity), the behavior of grain particles in the warehouse can be accurately simulated. The setting of these particle physical properties makes the simulation results more realistic and can truly reflect the movement, accumulation, and interaction of grains in the warehouse environment. By calculating the temperature, humidity, and air flow direction and velocity in the warehouse and combining with the grain particle data, the flow path of grains under different climate conditions can be simulated, which provides data support for air flow management, temperature and humidity control, and environmental optimization in the warehouse, and helps to ensure the storage quality and safety of grains. The grain flow simulation data provides the actual flow path of grain particles in the warehouse. This process not only helps to understand the movement mode of grains under different environmental conditions but also provides an important basis for warehouse management to optimize the structural layout of grain storage and reduce problems such as uneven accumulation. By simulating the interaction and collision between grain particles, the force conditions of grains under different conditions can be better evaluated, and how grains deform, compact, or move during the accumulation process can be understood, which is very important for ensuring the safe storage of grains and avoiding damage caused by over-dense accumulation. By simulating the grain accumulation simulation data, the accumulation pattern of grains in the warehouse can be predicted in advance, thereby optimizing the storage location layout of the warehouse. This not only improves the utilization efficiency of the warehouse space but also reduces the quality loss of grains caused by uneven accumulation or improper location.

[0048] Preferably, the virtual-real contrast warning of the bin 3D model according to the particle simulation results in step S3 includes:

[0049] Identify abnormal grain conditions in the bin 3D model according to the particle simulation results. When any of the following situations occurs, determine that the particle simulation result is abnormal grain flow and obtain abnormal grain flow data: the grain flow velocity deviates from the optimal range by more than ±10%; the grain flow deviates from the predetermined direction by more than 15°;

[0050] When the following situations occur simultaneously, determine that the particle simulation result is abnormal grain movement and obtain abnormal grain movement data: the grain acceleration detection is lower than 0.5 m / s² for 3 consecutive times; the lag time of the grain movement trajectory exceeds 20 min; the fluctuation of the grain movement speed is greater than ±15% and lasts for more than 30 min;

[0051] When the following situations occur simultaneously, determine that the particle simulation result is abnormal grain accumulation and obtain abnormal grain accumulation data: the grain accumulation density changes by more than 10% within 1 hour; the deviation of the grain accumulation height from the preset accumulation mode in the warehouse exceeds 20%; the temperature in the grain accumulation area deviates from the normal range by more than 5°C;

[0052] Integrate abnormal grain flow data, abnormal grain movement data, and abnormal grain accumulation data to obtain abnormal grain condition data;

[0053] Generate corresponding dynamic grain condition warning signals according to the types of abnormal grain condition data.

[0054] Through parsing the device control parameters of the dynamic grain condition warning signal, the present invention can generate an accurate device control instruction dataset. This process ensures that according to the changes in the grain storage environment, the device can respond in a timely manner and execute corresponding operations, improving the automation level of the device and reducing the need for human intervention. By verifying the instruction matching degree in combination with the grain condition monitoring data collected in real time by IoT sensors, the control instructions of the device can be dynamically adjusted, ensuring that the device operation is more flexible and accurate. This real-time feedback mechanism enables the warehouse equipment to be optimized in a timely manner according to the real-time conditions in the grain depot, ensuring the continuous suitability of the grain storage conditions. The generation and execution of the device state optimization instructions can optimize the device operation state, reduce energy consumption and improve work efficiency. By verifying the instruction matching degree, the system can further improve the response accuracy of the device, ensuring that each operation can be effectively executed and avoiding losses caused by incorrect or unmatched control instructions. Through pulse collaborative adaptive regulation, the warehouse management strategy can be automatically adjusted according to the real-time grain condition data. The automatic control and regulation can significantly improve the overall efficiency of grain depot management, reduce the error rate of manual operations, and improve the stability of the grain storage environment. By generating the optimal management strategy and executing it in the 3D model of the granary, a more scientific and efficient three-dimensional digital management operation of the grain depot can be realized. This data-driven decision-making method helps the warehouse manager make quick decisions on the optimal plan, effectively avoiding resource waste and management mistakes.

[0055] Preferably, step S4 includes the following steps:

[0056] Step S41: Parse the device control parameters of the dynamic grain condition warning signal to generate a device control instruction dataset;

[0057] Step S42: Based on the automated device control instruction dataset, verify the instruction matching degree in combination with the grain condition monitoring data collected in real time by IoT sensors to generate device state optimization instruction parameters;

[0058] Step S43: Input the device state optimization instruction parameters into the 3D model of the granary for pulse collaborative adaptive regulation to generate the optimal management strategy for executing the three-dimensional digital management operation of the grain depot.

[0059] Through the analysis of equipment control parameters for dynamic grain condition warning signals, the present invention can generate an accurate dataset of equipment control instructions. This process ensures that the equipment can respond in a timely manner and execute corresponding operations according to changes in the grain storage environment, improving the automation level of the equipment and reducing the need for human intervention. By combining the grain condition monitoring data collected in real time by IoT sensors for instruction matching degree verification, the control instructions of the equipment can be dynamically adjusted, ensuring that the equipment operation is more flexible and precise. This real-time feedback mechanism enables the warehouse equipment to be optimized in a timely manner according to the real-time conditions in the granary, ensuring the continuous suitability of the grain storage conditions. The generation and execution of equipment state optimization instructions can optimize the equipment operation state, reduce energy consumption and improve work efficiency. By verifying the instruction matching degree, the system can further improve the response accuracy of the equipment, ensuring that each operation can be effectively executed and avoiding losses caused by incorrect or mismatched control instructions. Through pulse collaborative adaptive regulation, the warehouse management strategy can be automatically adjusted according to real-time grain condition data. The automatic control and regulation can significantly improve the overall efficiency of granary management, reduce the error rate of manual operations, and improve the stability of the grain storage environment. By generating the optimal management strategy and executing it in the 3D model of the granary, a more scientific and efficient three-dimensional digital management operation of the granary can be realized. This data-driven decision-making method helps warehouse managers make quick decisions on the optimal plan, effectively avoiding resource waste and management mistakes. Through automatic equipment regulation and real-time grain condition warning, potential risk problems can be detected and solved in a timely manner during the grain storage process. By taking adaptive management measures in advance, the impact of environmental changes on grain quality can be reduced, and the safety and storage period of the grain can be improved.

[0060] In this specification, a three-dimensional digital management system for a granary based on digital twin is provided for implementing the above-mentioned three-dimensional digital management method for a granary based on digital twin. The three-dimensional digital management system for a granary based on digital twin includes:

[0061] A scene modeling module, configured to obtain granary structure data and granary environment data; perform three-dimensional fine modeling based on the granary structure data to obtain a panoramic digital twin scene of the granary; and adjust the geospatial accuracy of the panoramic digital twin scene of the granary according to the granary environment data to obtain a 3D model of the granary.

[0062] A data uploading module, configured to collect granary information by using IoT sensors and upload the granary information to the 3D model of the granary for data integration and annotation to establish a data visualization interface.

[0063] A digital management module, configured to extract the granary storage information of the data visualization interface to construct a virtual granary; perform a 3D particle simulation on the virtual granary, and perform a virtual-real contrast warning on the 3D model of the granary according to the particle simulation results to generate a dynamic grain condition warning signal.

[0064] The feedback control module is used to associate the grain depot's automated equipment with dynamic grain condition warning signals and perform adaptive regulation to obtain the optimal management strategy and execute three-dimensional digital management operations of the grain depot.

[0065] The beneficial effect of the present invention is that a highly accurate panoramic digital twin scene of the grain depot can be created by three-dimensional fine modeling of the grain depot structure data, so that the virtual representation of the grain depot is highly consistent with the actual situation, which helps to more realistically reflect the spatial layout and status of the grain depot. The geospatial accuracy adjustment of the panoramic digital twin scene based on the grain depot environmental data can ensure the matching of the model with the actual geographical environment, thereby enhancing the accuracy of warehouse space planning and avoiding potential problems caused by coordinate errors. By generating a 3D model of the granary, managers can intuitively view and manage the grain depot in the digital space, optimize space use and improve decision-making efficiency. IoT sensors are used to collect real-time information and upload these data to the 3D model of the granary to ensure that the management information of the grain depot is updated in real time, which improves the response speed and flexibility of grain depot management. Through data integration and annotation, a comprehensive data visualization interface is formed, allowing grain depot managers to clearly and intuitively view various storage information, thereby supporting better decision-making. By integrating multi-source data, the information island phenomenon can be effectively reduced, cross-system collaborative work can be achieved, and the availability and analysis depth of the overall data can be improved. By extracting storage information from the data visualization interface, building a virtual granary, and simulating different storage conditions, this not only enhances the transparency of grain depot management, but also provides an accurate basis for subsequent optimization decisions. Through 3D particle simulation of the virtual granary, the flow, accumulation and interaction of grain can be simulated, potential problems can be discovered in advance, and early warning signals can be provided for warehouse management. This simulation warning helps to adjust storage management in time to avoid adverse situations such as accumulation or abnormal flow. Through the virtual-real comparison warning of particle simulation results and the 3D model of the granary, the abnormal situation of grain in the warehouse can be detected in real time, and dynamic grain situation warning signals can be provided to decision makers, further improving the accuracy and timeliness of warehouse management. Through dynamic grain situation warning signals, the automation equipment is associated and adaptively controlled to achieve refined control of equipment in grain depot management, which can optimize equipment operation according to actual conditions, reduce human intervention, and improve the overall efficiency of the system. Generate the optimal management strategy based on the warning signal to ensure that all tasks in the warehouse management process, such as temperature and humidity control, material flow, etc., can be carried out in the best state, and improve the storage environment and management efficiency of grain. Finally, by executing the three-dimensional digital management operation of the grain depot, intelligent and accurate management operations can be achieved, the operational efficiency of the grain depot can be improved, and the possibility of errors can be reduced. Therefore, the present invention improves the accuracy and response speed of three-dimensional digital management through three-dimensional digital twin modeling, IoT data integration, particle simulation and automatic control. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a schematic diagram of the step - by - step process of a three - dimensional digital management method for grain depots based on digital twins;

