Information integration and intelligent decision method of intelligent stockyard integrated management and control platform

Abstract

The application discloses an information integration and intelligent decision method of an intelligent stockyard integrated management and control platform, and relates to the technical field of intelligent stockyard management. The specific steps of the decision method are as follows: S100, data acquisition: deploying Internet of Things sensors in the material storage area of the stockyard, the equipment operation site and the transport vehicle; by constructing a complete information integration and intelligent decision method of the intelligent stockyard integrated management and control platform, the application realizes comprehensive, real-time and accurate acquisition and processing of the material state of the stockyard, the equipment operation parameters and the transport information; through the comprehensive use of the Internet of Things sensors, satellite remote sensing data and 5G technology, the application can acquire multi-dimensional data of the stockyard in real time, and through the data quality processing, data fusion and monitoring steps, the accuracy and reliability of the data are effectively improved; this innovative data acquisition and processing mode not only improves the intelligent level of the stockyard management, but also provides a basis for subsequent resource allocation decision and dynamic optimization.

Description

Technical Field

[0001] This invention relates to the field of intelligent material yard management technology, specifically to information integration and intelligent decision-making methods for an integrated intelligent material yard management and control platform. Background Technology

[0002] With the rapid development of modern industrial production, material yards, as a crucial link in logistics and production, have a significant impact on the overall production process due to their management efficiency and level of intelligence. Traditional manual management methods are no longer sufficient to meet the demands of large-scale, high-efficiency production. Therefore, intelligent and information-based material yard management systems have emerged. As an advanced management tool, the intelligent integrated material yard control platform integrates IoT, big data, and artificial intelligence technologies to achieve comprehensive monitoring and optimized management of key aspects of material yard materials, equipment, and transportation, becoming an important development trend in current material yard management.

[0003] However, traditional technologies have many shortcomings in intelligent material yard management. On the one hand, traditional data collection methods often rely on manual inspection and recording, which leads to inaccurate and untimely data, making it difficult to achieve real-time control of the material yard status. On the other hand, traditional data processing and analysis methods often lack the organic integration and in-depth mining of multi-source data, resulting in a lack of scientific basis for resource allocation decisions and making it difficult to achieve optimal resource allocation and efficient utilization. In addition, traditional technologies also lack dynamic monitoring and pattern mining of material yard operation patterns, making it difficult to adjust management strategies based on real-time data, thus limiting the improvement of material yard management level.

[0004] Therefore, developing information integration and intelligent decision-making methods for an integrated intelligent material yard management platform can provide a scientific basis for enterprise decision-making and help promote the transformation and upgrading of enterprises towards intelligence and informatization. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an information integration and intelligent decision-making method for an integrated intelligent material yard management platform. This method collects material yard data in real time through IoT sensors, combines it with satellite remote sensing image data, uses advanced data processing technology for data quality processing, fusion and monitoring, constructs a resource allocation model, introduces resource fairness indicators, formulates the optimal resource allocation strategy, and monitors data in real time to discover potential patterns and achieve dynamic optimization.

[0006] To solve the above-mentioned technical problems, this invention provides the following technical solution: an information integration and intelligent decision-making method for an integrated intelligent material yard management platform, the specific steps of which are as follows:

[0007] S100, Data Acquisition: IoT sensors are deployed in the material storage area, equipment operation site, and transport vehicles of the material yard to collect material status, equipment operating parameters and transportation information in real time. At the same time, satellite remote sensing image data is acquired periodically through the data interface of the satellite remote sensing data provider, and these data are transmitted to the management and control platform using 5G technology.

[0008] S200, Data Quality Processing: The management and control platform uses data validity to judge data quality. For poor quality data, adaptive data interpolation repair formula is used to repair it. The data collection items and frequency are determined according to the collection plan and business needs, and the time format of data from different sources is unified.

[0009] S300, Data Fusion and Monitoring: It performs format unification and noise reduction processing on data collected by IoT sensors and constructs micro data feature vectors. It processes satellite remote sensing image data using image recognition and geographic information system technologies, constructs macro data feature vectors, and fuses them through a multi-dimensional feature fusion mapping algorithm.

[0010] S400, Resource Allocation Decision: By surveying the needs of various departments in the material yard, a resource allocation model is established using game theory. A resource fairness index is introduced to quantify the degree of fairness. An iterative algorithm is used to adjust the resource allocation scheme and formulate the optimal resource allocation strategy.