[0067] Figure 2 is Figure 1 a detailed schematic diagram of the implementation steps of step S1 in

[0068] Figure 3 is Figure 1 a detailed schematic diagram of the implementation steps of step S4 in

[0069] The realization, functional features, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0070] The following clearly and completely describes the technical method of the present invention with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0071] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. Functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0072] It should be understood that although terms such as "first", "second", etc. may be used here to describe each unit, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit can be called the second unit, and similarly the second unit can be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed associated items.

[0073] To achieve the above object, please refer to Figures 1 to 3 , a three - dimensional digital management method for grain depots based on digital twins, the method includes the following steps:

[0074] Step S1: Obtain the grain depot structure data and the grain depot environment data; perform three-dimensional refined modeling based on the grain depot structure data to obtain a panoramic digital twin scene of the grain depot; adjust the geospatial accuracy of the panoramic digital twin scene of the grain depot according to the grain depot environment data to obtain a bin 3D model;

[0075] Step S2: Use IoT sensors to collect grain depot information and upload the grain depot information to the bin 3D model for data integration and annotation to establish a data visualization interface;

[0076] Step S3: Extract the grain storage information of the data visualization interface to construct a virtual granary; perform 3D particle simulation on the virtual granary, and conduct virtual-real comparison warning on the bin 3D model according to the particle simulation results to generate a dynamic grain condition warning signal;

[0077] Step S4: Associate the dynamic grain condition warning signal with the automated equipment of the grain depot and perform adaptive regulation to obtain an optimal management strategy and execute the three-dimensional digital management operation of the grain depot.

[0078] Through three-dimensional refined modeling, the present invention establishes a digital twin scene, making the spatial information of the grain depot more intuitive and comprehensive, which helps to improve the understanding of the grain depot layout, structure and environment by grain depot management personnel, so as to conduct daily management more efficiently. Through the real-time collection and data integration of IoT sensors, information such as environmental changes and equipment status of the grain depot can be dynamically presented on the data visualization interface, providing a tool for management personnel for real-time monitoring and data analysis, and enhancing the monitoring ability of the operation of the grain depot. The 3D particle simulation of the virtual granary and the virtual-real comparison warning of the bin 3D model can monitor the changes in the grain condition in the warehouse in real time, including temperature and humidity, ventilation conditions, etc., timely discover potential problems, improve the warning ability, and reduce grain loss or quality problems. Through the association with automated equipment, the dynamic grain condition warning signal can trigger an automatic regulation mechanism to achieve adaptive adjustment of the grain depot environment. Specifically, automatically adjust the temperature and humidity, air circulation, etc. to improve the storage conditions of the granary and ensure the storage quality of the grain. Through adaptive regulation and optimized management strategies, the resource use of the grain depot will be more efficient, reducing unnecessary energy consumption and operating costs, and enhancing the economic benefits of the grain depot. The entire process from data collection, modeling, warning to intelligent regulation constructs an intelligent grain depot management system, promoting the digital transformation of grain depot management and making the management more accurate, automated and efficient. Therefore, the present invention improves the accuracy and response speed of three-dimensional digital management through three-dimensional digital twin modeling, IoT data integration, particle simulation and automated regulation.

[0079] In the embodiment of the present invention, refer to Figure 1As shown in the figure, it is a schematic diagram of the step flow of a three-dimensional digital management method for a grain depot based on digital twin. In this example, the three-dimensional digital management method for a grain depot based on digital twin includes the following steps:

[0080] Step S1: Obtain the grain depot structure data and grain depot environment data; perform three-dimensional refined modeling based on the grain depot structure data to obtain a panoramic digital twin scene of the grain depot; adjust the geographical spatial accuracy of the panoramic digital twin scene of the grain depot according to the grain depot environment data to obtain a bin 3D model;

[0081] In the embodiment of the present invention, by obtaining the grain depot structure data, including the length of the grain depot is 80m, the width is 40m, the height is 15m, the number of bins is 10, the size of each bin is 20m×10m×12m, the wall thickness is 0.3m, the spacing of the support columns is 5m, and it includes a top ventilation system (the spacing between the air vents is 2m, and the diameter of each air vent is 0.5m). At the same time, collect the grain depot environment data, including geographical coordinates (latitude 34.265N, longitude 108.953E), altitude 450m, average annual temperature 15°C, average annual humidity 60%, normal wind speed 3.5m / s, extreme wind speed 20m / s, and ground slope 3° (north-south direction). Then, based on the grain depot structure data, use BIM (Building Information Modeling) + three-dimensional GIS technology for refined modeling, use Autodesk Revit or Rhino+Grasshopper to establish three-dimensional objects such as walls, support structures, and roof air vents, and set a grid accuracy of 0.05m, the thermal conductivity of the wall material is 1.4W / (m·K), the smoothness is 0.8, and the diffuse reflectivity is 0.3. At the same time, arrange 5×5 temperature and humidity monitoring points inside each bin. After the modeling is completed, import the model into Unity3D or Unreal Engine to generate a VR interactive visual panoramic digital twin scene. Subsequently, perform geographical space calibration according to the grain depot environment data, accurately locate the bin 3D model according to the GPS coordinates (34.265N, 108.953E), and use Google Earth DEM (Digital Elevation Model) to adjust the bin foundation to an altitude of 450m. At the same time, correct the bottom of the bin based on IDW interpolation to match the real terrain. In addition, optimize and adjust according to environmental factors, including optimizing the roof air vent design for a wind speed of 3.5m / s, setting the air outlet temperature of the bin air conditioner to 18°C, the air volume to 500m³ / h, and adjusting the humidity control system based on a relative humidity of 60%. Finally, generate a bin 3D model with an accuracy of ±5cm, and integrate the grain depot environment data to support structural analysis, temperature and humidity control optimization, and intelligent management.

[0082] Step S2: Use loT sensors to collect grain depot information and upload the grain depot information to the bin 3D model for data integration and annotation to establish a data visualization interface;

[0083] In the embodiments of the present invention, multiple types of IoT sensors are deployed inside and outside the grain depot, including temperature and humidity sensors, CO2 sensors, gas detection sensors, light sensors, wind speed and direction sensors, grain stacking temperature sensors, etc. The specific layout is as follows: 25 groups of temperature and humidity sensors (5×5 grid layout) are set in each granary, covering the surface and interior of the grain pile, and 1 group is arranged every 1m in depth to ensure that the accuracy of temperature and humidity monitoring reaches ±0.5°C and ±2%RH. A CO2 sensor (monitoring range 400 - 5000ppm, error ±50ppm) and an oxygen concentration sensor are installed on the top of each granary to determine whether abnormal respiration occurs in the grain pile. Hydrogen sulfide (H2S) and ammonia (NH3) sensors are installed around the granary to detect whether harmful gases are generated due to mildew of the grain. A wind speed sensor (measurement range 0 - 20m / s, error ±0.1m / s) and a wind direction sensor are installed at the ventilation opening of the granary to optimize the ventilation strategy. A light sensor (measurement range 0 - 100,000lux) is installed in the window area of the grain depot to evaluate the impact of external light on the storage environment. Secondly, all sensors transmit data through LoRa wireless network or NB-IoT (Narrow Band Internet of Things), collect data every 10 seconds, upload it to the grain depot data center every 1 minute, and store it in a cloud database (such as MySQL, InfluxDB), and at the same time transmit it to the 3D model of the granary. In the 3D model of the granary, using data integration and annotation technology, the real-time data is mapped to the corresponding positions of the 3D model. Specifically, the temperature and humidity data of each granary are displayed in the form of a heatmap, with the areas of high temperature marked in red and the normal areas marked in green; gas data such as CO2 and H2S are presented through alarm signs on the 3D interface, such as yellow or red warning lights are displayed in the exceeded standard areas. At the same time, using BIM+GIS data fusion, the real-time status of the sensors is displayed on the 3D interface. Specifically, the numerical dynamic change of the wind speed sensor can be intuitively displayed through an arrow flow animation. Finally, a data visualization interface is constructed, adopting a Web-side and VR / AR interactive interface, and 3D visualization is realized through front-end technologies (such as WebGL, Three.js). Grain depot managers can enter the system through the PC side or the mobile side, view various environmental parameters of the 3D model of the granary in real time, and combine with the AI early warning system to identify potential risks. Specifically, an alarm is automatically triggered when the abnormal rise of the grain stacking temperature or the CO2 exceeds the standard, guiding the ventilation and regulation strategies of the grain depot.

[0084] Step S3: Extract the grain storage information of the data visualization interface to construct a virtual granary; perform 3D particle simulation on the virtual granary, and compare the virtual and real situations of the 3D model of the granary according to the particle simulation results to generate a dynamic grain situation early warning signal;

[0085] In the embodiments of the present invention, key grain depot storage information is extracted from the data visualization interface, including the temperature and humidity data of the granary, gas composition data, temperature gradient of the grain accumulation layer, ventilation parameters, structural characteristics of the granary, etc. Assume: the height of the grain pile inside the granary is 10m, the initial filling density is 750kg / m³, the moisture content is 13%, and the ventilation rate is 2m³ / h. The sensor measures that the local grain pile temperature rises to 35°C and the CO2 concentration rises to 900ppm, indicating a risk of pest infestation or mildew. Based on the above data, a virtual granary is constructed using three-dimensional grid modeling technology (BIM+GIS), whose structure is the same as that of the actual granary. At the same time, voxel storage units are established on the surface and inside of the grain pile, and the grain pile is divided into basic units of 10cm×10cm×10cm for particle simulation calculation. In the virtual granary, discrete particle dynamics (DPM) or computational fluid dynamics (CFD) is used for 3D particle simulation to simulate dynamic processes such as air flow, temperature diffusion, and humidity change in the grain pile. The lattice Boltzmann method (LBM) is used to establish a temperature and humidity diffusion model to simulate the heat transfer in the grain pile and predict potential high-temperature areas. Fick's Law is used to calculate the changes in CO2 and O2 to judge whether the grain respiration is abnormal. Based on the multi-agent system (MAS), the diffusion behavior of pests inside the grain pile is simulated, and the pest reproduction rate is calculated in combination with the temperature and humidity field. During the simulation process, different initial conditions (temperature, humidity, gas concentration) are set, and 1000-5000 iterative calculations are performed to generate dynamic particle field data, and the temperature and humidity isosurfaces, gas concentration distribution maps, and pest propagation trajectories are output. The 3D particle simulation results are compared with the actual 3D model of the granary to analyze the differences between the virtual granary simulation data and the real-time monitoring data: calculate the mean square error (MSE) between the virtual granary temperature field and the actual sensor temperature field. When the error exceeds the set threshold (such as 2°C), there is a risk of abnormal temperature rise. Analyze the trend of CO2 concentration change. If the growth rate of the actual data exceeds 20% of the simulation prediction, there is excessive grain respiration or pest infestation. Combine the predicted pest propagation path from the simulation with the video monitoring data or the pest location identified by AI inside the granary to judge the degree of pest diffusion. According to the results of the virtual-real comparison, a dynamic grain condition warning signal is generated using machine learning (such as the LSTM prediction model): Yellow warning (low risk): The temperature rises by 1-2°C, and the CO2 increases by 100-300ppm. The system recommends optimizing the ventilation strategy. Orange warning (medium risk): The temperature rises by 3-5°C, the CO2 exceeds 800ppm, and the humidity increases by more than 5%, indicating local mildew.Red alert (high risk): When the temperature exceeds 35°C, CO2 exceeds 1000 ppm, and pest activities are detected, the system automatically triggers the fan to accelerate ventilation, and sends text messages / emails to alarm the granary management staff. Finally, all warning signals are marked and displayed in the 3D model of the granary warehouse with dynamic colors. Specifically, the specific high-temperature area turns into a red gradient, and the pest risk area is marked in black, forming an intuitive dynamic grain condition monitoring system.

[0086] Step S4: Associate the automated equipment in the granary through the dynamic grain condition warning signal and perform adaptive regulation to obtain the optimal management strategy, and execute the three-dimensional digital management operation of the granary.

[0087] In the embodiment of the present invention, the automated equipment in the granary is associated through the dynamic grain condition warning signal and adaptive regulation is performed to form the optimal management strategy and execute the three-dimensional digital management operation of the granary. First, the system analyzes warning information such as temperature, humidity, gas concentration, and pest detection, and automatically matches automated equipment such as ventilation systems, gas replacement systems, intelligent dehumidification equipment, and automatic fumigation systems. Specifically, when the temperature of the grain pile exceeds 35°C and the CO2 concentration is higher than 900 ppm, the system triggers the ventilation and cooling mode and starts the fans of levels 1-3; if the local humidity is higher than 18%, the dehumidification equipment is turned on and the ventilation rate is reduced to prevent the diffusion of moisture. After the equipment is associated, the system uses the adaptive regulation algorithm to perform intelligent parameter adjustment. Specifically, the rotation speed of the fan is dynamically adjusted through the PID control algorithm to ensure that the temperature, humidity, and CO2 concentration of the grain pile are maintained within a safe range. In addition, in terms of pest control, the system automatically selects different strategies such as ultraviolet sterilization, low-oxygen regulation, and phosphine fumigation according to the pest level, and continuously optimizes the regulation plan through Reinforcement Learning (RL). The reinforcement learning model takes the temperature, humidity, gas concentration, and pest distribution in the granary as the state, takes adjusting the fan, starting fumigation, and adjusting the gas concentration as the actions, and continuously adjusts the management strategy through the reward function to ensure the stability of the grain condition. Finally, after all the optimized strategies are executed, the system synchronizes the operation records to the 3D digital twin system of the granary warehouse, and presents the temperature and humidity dynamics, equipment status, and historical operation records in the form of a heat map, color coding, and equipment operation logs on the three-dimensional visualization management interface. At the same time, it supports remote control and intelligent inspection, realizing the digital, automated, and intelligent management of the granary.

[0088] As an example of the present invention, refer to Figure 2 As shown, in this example, step S1 includes:

[0089] Step S11: Perform three-dimensional point cloud scanning on the granary through lidar to obtain the granary structure data;

[0090] Step S12: Use meteorological monitoring sensors to obtain the granary environment data;

[0091] Step S13: Perform point cloud fusion on the grain depot structure data to generate a 3D structure framework of the grain depot; perform fine-grained hierarchical modeling on the 3D structure framework of the grain depot to generate high-precision grain depot hierarchical information, where the hierarchical modeling includes structural layer modeling, functional layer modeling, and equipment layer modeling;

[0092] Step S14: Extract the material, lighting, and color of the grain depot structure data, perform texture mapping on the high-precision grain depot hierarchical information to generate a panoramic digital twin scene of the grain depot;

[0093] Step S15: Perform geographic coordinate calibration based on the grain depot environment data to obtain geocalibration data; use the geocalibration data to optimize the space of the panoramic digital twin scene of the grain depot to obtain the 3D model of the granary.

[0094] In the embodiments of the present invention, by using a Light Detection and Ranging (LiDAR) to scan the entire structure of the grain depot, accurate three-dimensional point cloud data is obtained. Ensure that the LiDAR system can cover all corners of the grain depot, especially high places and complex structures, in order to obtain comprehensive structural information. The data collection should be divided into multiple perspectives to ensure high-precision point cloud data without blind spots. The scanning results are preliminarily processed to remove noise and redundant data to ensure the accuracy of the point cloud data. Meteorological monitoring sensors (such as temperature, humidity, wind speed, air pressure, etc. sensors) are deployed in different areas of the grain depot to obtain environmental data. The sensors are configured to collect and record environmental parameters in real time, and the data should be updated at regular time intervals. Through the point cloud registration algorithm, the LiDAR scanning data from different angles or positions is fused to generate a complete three-dimensional point cloud data set. Using the point cloud data to generate a three-dimensional framework of the grain depot, it can be processed using modeling software (such as Revit, AutoCAD, etc.). According to the actual functions and structures of the grain depot, fine-grained hierarchical modeling is carried out, including: constructing a structural layer based on the main framework, support structures, etc. of the building, constructing a functional layer according to different functional areas of the grain depot (such as storage areas, equipment areas, channels, etc.), and the equipment layer modeling includes the equipment in the grain depot, such as conveyor belts, cooling equipment, storage equipment, etc. Extract the material properties related to the grain depot structure (such as the materials of the walls, floors, and ceilings) from the point cloud data and calibrate them. Use the light data collected during the LiDAR scanning process, or extract color information from the image data, to assign colors to different areas. Apply the extracted material, color, and light information to the three-dimensional model to generate a realistic texture mapping. Combining multi-perspective and high-precision point cloud data, a panoramic digital twin scene of the grain depot is generated to make it have a sense of reality and interactivity. Align the meteorological monitoring sensor and LiDAR data with the actual geographical coordinates. Use GPS and other geolocation technologies for high-precision calibration. Based on the obtained geographical calibration data, spatial optimization of the panoramic digital twin scene of the grain depot is carried out to ensure that the virtual scene is consistent with the actual geographical location. Use spatial data processing tools (such as GIS software) to perform optimization operations on the model, such as coordinate transformation, spatial alignment, and scale correction. Through these optimizations, an accurate three-dimensional model of the grain depot, that is, the bin 3D model, can be obtained, which can accurately present the structure and functions of the grain depot in the virtual environment.