[0011] S500, Dynamic Optimization and Pattern Application: Real-time monitoring of various data in the material yard, analysis of causal relationships between data to uncover potential patterns, dynamic adjustment of resource allocation strategies based on potential patterns, and optimization of the overall operation of the material yard based on potential patterns.

[0012] Furthermore, in S100, the sensors used at each location during data acquisition are:

[0013] Material storage area:

[0014] Temperature and humidity sensors: installed at different heights of material stacks, in warehouse corners, and at ventilation openings;

[0015] Composition analyzer: installed on the material conveyor belt or near the sampling point of the material stack;

[0016] Liquid level sensor: Installed inside the liquid storage tank, vertically downwards from the top of the tank;

[0017] Equipment in operation site:

[0018] Vibration sensor: installed in the bearing housing of the motor or the outer casing of the gearbox;

[0019] Speed ​​sensor:

[0020] Installation location: Install near the rotating axis of the equipment;

[0021] Load sensor: installed on the hook of a crane or the support structure of a conveyor belt;

[0022] Transport vehicles:

[0023] GPS positioning device: installed on the top or inside of the vehicle;

[0024] Speed ​​sensor: Installed near the vehicle's wheels;

[0025] Weight sensor: Installed on the chassis or cargo box of the vehicle.

[0026] Furthermore, in step S200, the data quality processing combines data validity to determine data quality, and includes... Data The data validity is The calculation formula is: ,in and The boundary values ​​for the normal data fluctuation range are set based on the characteristics of the equipment. The total number of data collected. The data quality is acceptable; otherwise, the data quality is unacceptable. and It is the set threshold for judging data quality.

[0027] Furthermore, in step S200, during data quality processing, abnormal data is repaired using an adaptive data interpolation repair formula. Let the existing abnormal data points be... Its adjacent normal data points are and The data change trend coefficient is The repaired data The calculation formula is: ,in, This is the repaired data. These are the original outlier data points, and the data change trend coefficient. Through the front and back Calculate the average slope of each normal data point.

[0028] Furthermore, in S300, the specific steps for constructing microscopic data feature vectors in data fusion and monitoring are as follows:

[0029] (1) Determine the standard data format, parse the raw data from different sensors and convert them into a unified format;

[0030] (2) Identify noise in the data. For high-frequency noise, use a low-pass filter to filter it, and for low-frequency interference, use a high-pass filter to process it.

[0031] (3) Calculate the basic statistical characteristics of the data, extract time-related features, convert the time-domain data to the frequency domain, and extract frequency features;

[0032] (4) Determine the dimensions according to management needs, and arrange the extracted features in an orderly manner to construct a micro data feature vector.

[0033] Furthermore, in S300, the specific steps for constructing macroscopic data feature vectors in data fusion and monitoring are as follows:

[0034] (1) Perform radiometric correction, geometric correction and image enhancement operations on satellite remote sensing image data;

[0035] (2) Classify different land features in remote sensing images and accurately identify and locate specific targets in the material yard;

[0036] (3) Register the processed images with the basic data in the geographic information system, conduct spatial analysis, and integrate multi-source geographic data to construct a geographic information database for the material yard;

[0037] (4) Screen and extract features related to material yard management, and combine them to form a macro data feature vector.

[0038] Furthermore, in S300, the data fusion and monitoring process fuses microscopic data feature vectors and macroscopic data feature vectors using a multi-dimensional feature fusion mapping algorithm. Let the microscopic data feature vector be... The macroscopic data feature vector is The fusion weight matrix is It is The matrix, elements It is the first in the matrix Line number Column element, representing the first The weights of each feature are set based on the importance and relevance of the data, resulting in the fused feature vector. The calculation formula is: .

[0039] Furthermore, in S400, the resource allocation decision uses game theory to establish a resource allocation model to formulate intelligent material yard resource allocation, assuming that there are [resources] in the material yard. Participating entities The resource demand vector for each entity is: , Represents the number of resource types, and the revenue function is: , It refers to the number of participants in the material yard, introducing a resource equity index. The calculation method is as follows: ,in It is the first The average allocation of various resources is calculated by surveying the needs of each department in the material yard, and the resource allocation vectors of each entity are continuously adjusted. To maximize the resource equity index while meeting the basic needs of all stakeholders. At the same time, it takes into account the revenue functions of each subject. The optimal resource allocation scheme is solved by using an iterative algorithm.