[0095] Preferably, step S15 includes the following steps:

[0096] Step S151: Extract the central point, boundary points, and key structural points of the grain depot environment data as the geographical coordinate reference benchmark data;

[0097] Step S152: Based on the geographical coordinate reference benchmark data, confirm the global coordinates and regional standard coordinates, thereby obtaining the global coordinate data and regional standard coordinate data respectively;

[0098] Step S153: Calculate the coordinate offset vector between the global coordinate data and the regional standard coordinate data to obtain the geographic calibration data;

[0099] Step S154: Perform scene space matching on the panoramic digital twin scene of the grain depot through the geographic calibration data to generate the 3D model of the granary.

[0100] In the embodiments of the present invention, the geometric center point of the grain depot is identified based on environmental data, such as lidar scans or GPS information. Generally, the center position of the grain depot can be obtained through the centroid calculation method of point cloud data. By lidar scanning or other measurement means, the external boundary points of the grain depot are extracted, and these points usually represent the outermost ends of the grain depot outline. Determine the key structural points of the grain depot, such as entrances, exits, main support columns, partition walls and other important building elements, which play a key role in subsequent coordinate calibration. Store the coordinates of the extracted center points, boundary points and key structural points in a database or a geographic information system (GIS) to ensure that they can be used as reference benchmark data for subsequent calibration processes. Use the Global Positioning System (GPS) or other global coordinate reference systems (such as WGS84) to convert the data to the global coordinate system through known reference points (such as the center point or boundary point of the grain depot) to ensure that it can be docked with the global map system. Based on the regional standard coordinate system (such as the Chinese National Geodetic Coordinate System CGCS2000, etc.), match the coordinates of the grain depot environmental data with the regional standard coordinate system. This process includes coordinate transformation, application of projection methods, etc., to ensure that the data is consistent with the regional standard. According to the existing coordinate reference data, convert the original point cloud data or measurement data into global coordinates and regional standard coordinates through conversion tools (such as coordinate conversion algorithms or dedicated software). Calculate the difference between the global coordinate data and the regional standard coordinate data in the grain depot environmental data. This usually involves the difference between the two coordinate systems, which is called the coordinate offset (or coordinate error) vector. Based on reference points (such as the center point, boundary point, etc. of the grain depot), calculate the coordinate offset of each point. By calculating the offset vector, the parameters required for coordinate transformation, such as translation amount, rotation angle, etc., can be obtained. Use the offset vector to quantify the difference between the global coordinate and the regional standard coordinate system, so as to obtain a set of geographic calibration data, which can include specific translation, rotation and scaling adjustment parameters. According to the geographic calibration data obtained in step S153, perform spatial matching on the point cloud data in the panoramic digital twin scene of the grain depot through a coordinate transformation algorithm (such as affine transformation, translation-rotation transformation, etc.). This step is to align the virtual grain depot model with the actual geographic coordinate system. After completing the spatial matching, use 3D modeling software (such as Blender, Revit, etc.) to perform modeling processing on the calibrated digital twin scene. The calibrated scene will have higher geographic accuracy and be able to be consistent with the coordinate system of the real world. Through the spatial matching process, ensure the alignment of the 3D model of the grain depot with the actual geographical environment, and optimize it to improve the accuracy and visualization effect of the model. The finally generated 3D model of the granary can be exported in a format that can be used for applications such as virtual reality (VR) and augmented reality (AR), or used for subsequent warehouse management and optimization analysis.

[0101] Preferably, step S2 includes the following steps:

[0102] Step S21: Collect grain depot information using IoT sensors;

[0103] Step S22: Calculate the data capacity of the grain depot information and set data transmission nodes to obtain data transmission nodes;

[0104] Step S23: Distributively upload the grain depot information to the 3D model of the granary for data integration according to the data transmission nodes to generate integrated grain depot information;

[0105] Step S24: Bind and visualize the module data of the integrated grain depot information and the 3D model of the granary to generate a data visualization interface.

[0106] In the embodiments of the present invention, different types of IoT sensors (such as temperature and humidity sensors, gas sensors, pressure sensors, light sensors, etc.) are arranged in the grain depot to obtain data related to the grain depot environment. These sensors can be installed at various key positions in the grain depot, such as the storage area, equipment area, passageway, etc. Enable IoT devices to perform real-time data collection, and the data includes environmental temperature, humidity, air quality, gas concentration, etc. The sensors need to have the ability to automatically collect, store, and upload data. The data collected by the sensors is transmitted to the central data processing system or cloud platform through wireless transmission methods (such as Wi-Fi, Zigbee, LoRa, etc.), providing a basis for subsequent data processing and analysis. Calculate the overall data capacity according to the number, collection frequency, data type, and required storage space of various sensors in the grain depot. The data capacity calculation should consider the data flow generated by the sensors, storage requirements, and network bandwidth requirements. Design a reasonable node transmission scheme according to the characteristics of the data flow in the grain depot. These nodes can include local data collection points (specific sensor clusters), data relay nodes, and forwarding nodes for uploading to the central server or cloud platform. In a large grain depot, using relay nodes can reduce the transmission distance and optimize the data transmission efficiency. According to the settings of the transmission nodes, plan the data flow between the nodes to ensure the stability and efficiency of data transmission. Configure the transmission frequency, transmission rate, data compression strategy, etc. for each node to ensure that each node can efficiently process and transmit grain depot information. Distribute and upload the grain depot data collected from each sensor through the configured transmission nodes. According to the distribution of the transmission nodes, batch upload the data by region or module to avoid network pressure during centralized data upload. Integrate the uploaded grain depot information data into the 3D model of the granary. The data integration process involves associating the sensor data with the corresponding parts in the 3D model to ensure that each data point accurately matches its corresponding physical location and environmental characteristics. Specifically, the temperature data will correspond to the storage area of the grain depot, and the humidity data will correspond to the passageway area in the grain depot. Databases (such as SQL, NoSQL) can be used to store and manage sensor data, and at the same time map the data to the corresponding spatial coordinate positions in the 3D model. Implement a real-time update mechanism for the data to ensure that the grain depot information can be continuously updated to the 3D model of the granary as the sensors collect and upload data, providing the latest grain depot status data. On the basis of the integrated grain depot information, bind different sensor data (such as temperature and humidity, gas concentration, etc.) to the modules in the 3D model of the granary. Specifically, the temperature and humidity information in the storage area can be corresponding to the grain storage area in the model. Use a unified interface standard (such as RESTful API or WebSocket) for data transmission and binding to ensure the consistency and real-time nature of the information. Integrate the bound integrated data with the 3D model of the granary to generate a data visualization interface.The interface can display each module of the grain depot and the corresponding environmental data through 3D rendering software (such as Unity3D, Blender, Three.js, etc.). The visualization interface should allow users to view real-time data (such as temperature and humidity, gas concentration, pressure, etc.) in different areas through interactive operations and support dynamic updates. It supports the switching of different perspectives, specifically the first-person perspective, panoramic perspective, bird's-eye view perspective, etc., providing a diverse interactive experience. Through the visualization interface, users can monitor in real time information such as environmental changes, equipment status, and resource usage inside the grain depot and perform remote operations and management. The visualization interface should have an alarm function. When certain environmental parameters exceed the preset threshold, the system can trigger an alarm and notify the management personnel to ensure the safe and stable operation of the grain depot.

[0107] Preferably, step S22 includes the following steps:

[0108] Calculate the data capacity of the grain depot information;

[0109] Set the data transmission node parameters according to the data capacity: the data transmission rate is from 10 kbps to 100 Gbps; the bandwidth requirement is from 50 kbps to 100 Gbps; the transmission frequency is from 1 time per second to 1 time per hour; the data volume is from 10 KB / day to 50 GB / day; the latency requirement is ≤200 ms; the maximum number of node connections is 1000 nodes / base station;

[0110] Deploy the data transmission nodes for the grain depot information based on the data transmission node parameters to obtain the data transmission nodes.

[0111] In the embodiments of the present invention, the data acquisition frequency is determined according to different types of IoT sensors (such as temperature and humidity sensors, gas sensors, etc.). Specifically, the temperature and humidity sensor acquires data once per second, and the gas sensor acquires data once every 5 seconds. Determine the unit and format of the data acquired by each sensor. Specifically, the data output by the temperature and humidity sensor each time is a 16-bit integer, and the gas sensor outputs 32-bit floating-point numbers. Calculate the daily storage requirement of a single sensor according to the acquisition frequency and data unit of the sensor. Specifically, a sensor with 16-bit data generates 2 bytes of data per second, and generates 2 bytes × 60 seconds × 60 minutes × 24 hours = 172,800 bytes (about 0.17MB) of data per day. Perform similar calculations for each sensor, and then accumulate the total data volume of all sensors to obtain the total data capacity of the grain depot information. Set the transmission rate of each data transmission node according to the data capacity of the grain depot information and the bandwidth requirement of the transmission node. The optional rate range is from 10 kbps to 100 Gbps. The specific rate setting is based on the following factors: calculate the average transmission rate according to the daily data volume and the number of nodes, and according to the specific application requirements, ensure that the transmission rate between nodes meets the optimal performance and avoid bottlenecks. Specifically, for a high-density sensor layout, a higher transmission rate can be selected. Set the bandwidth requirement of the data transmission node, and the bandwidth requirement range is from 50 kbps to 100 Gbps. The bandwidth setting needs to meet the following requirements: set the bandwidth requirement according to the daily uploaded data volume and the transmission rate, and considering the network type (such as Wi-Fi, LoRa, Zigbee, etc.) and the stability of network transmission, ensure that the transmission bandwidth is sufficient to support the data traffic. Set the transmission frequency of each data transmission node, and the range of the transmission frequency is from once per second to once per hour. The transmission frequency needs to be set according to the data update frequency of the sensor and the real-time requirement: if the sensor requires high real-time performance (specific temperature and humidity monitoring), the transmission frequency can be set to once per second. For environmental monitoring data (such as gas concentration), the transmission frequency can be set to once per minute or once per hour. Set the data volume range from 10 KB / day to 50 GB / day. The delay requirement is set to ≤200 ms. The maximum number of node connections for each base station should be 1000 nodes / base station. Design a reasonable network topology according to parameters such as the number of nodes, transmission rate, and bandwidth requirement. Topologies such as star, mesh, and tree can be adopted to ensure that each sensor node can efficiently upload data to the central system or cloud platform. Arrange the positions of the transmission nodes reasonably according to the actual structure of the grain depot and the data transmission requirements. The transmission nodes should be set in places that can cover all sensor areas to reduce the data transmission distance and network delay. For a large warehouse or multi-layer structure, multiple data forwarding nodes can be set in different areas to ensure signal coverage of the entire area. After the deployment is completed, regularly check the working status of the transmission nodes to ensure that their data transmission performance meets the expectations.

[0112] Preferably, the construction of the virtual granary with the grain depot storage information in the data visualization interface in step S3 includes:

[0113] Extracting the grain depot storage information in the data visualization interface;

[0114] Performing time series analysis on the grain depot storage information to generate grain depot storage time series data;

[0115] Based on the grain depot storage time series data, mirror-copying the storage structure of the bin 3D model to generate a virtual storage framework;

[0116] Performing simulated node deployment on the virtual storage framework to generate a virtual granary.

[0117] In the embodiments of the present invention, data scraping is performed through interfaces with a grain depot information management system, an IoT sensor system, and other relevant data sources (such as an inventory management system, a climate monitoring system, etc.) to extract the warehousing information of the grain depot. The warehousing information includes, but is not limited to, data such as the location of the warehouse, the size of the storage unit, the inventory quantity, the type of goods, the warehousing conditions, the storage temperature and humidity, and the gas concentration. The extracted raw data needs to be structurally processed, including: integrating the warehousing information from different sources into a unified data structure, such as a database table or JSON format, handling missing data or outliers, and ensuring the integrity and accuracy of the data. Mapping the extracted data to specific warehousing information modules on the visualization interface, specifically: the specific location of the granary, the temperature and humidity of each storage unit, the inventory quantity, etc. information, and the environmental conditions (climate, ventilation, etc.). Converting the warehousing information extracted from the data visualization interface into a time series data format to ensure that it can reflect the time changes of the grain depot information. Specifically, information such as the inventory quantity, temperature and humidity, and gas concentration is dynamically changing, so each data collection needs to be marked with a timestamp. Processing the collected time series data, specifically including: analyzing the change trends of each item of data through time series analysis algorithms (such as moving average, Fourier transform, etc.), identifying whether there are periodic fluctuations in the warehousing conditions (such as temperature and humidity), using statistical methods or machine learning algorithms (such as outlier detection algorithms) to identify abnormal fluctuations or events in the data, and standardizing the time series data for subsequent modeling use to ensure the unity of the dimensions of each item of data. Extracting the existing 3D model of the granary, which can be structural data generated by lidar scanning or other modeling tools (such as CAD, BIM). This model contains detailed information such as the building appearance of the grain depot, the internal structure (such as shelves, storage areas, passages, etc.), and the equipment locations (such as sensors, ventilation equipment). Using the time series data to mirror and copy the existing 3D model of the granary to construct a virtual warehousing framework. This framework will include: based on the warehousing time series data, the warehousing structure mirror image will be continuously adjusted and updated to reflect the real-time changes in the inventory quantity and storage conditions. Using 3D modeling software (such as Blender, Unity, Unreal Engine, etc.) to generate the virtual framework. According to the data at different time nodes, simulate the dynamic effects of inventory changes, item movements, and environmental conditions (such as temperature and humidity changes) on the shelves. Adding functional modules to the mirrored warehousing framework, specifically: dynamically displaying warehousing information based on the time series data, such as the inventory quantity, temperature and humidity, gas concentration, etc., adding various equipment (such as ventilation equipment, temperature and humidity control equipment, etc.) and sensor nodes in the virtual warehouse to monitor the warehousing environment in real time. In the virtual warehousing framework, according to the structural and functional requirements of the actual grain depot, deploy simulation nodes, which include: such as temperature and humidity sensors, gas sensors, etc., and these nodes will simulate the actual warehouse environment monitoring. Specifically, temperature control equipment, ventilation equipment, etc., to simulate the control system in the actual grain depot.Allow users to perform interactive operations in the virtual warehouse, such as controlling temperature and humidity, adjusting storage locations, etc. Each simulation node makes dynamic responses according to the changes in the time-series data. Specifically: the temperature and humidity sensor nodes feedback the temperature and humidity data in real time according to the environmental conditions in the virtual warehouse, and the control device nodes automatically start or stop the devices according to the environmental changes, such as starting the refrigeration device when the temperature is too high. After the simulation node deployment is completed, a complete virtual granary is generated through the rendering tool.

[0118] Preferably, the 3D particle simulation of the virtual granary in step S3 includes:

[0119] Extract the temperature and humidity inside the virtual granary, the layout of the storage locations inside the granary, and the grain particle data;

[0120] Add initial simulation attributes to the grain particle data to generate particle physical attributes, where the attribute addition includes particle size distribution, density, and elasticity;

[0121] Calculate the air flow direction and speed of the storage location layout inside the granary according to the temperature and humidity, and combine the grain particle data to simulate the flow path of the grain particles to generate grain flow simulation data;

[0122] Simulate the interactions and collisions between the grain particles in the grain flow simulation data based on the particle physical attributes to obtain the grain movement simulation data;

[0123] Simulate the stacking and movement of the particles in the warehouse for the grain movement simulation data through the storage location layout inside the granary to obtain the grain stacking simulation data;

[0124] Integrate the grain flow simulation data, the grain movement simulation data, and the grain stacking simulation data into the particle simulation result.

[0125] In the embodiments of the present invention, the temperature and humidity information inside the warehouse is extracted from the sensor data of the virtual granary, and these data will be used for the subsequent calculation of the air flow direction and speed. The shelf layout data inside the warehouse is extracted, including information such as the position, size, and shape of each storage location, especially the storage location and form of the grain. The basic data of the grain particles are extracted, and these data include basic physical properties such as the type, size, and density of the particles. In addition, the form of the grain particles (such as long strip shape, spherical shape, etc.) and the interaction between them also need to be considered. Statistical analysis is carried out on the size of the grain particles, and they are classified according to the particle size distribution. Specifically, the grain particles can be divided into different size ranges to reflect the influence of different particle sizes on flow and accumulation. According to the type and characteristics of the grain particles, appropriate density values are assigned to each particle. Different types of grains (such as wheat, corn, etc.) have different densities, which will affect their fluidity and accumulation in the warehouse. An elastic property is added to the grain particles, which includes the recovery coefficient when the particles collide with each other. This property can be estimated based on the physical properties of the actual grain to ensure that the particle collision behavior in the simulation is consistent with the real physical environment. According to the temperature and humidity data inside the warehouse, the Navier-Stokes equation in the computational fluid dynamics model is used to calculate the air flow direction and speed inside the warehouse. The direction and speed of the air flow will be affected by the temperature and humidity. Specifically, the air flow speed in the high-temperature area is relatively fast. The storage location layout inside the warehouse (such as the distribution of shelves, the width of the channels, etc.) will have an impact on the air flow. It is necessary to consider how these physical structures guide the air flow and their influence on the flow path of the grain particles. Based on the calculated air flow direction and speed, combined with the initial position and physical properties of the grain particles, the discrete element method in the particle flow simulation method is used to calculate the flow path of the grain particles in the warehouse. At this time, the flow of the grain particles will be jointly affected by the air flow, temperature and humidity, and the storage location layout. Based on the physical properties of the particles (such as elasticity, density, friction coefficient, etc.), the interaction between the grain particles is simulated in the simulation, which includes: the grain particles will collide during the flow process. The physical collision model (such as elastic collision or inelastic collision model) is used to simulate the interaction between the particles. The frictional force between the grain particles and the warehouse surface (such as the ground, shelves, etc.) will affect their flow and accumulation. The frictional force is calculated according to the surface properties of the particles (such as roughness). In some cases, the particles will aggregate together (such as in a narrow channel) and will also be dispersed due to the influence of the air flow. Appropriate mathematical models are used to simulate these phenomena. During the simulation of the movement of the grain particles, considering the influence of the storage location layout inside the warehouse, the accumulation behavior of the particles on the shelves and the ground is simulated. The accumulation behavior of the particles is affected by the following factors: the accumulation patterns of large particles and small particles are different. Larger particles will get stuck in a smaller space, while smaller particles can more easily fill the space. The layout, spacing of the storage locations inside the warehouse, and the inclination angle of the shelves, etc. will all have an impact on the accumulation pattern of the grain particles.In the simulation, it is also necessary to consider the movement of particles during the stacking process, and simulate how particles migrate from one position to another under the action of air flow and gravity. This process can be achieved through the discrete element method (DEM) or other particle motion models. Integrate the simulation data of grain flow, movement and stacking to generate comprehensive particle simulation results, which will reflect the overall behavior of grain in the warehouse. The final particle simulation results can be rendered through 3D visualization tools (such as Blender, Unity, Unreal Engine, etc.) to form a dynamic display of the movement, stacking and flow of grain particles in the virtual granary.