[0040] Furthermore, in S500, the calculation of the causal correlation strength coefficient in dynamic optimization and regular application is assumed to be based on the material moisture content. and equipment operating efficiency Collected over a period of time Group data , It refers to the number of data sets collected and the causal relationship strength coefficient. The calculation formula is: ,in, and These are the average values ​​of humidity and equipment operating efficiency, respectively. This indicates the moisture content of the material. and equipment operating efficiency There is a significant correlation between them, and The larger the size, the stronger the causal relationship. They concluded that there was no significant correlation between the two, among which It is the threshold for determining the strength of causal association through statistical significance testing.

[0041] Compared with existing technologies, the information integration and intelligent decision-making methods of this intelligent material yard integrated management and control platform have the following beneficial effects:

[0042] I. This invention, through the construction of a complete intelligent integrated management and control platform for material yards, achieves comprehensive, real-time, and accurate collection and processing of material status, equipment operating parameters, and transportation information. By comprehensively utilizing IoT sensors, satellite remote sensing data, and 5G technology, this invention can acquire multi-dimensional data from the material yard in real time. Through data quality processing, data fusion, and monitoring steps, it effectively improves the accuracy and reliability of the data. This innovative data collection and processing mode not only enhances the intelligence level of material yard management but also provides a solid foundation for subsequent resource allocation decisions and dynamic optimization. Furthermore, this invention designs detailed data collection and abnormal data processing schemes for different types of sensors and data characteristics, further enhancing the system's adaptability and stability.

[0043] Second, by introducing game theory and resource fairness indicators, this invention can take into account the benefits of various departments in the material yard and formulate the optimal resource allocation strategy. At the same time, by monitoring and analyzing various types of data in the material yard in real time, this invention can uncover potential data correlation patterns and dynamically adjust the resource allocation strategy and optimize the overall operation of the material yard accordingly. This resource allocation and dynamic optimization method not only improves the utilization efficiency of material yard resources, but also promotes the refinement and intelligence of material yard management. In addition, this invention also achieves the organic integration of multi-source data by constructing micro-data feature vectors and macro-data feature vectors, providing more comprehensive and accurate information support for resource allocation decisions.

[0044] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0046] Figure 1 A flowchart illustrating the information integration and intelligent decision-making methods for an integrated intelligent material yard management platform;

[0047] Figure 2 This is a framework diagram of information integration and intelligent decision-making methods for an integrated intelligent material yard management platform. Detailed Implementation

[0048] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0049] Example 1

[0050] Intelligent material yard management in steel plants.

[0051] Data Acquisition: In large ore yards of steel plants, the layout and planning of material storage areas are extremely critical. Temperature and humidity sensors are installed at different heights (top, middle, and bottom) of material stacks to accurately monitor humidity differences within the stacks and prevent ore oxidation and rusting caused by excessively high local humidity. Temperature and humidity sensors are also deployed in the corners and vents of the warehouse to monitor the overall temperature and humidity environment in real time. Composition analyzers are installed near key nodes of the material conveyor belt and multiple sampling points on the material stacks to ensure timely and accurate detection of ore composition, providing a stable raw material quality guarantee for subsequent steel smelting. Liquid level sensors are vertically installed inside liquid storage tanks used for ore washing and cooling to accurately measure the liquid level in the tanks, avoiding production disruptions due to low liquid levels or overflow risks due to excessively high liquid levels.

[0052] At the equipment operation site, the motor is the core power source for equipment operation. Vibration sensors are installed on its bearing housing and gearbox housing to monitor the parameters of vibration amplitude and frequency, and to detect potential mechanical failures in a timely manner. Speed ​​sensors are installed near the rotating shaft of the equipment to accurately obtain the operating speed of the equipment and ensure that the equipment operates within a reasonable speed range. Load sensors are installed on the hook of the crane and the support structure of the conveyor belt to monitor the load on the equipment in real time and prevent safety accidents caused by overload operation.