[0126] Preferably, the virtual-real contrast warning for the 3D model of the granary according to the particle simulation results in step S3 includes:

[0127] Identify abnormal grain conditions for the 3D model of the granary according to the particle simulation results. When any of the following situations occur, determine that the particle simulation result is abnormal grain flow and obtain abnormal grain flow data: the grain flow rate deviates from the optimal range by more than ±10%; the grain flow deviates from the predetermined direction by more than 15°;

[0128] When the following situations occur simultaneously, determine that the particle simulation result is abnormal grain movement and obtain abnormal grain movement data: the grain acceleration detection is lower than 0.5 m / s² for 3 consecutive times; the lag time of the grain movement trajectory exceeds 20 min; the fluctuation of the grain movement speed is greater than ±15% and lasts for more than 30 min;

[0129] When the following situations occur simultaneously, determine that the particle simulation result is abnormal grain stacking and obtain abnormal grain stacking data: the change in grain stacking density exceeds 10% within 1 hour; the deviation of the grain stacking height from the preset stacking mode in the warehouse exceeds 20%; the temperature in the grain stacking area deviates from the normal range by more than 5°C;

[0130] Integrate the abnormal grain flow data, abnormal grain movement data and abnormal grain stacking data to obtain abnormal grain condition data;

[0131] Generate corresponding dynamic grain condition warning signals according to the type of abnormal grain condition data.

[0132] In the embodiments of the present invention, potential abnormal situations are analyzed and identified based on the grain flow, movement, and accumulation data in the particle simulation results. The specific situations are as follows: When the flow rate of the grain in the particle simulation deviates from the optimal flow rate range by more than ±10%, it is determined that the grain flow is abnormal. The optimal range of the flow rate is usually preset according to the warehouse design, the type of grain, and its characteristics. If the flow direction of the grain particles deviates from the predetermined direction by more than 15°, it is considered that the grain flow is abnormal. The predetermined direction is usually determined based on the airflow simulation, the warehouse layout, and the flow characteristics of the grain particles. By detecting the acceleration of the grain particles, if the grain acceleration is detected to be lower than 0.5 m / s² for three consecutive times, it is determined that the grain movement is abnormal. An acceleration lower than this value indicates that the grain particles are in a static state or have poor flow. If the movement trajectory of the grain particles has a lag time exceeding 20 minutes, that is, the particles fail to reach the predetermined target position within the expected time, it is determined that the grain movement is abnormal. The lag time indicates system failure or poor airflow. If the movement speed of the grain particles fluctuates by more than ±15% and lasts for more than 30 minutes, it is considered that the grain movement is abnormal. The continuous speed fluctuation indicates unstable movement caused by changes in the internal environment of the warehouse or the characteristics of the grain particles. If the density change of the grain accumulation exceeds 10% within 1 hour, it is determined that the grain accumulation is abnormal. An excessive density change in the accumulation indicates abnormal interaction between particles or problems such as poor ventilation in the warehouse. If the height of the grain accumulation deviates from the preset accumulation pattern in the warehouse by more than 20%, it is determined that the grain accumulation is abnormal. The warehouse design usually has a predetermined accumulation pattern and height, and an excessive deviation indicates uneven accumulation or abnormal flow. If the temperature in the accumulation area deviates from the normal range by more than 5°C, it is determined that the grain accumulation is abnormal. The abnormal temperature affects the quality of the grain and indicates problems with the ventilation or environmental control system in the warehouse. By identifying the abnormal situations of grain flow, movement, and accumulation, collecting all abnormal information (such as flow rate deviation, low acceleration, excessive lag time, large density change, etc.), and integrating it into a comprehensive abnormal grain situation data. The specific integration is as follows: Each abnormal type (flow abnormality, movement abnormality, accumulation abnormality) is quantified and scored separately. For each abnormality, different weight values are assigned, and the severity of each abnormality and its impact on warehouse operations are comprehensively considered to generate comprehensive abnormal data. According to the integrated abnormal grain situation data, different levels of dynamic grain situation warning signals are generated through the set rules. Different types of abnormalities will generate different warning signals.

[0133] As an example of the present invention, refer to Figure 3 shown, in this example, step S4 includes:

[0134] Step S41: Analyze the device control parameters of the dynamic grain situation warning signal to generate a device control instruction dataset;

[0135] Step S42: Based on the automation equipment control instruction data set, combined with the grain condition monitoring data collected in real time by the IoT sensor, the instruction matching degree is verified to generate equipment status optimization instruction parameters;

[0136] Step S43: Input the equipment status optimization instruction parameters into the 3D model of the granary to perform pulse coordinated adaptive control and generate the optimal management strategy to execute the three-dimensional digital management operation of the grain warehouse.

[0137] In the embodiment of the present invention, by analyzing the various control parameters involved in the dynamic grain condition warning signal according to the type (such as abnormal flow, abnormal movement, abnormal accumulation, etc.), these parameters generally include: temperature and humidity control, air flow speed adjustment, grain flow direction control, grain accumulation area control, etc. The abnormal condition of each warning signal is converted into a specific equipment control demand. Specifically: abnormal flow requires adjustment of air flow direction and speed; abnormal accumulation requires adjustment of accumulation density or temperature of grain accumulation area, etc. Based on the above-analyzed warning signal, each control demand is mapped to a control instruction of the automation equipment, including specific control parameters, thresholds, operation time, etc. Specifically: adjust the running speed and direction of the fan; set the temperature and humidity target value of the temperature and humidity control device; adjust the speed, direction or other control settings of the grain flow device. All control instructions and their parameters are summarized to generate an equipment control instruction data set, which will provide a reference basis for the execution of subsequent steps. IoT sensors are arranged in the grain warehouse to monitor grain condition data (such as temperature and humidity, grain flow state, accumulation density, etc.) in real time. The data collected by the sensor include: actual temperature and humidity values; actual grain flow speed and direction; density and accumulation height of the accumulation area, etc. Verify the matching degree between the real-time collected grain monitoring data and the equipment control instruction data set, and analyze the execution effect of the control instructions in the actual environment. Specifically: compare the difference between the temperature and humidity values measured by the sensor and the target value;

[0138] Check the compliance of the grain flow speed with the target speed; verify the deviation of the grain stacking height from the target stacking pattern. According to the result of the instruction matching verification, adjust the parameters in the original equipment control instruction to achieve more efficient and accurate equipment control. Specifically: adjust the setting range of the temperature and humidity control equipment; optimize the wind speed, wind direction, etc. of the air flow regulating equipment; adjust the operation mode and speed of the grain flow device. Based on the results of verification and adjustment, generate the optimized instruction parameters for the equipment state, and update the equipment control instruction dataset to improve the accuracy and efficiency of the system response. Input the generated optimized instruction parameters for the equipment state into the 3D digital model of the granary, which should include information such as the spatial layout of the granary, equipment location, sensor settings, etc. Through the pulse cooperative adaptive regulation mechanism, coordinate the optimized instructions with the actual control equipment in the 3D model of the granary. The specific methods include: making pulse adjustments to the equipment according to the real-time feedback data (such as temperature and humidity, air flow speed, etc.), enabling the system to quickly respond to the changes in the grain situation and promptly correct the deviations that occur. Adaptively adjust the equipment control strategy according to the actual operation of the granary. Specifically, when the grain stacking density is too high, automatically strengthen ventilation or adjust the grain flow path. Generate the optimal management strategy by simulating the cooperative effect of the equipment in the 3D model of the granary.

[0139] Particularly important is that the input of the optimized instruction parameters for the equipment state into the 3D model of the granary for pulse cooperative adaptive regulation in step S43 further includes:

[0140] Perform spatio-temporal correlation on the optimized instruction parameters for the equipment state, identify the dynamic change rules of the equipment state in the spatial and temporal dimensions, and generate spatio-temporal feature data of the equipment state;

[0141] Perform grid-based spatial matching on the spatio-temporal feature data of the equipment state and the 3D model data of the granary to generate matching data of the equipment and the 3D spatial position of the granary;

[0142] Conduct pulse regulation modeling on the matching data of the equipment and the granary spatial position to generate pulse regulation data;

[0143] Perform adaptive optimization on the pulse regulation data and conduct comprehensive performance evaluation based on the optimized pulse regulation data to screen the optimal management strategy.

[0144] In the embodiments of the present invention, time series data of equipment state optimization instructions are collected, including change records of various control parameters (such as temperature and humidity, air flow velocity, grain flow direction, etc.) at different time points. The changes of equipment control parameters are analyzed in spatio-temporal correlation with spatial positions (equipment positions, each area of the granary). By analyzing the fluctuation law of equipment state in the time dimension (such as the diurnal change of temperature) and the distribution in the spatial dimension (such as the temperature and humidity differences in different warehouse areas), the dynamic change law between the equipment state and the granary environment is established. Spatial features of the equipment location are extracted, such as the specific location of the equipment in the warehouse, functional zoning, etc. Temporal features of equipment control are extracted, such as the frequency and amplitude of control instruction changes. Based on spatio-temporal correlation analysis, spatio-temporal feature data of equipment state are generated, and these data reflect the laws of equipment changes in space and time, specifically the change trend, periodicity, etc. of the equipment state. The spatio-temporal feature data are formatted for subsequent matching with the spatial data in the 3D model of the granary. The 3D model of the granary is divided into several grid cells according to the spatial dimension, and each grid cell represents the coverage range of an area or equipment in the warehouse. Each cell has independent spatial coordinates and can reflect the environmental conditions such as temperature, humidity, and air flow in this area. For each grid cell, information on the interaction between the equipment and the environment is recorded, such as the spatial features of the equipment location and the relative position relationship between the equipment and the environment. According to the spatio-temporal feature data of the equipment, the changes in the spatial and temporal dimensions are analyzed, and these data are spatially matched with the grid cells in the 3D model of the granary. The specific steps include: docking according to the location of the equipment and the grid cells to ensure that the changes of the equipment can be mapped to the relevant areas of the warehouse. Translating the spatio-temporal feature data of the equipment in time to ensure that the data can correspond to the actual environmental changes in the granary. Through the matching of the spatio-temporal feature data and the 3D spatial position of the granary, matching data of the equipment and the 3D spatial position of the granary are generated, and these data provide a basis for subsequent pulse regulation modeling. Based on the matching data of the equipment and the granary spatial position, a pulse regulation model is established. This model dynamically adjusts the operating state of the equipment according to different granary environments and equipment control requirements. The purpose of pulse regulation is to respond to the changes in the grain situation in the granary in real time and quickly. A pulse regulation controller is designed to generate control signals according to the state changes of the equipment in the granary and drive the equipment to perform necessary adjustment operations (such as increasing air flow, adjusting temperature and humidity, etc.). Different pulse regulation strategies are designed, including continuous adjustment, periodic adjustment, emergency response, etc. to cope with different grain situation changes. Through pulse regulation modeling, corresponding pulse regulation data are generated. This data includes information such as specific regulation timing, regulation amplitude, and regulation frequency, and these data will be used for the actual operation adjustment of the equipment. An adaptive optimization algorithm is used to adjust the pulse regulation parameters according to the actual operation feedback of the equipment and the granary environment change data.The optimization process includes: adjusting the pulse control strategy according to real-time monitoring data (such as temperature, humidity, and grain flow status) to ensure that the control operations of the equipment always adapt to environmental requirements. Considering multi-dimensional objectives such as equipment energy efficiency and grain condition stability, optimize the control strategy to reduce energy consumption and improve management efficiency. Based on the optimized pulse control data, conduct a comprehensive performance evaluation. The evaluation indicators include: the quality of grain storage (such as the stability of temperature and humidity, the smoothness of grain flow, the uniformity of stacking, etc.); the operating efficiency of the equipment (such as energy-saving effect, equipment load conditions, etc.); the system response time and stability, etc. According to the evaluation results, select the optimal pulse control management strategy and apply it to the digital management of the grain depot.

[0145] Particularly important is that the pulse control modeling of the matching data of the equipment and the bin space position specifically includes:

[0146] Extract the dynamic spatio-temporal characteristics of the matching data of the equipment and the bin space position, and conduct pulse signal modeling on the matching data according to the dynamic spatio-temporal characteristics to generate preliminary pulse control data;

[0147] Conduct global coupling optimization on the preliminary pulse control data, optimize the amplitude, frequency, and phase of the pulse control, and generate pulse control data after coupling optimization;

[0148] Conduct multi-objective constraint optimization on the pulse control data after coupling optimization, where the multi-objective optimization includes energy consumption, response time, and load balance, to generate multi-objective optimized pulse control data;

[0149] Conduct spatio-temporal spectrum analysis on the multi-objective optimized pulse control data to generate pulse spectrum analysis data;

[0150] According to the pulse spectrum analysis data, conduct real-time pulse control adjustment on the optimized pulse control data to generate pulse control data.

[0151] In the embodiments of the present invention, relevant dynamic spatio-temporal features are extracted from the matching data of the device and the silo space position. The spatio-temporal features include the change in the device position (such as the movement path of the device in the warehouse), the granularity of the time change (the update frequency of the specific device state), and the relationship between these changes and the silo environment (such as temperature and humidity, grain flow, etc.). According to the extracted dynamic spatio-temporal features, a pulse signal model is constructed using signal modeling methods (such as Fourier transform-based or time-domain analysis). The modeling of the pulse signal will consider the control requirements of the devices in the silo and the real-time changes in environmental conditions (such as temperature and humidity, air flow, etc.). Pulse signals (specifically in the forms of Gaussian pulses, sine wave pulses, etc.) are constructed using spatio-temporal features to simulate the responses of the devices in different states. The pulse signal will reflect factors such as the amplitude, frequency, and timing of the control command. Through the pulse signal data generated by the above modeling, the preliminary control data includes features such as the amplitude, period, and phase of the signal, which are used to preliminarily control the device state. The amplitude, frequency, and phase of the pulse signal are used as optimization parameters, and optimization objectives are set to achieve the improvement of the global performance. The optimization objectives include: ensuring that the amplitude of the pulse signal is suitable for the control requirements of each area in the silo, avoiding over-adjustment or under-adjustment, adjusting the frequency of the pulse signal to achieve efficient regulation of the environmental conditions in the warehouse, and ensuring the coordinated operation of each device and improving the overall working efficiency by adjusting the phase of the pulse signal. A global optimization algorithm (such as genetic algorithm, particle swarm optimization, or simulated annealing algorithm) is used for optimization to find the best combination of pulse regulation amplitude, frequency, and phase. The algorithm calculates the advantages and disadvantages of each parameter combination by simulating the influence of different pulse parameters on the performance of the silo system, and finally obtains the optimized pulse regulation data. When optimizing the pulse regulation, the minimization of device energy consumption is considered to ensure the energy efficiency of the silo operation and reduce unnecessary energy consumption. Ensure that the pulse signal can respond to environmental changes in a timely manner, such as temperature and humidity fluctuations, abnormal grain flow, etc., and reduce the response delay. Reasonably allocate the device load to avoid device overload or inefficient operation and ensure the rational use of resources. A multi-objective optimization algorithm (such as weighted sum method, ε-constraint method, Pareto optimization, etc.) is used to optimize the pulse regulation data. In this step, all objective functions (energy consumption, response time, load balance) will be integrated into a composite objective, and the balance point between the objectives is obtained through the optimization algorithm to generate pulse regulation data that meets multi-objective constraints. Spatio-temporal spectrum analysis is performed on the optimized pulse regulation data. Through methods such as Fourier transform and wavelet analysis, the spectral characteristics of the pulse signal are analyzed, and the frequency components in the signal are extracted. According to the spatio-temporal distribution of the pulse signal, the spectral characteristics of the signal, specifically the frequency range, energy distribution, etc., are extracted, and its variation law in space and time is analyzed. This data reflects how the intensity, frequency, and phase of the pulse regulation signal change over time and space. The results obtained through spectrum analysis can provide a basis for further optimization of pulse regulation.Spectrum analysis data includes the frequency components, phase relationships, amplitude information, etc. of a signal, which can help evaluate the adaptability and stability of a pulse signal. According to the pulse spectrum analysis data, the pulse regulation parameters are adjusted in real time. This includes: adjusting the frequency of the pulse signal according to the change in frequency components in the spectrum to meet real-time requirements; dynamically adjusting the amplitude of the pulse signal according to environmental changes such as temperature, humidity, and grain flow in the granary to achieve the best control effect; adjusting the phase of the pulse signal according to the equipment status and granary requirements to ensure the coordination and effectiveness of the regulation signal. According to the real-time pulse regulation adjustment results, the final pulse regulation data is generated.

[0152] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be embraced within the present invention.

[0153] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A three-dimensional digital management method for grain depots based on digital twins, characterized in that, It includes the following steps: Step S1: Obtain the grain depot structure data and the grain depot environment data; Based on the grain depot structure data, conduct three-dimensional refined modeling to obtain a panoramic digital twin scenario of the grain depot; adjust the geospatial accuracy of the panoramic digital twin scenario of the grain depot according to the grain depot environment data to obtain the bin 3D model; Step S2: Use IoT sensors to collect grain depot information and upload the grain depot information to the bin 3D model for data integration and annotation to establish a data visualization interface; Step S3: Extract the grain storage information of the data visualization interface to construct a virtual granary; Conduct 3D particle simulation on the virtual granary, and perform virtual-real contrast warning on the bin 3D model according to the particle simulation results to generate a dynamic grain condition warning signal; among them, the 3D particle simulation of the virtual granary in Step S3 includes: Extract the temperature and humidity inside the virtual granary, the layout of in-bin storage locations, and the grain particle data; Add simulation initial attributes to the grain particle data to generate particle physical attributes, where the attribute addition includes particle size distribution, density, and elasticity; Calculate the air flow direction and velocity of the in-bin storage location layout according to the temperature and humidity, and combine the grain particle data to simulate the flow path of the grain particles to generate grain flow simulation data; Based on the particle physical attributes, simulate the interaction and collision between the grain particles in the grain flow simulation data to obtain grain movement simulation data; Through the in-bin storage location layout, simulate the accumulation and movement of particles in the warehouse for the grain movement simulation data to obtain grain accumulation simulation data; Integrate the grain flow simulation data, the grain movement simulation data, and the grain accumulation simulation data into the particle simulation result; Step S4: Associate the automated equipment of the grain depot through the dynamic grain condition warning signal and perform adaptive control to obtain an optimal management strategy and execute the three-dimensional digital management operation of the grain depot. Among them, Step S4 includes the following steps: Step S41: Analyze the equipment control parameters of the dynamic grain condition warning signal to generate a device control instruction data set; Step S42: Based on the automated equipment control instruction data set, combine the grain condition monitoring data collected in real time by IoT sensors to verify the instruction matching degree and generate device state optimization instruction parameters; Step S43: Input the device state optimization instruction parameters into the bin 3D model for pulse collaborative adaptive control to generate an optimal management strategy to execute the three-dimensional digital management operation of the grain depot; among them, the input of the device state optimization instruction parameters into the bin 3D model for pulse collaborative adaptive control in Step S43 further includes: Perform spatio-temporal association on the device state optimization instruction parameters to identify the dynamic change rules of the device state in the spatial and temporal dimensions and generate device state spatio-temporal feature data; Perform grid-based spatial matching on the device state spatio-temporal feature data and the bin 3D model data to generate matching data of the device and the bin 3D spatial position; Perform pulse control modeling on the matching data of the device and the bin spatial position to generate pulse control data; Adaptive optimization is performed on the pulse regulation data, and comprehensive performance evaluation is carried out based on the optimized pulse regulation data to screen the optimal management strategy; among them, the pulse regulation modeling of the matching data of the equipment and the warehouse space position specifically includes: Extract the dynamic spatio-temporal features of the matching data of the equipment and the warehouse space position, and perform pulse signal modeling on the matching data according to the dynamic spatio-temporal features to generate preliminary pulse regulation data; Perform global coupling optimization on the preliminary pulse regulation data, optimize the amplitude, frequency and phase of the pulse regulation, and generate the pulse regulation data after coupling optimization; Perform multi-objective constraint optimization on the pulse regulation data after coupling optimization, where the multi-objective optimization includes energy consumption, response time, and load balancing, and generate multi-objective optimized pulse regulation data; Perform spatio-temporal spectrum analysis on the multi-objective optimized pulse regulation data to generate pulse spectrum analysis data; Perform real-time pulse regulation adjustment on the optimized pulse regulation data according to the pulse spectrum analysis data to generate pulse regulation data.

2. The three-dimensional digital management method of a grain depot based on digital twin according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Perform three-dimensional point cloud scanning on the grain depot by using a lidar to obtain grain depot structure data; Step S12: Use meteorological monitoring sensors to obtain grain depot environment data; Step S13: Perform point cloud fusion on the grain depot structure data to generate a three-dimensional structure framework of the grain depot; perform fine-grained hierarchical modeling on the three-dimensional structure framework of the grain depot to generate high-precision grain depot hierarchical information, where the hierarchical modeling includes structure layer modeling, function layer modeling, and equipment layer modeling; Step S14: Extract the material, illumination, and color of the grain depot structure data, and perform texture mapping on the high-precision grain depot hierarchical information to generate a panoramic digital twin scene of the grain depot; Step S15: Perform geographic coordinate calibration based on the grain depot environment data to obtain geographic calibration data; use the geographic calibration data to perform spatial optimization on the panoramic digital twin scene of the grain depot to obtain a 3D model of the warehouse.

3. The three-dimensional digital management method of a grain depot based on digital twins according to claim 2, wherein Step S15 includes the following steps: Step S151: Extract the grain depot center point, boundary point, and key structure point in the grain depot environment data as geographic coordinate reference benchmark data; Step S152: Confirm the global coordinates and regional standard coordinates based on the geographic coordinate reference benchmark data, so as to obtain global coordinate data and regional standard coordinate data respectively; Step S153: Calculate the coordinate offset vector of the global coordinate data and the regional standard coordinate data to obtain geographic calibration data; Step S154: Perform scene space matching on the panoramic digital twin scene of the grain depot through the geographic calibration data to generate a 3D model of the warehouse.

4. The three-dimensional digital management method of a grain depot based on digital twin according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Use loT sensors to collect grain depot information; Step S22: Calculate the data capacity of the grain depot information and set data transmission nodes to obtain data transmission nodes; Step S23: Distributively upload the grain depot information to the 3D model of the warehouse according to the data transmission nodes for data integration to generate integrated grain depot information; Step S24: Bind and visualize the module data of the integrated grain depot information and the 3D model of the warehouse to generate a data visualization interface.

5. The three-dimensional digital management method for grain depots based on digital twins according to claim 4, characterized in that, Step S22 includes the following steps: Calculate the data capacity of the grain depot information; Set data transmission node parameters according to the data capacity: the data transmission rate is from 10 kbps to 100 Gbps; the bandwidth requirement is from 50 kbps to 100 Gbps; the transmission frequency is from 1 time per second to 1 time per hour; the data volume is from 10 KB / day to 50 GB / day; the latency requirement ≤ 200 ms; the maximum number of node connections is 1000 nodes / base station; Deploy data transmission nodes for the grain depot information based on the data transmission node parameters to obtain data transmission nodes.

6. The three-dimensional digital management method of a grain depot based on digital twin according to claim 1, characterized in that The construction of a virtual granary by extracting the grain depot storage information of the data visualization interface described in step S3 includes: Extract the grain depot storage information of the data visualization interface; Conduct time series analysis on the grain depot storage information to generate grain depot storage time series data; Based on the grain depot storage time series data, perform a mirror copy of the storage structure on the 3D model of the granary to generate a virtual storage framework; Perform simulated node deployment on the virtual storage framework to generate a virtual granary.

7. The three-dimensional digital management method of a grain depot based on digital twin according to claim 1, wherein The virtual-real contrast warning for the 3D model of the granary according to the particle simulation results described in step S3 includes: Identify abnormal grain conditions for the 3D model of the granary according to the particle simulation results. When any of the following situations occurs, determine that the particle simulation result is abnormal grain flow and obtain abnormal grain flow data: the grain flow velocity deviates from the optimal range by more than ±10%; the grain flow deviates from the predetermined direction by more than 15°; When the following situations occur simultaneously, determine that the particle simulation result is abnormal grain movement and obtain abnormal grain movement data: the grain acceleration detection is lower than 0.5 m / s² for 3 consecutive times; the lag time of the grain movement trajectory exceeds 20 min; the fluctuation of the grain movement speed is greater than ±15% and lasts for more than 30 min; When the following situations occur simultaneously, determine that the particle simulation result is abnormal grain accumulation and obtain abnormal grain accumulation data: the grain accumulation density changes by more than 10% within 1 hour; the deviation of the grain accumulation height from the preset accumulation mode in the granary exceeds 20%; the temperature in the grain accumulation area deviates from the normal range by more than 5°C; Integrate the abnormal grain flow data, abnormal grain movement data, and abnormal grain accumulation data to obtain abnormal grain condition data; Generate corresponding dynamic grain condition warning signals according to the types of abnormal grain condition data.

8. A three-dimensional digital management system for grain depots based on digital twins, characterized in that, For implementing the three-dimensional digital management method of a grain depot based on digital twin as described in claim 1, the three-dimensional digital management system of the grain depot based on digital twin includes: A scene modeling module, used to obtain grain depot structure data and grain depot environment data; perform three-dimensional fine modeling based on the grain depot structure data to obtain a panoramic digital twin scene of the grain depot; adjust the geospatial accuracy of the panoramic digital twin scene of the grain depot according to the grain depot environment data to obtain a 3D model of the granary; A data upload module, used to collect grain depot information using IoT sensors and upload the grain depot information to the 3D model of the granary for data integration and annotation to establish a data visualization interface; A digital management module, used to extract the grain depot storage information of the data visualization interface to construct a virtual granary; perform 3D particle simulation on the virtual granary, and conduct virtual-real contrast warning on the 3D model of the granary according to the particle simulation results to generate dynamic grain condition warning signals; A feedback control module is used to associate the automated equipment of the grain depot through dynamic grain condition warning signals and perform adaptive regulation to obtain an optimal management strategy and execute the three-dimensional digital management operation of the grain depot.

Images