[0053] Regarding transport vehicles, GPS positioning devices are installed on the top or inside. Utilizing a satellite positioning system, the vehicle's location within the material yard and during transport can be tracked in real time, optimizing transport route planning. Speed ​​sensors are installed near the wheels to accurately measure vehicle speed, ensuring transport safety. Weight sensors are installed on the chassis or cargo box floor to accurately acquire vehicle load data, preventing overloading and damage to roads and vehicles. Furthermore, high-resolution satellite remote sensing image data is regularly acquired from professional satellite remote sensing data providers. Using 5G's high-speed, low-latency communication technology, data collected by various sensors and remote sensing image data are quickly and stably transmitted to the management platform, providing comprehensive data support for subsequent data processing and decision analysis. Figure 1 The specific implementation steps are shown below.

[0054] Data quality processing: After receiving data, the management platform assesses the data quality according to a pre-set data validity calculation formula, and has... Data The data validity is The calculation formula is: ,in and The boundary values ​​for the normal data fluctuation range are set based on the characteristics of the equipment. The total number of data collected. Taking ore moisture data as an example, based on the requirements of steel production processes for ore moisture, the normal fluctuation range is set as follows: If the ore moisture data collected at a certain moment is If the data points are below the normal range, and the data validity is determined to be below a set threshold, then the data quality is considered unqualified. For this type of abnormal data, an adaptive data interpolation repair formula is used. Let the number of abnormal data points be denoted as . Its adjacent normal data points are and The data change trend coefficient is The repaired data The calculation formula is: ,in, This is the repaired data. These are the original outlier data points, and the data change trend coefficient. Through the front and back The average slope of each normal data point is calculated to obtain the repaired data. Simultaneously, based on the steel plant's production plan, such as the different requirements for ore composition and moisture content for different furnace runs, and the operational needs of daily equipment maintenance plans, the data collection items and frequencies are precisely determined. For example, before blast furnace ironmaking and furnace changeover, the frequency of collecting ore composition and moisture data is increased; during equipment maintenance, the focus is on collecting operating parameters of relevant equipment. Finally, the time format of data from different sources is standardized to ensure data consistency and comparability across time dimensions, laying a solid foundation for subsequent data fusion and analysis. Figure 2 As shown.

[0055] Data Fusion and Monitoring: For various types of data collected by IoT sensors, the format is first standardized by converting the diverse data formats output by sensors from different manufacturers and types into a standard format that the platform can recognize. Then, noise reduction is performed. For high-frequency noise, a low-pass filter is used to remove high-frequency noise caused by electromagnetic interference. For low-frequency interference, a high-pass filter is used to eliminate interference signals caused by low-frequency vibration of equipment. After data preprocessing, the basic statistical characteristics of the data are calculated, and time-related features are extracted, such as the data change cycle and peak occurrence time. The time-domain data is then converted to the frequency domain to extract frequency features. Based on the actual needs of steel plant material yard management, the dimensions of the micro data feature vector are determined. For example, key features such as humidity, composition, and equipment operating parameters are arranged in an orderly manner to construct the micro data feature vector.

[0056] For satellite remote sensing image data, radiometric correction is first performed to eliminate radiometric errors caused by differences in sensor sensitivity and atmospheric scattering. Then, geometric correction is performed to correct geometric distortions in the image, ensuring the accurate position and shape of objects in the image. Next, image enhancement operations are used to highlight relevant information about the material yard, improving image clarity and readability. Using image recognition and geographic information system (GIS) technologies, different features in the remote sensing image are classified to accurately identify and locate specific targets in the material yard, such as ore piles, equipment facilities, and transportation roads. The processed image is then registered with basic data in the GIS, such as topographic data and geographic location data, to conduct spatial analysis. Multi-source geographic data is integrated to construct a material yard geographic information database. Features related to material yard management, such as the area, volume, and distribution location of ore piles, are extracted from this database and combined to form a macroscopic data feature vector.

[0057] The micro-data feature vector and the macro-data feature vector are fused using a multi-dimensional feature fusion mapping algorithm. Let the micro-data feature vector be... The macroscopic data feature vector is The fusion weight matrix is It is The matrix, elements It is the first in the matrix Line number Column element, representing the first The weights of each feature are set based on the importance and relevance of the data, resulting in the fused feature vector. The calculation formula is: This provides richer and more accurate information for comprehensive monitoring of the material yard.

[0058] Resource allocation decision-making: A steel mill's material yard involves multiple departments, including mining, transportation, and smelting. These departments have different and competing resource needs. A resource allocation model is established using game theory, assuming the material yard has... Each participating entity, such as the mining sector Transportation Department Smelting Department Each entity's resource requirement vector is determined based on its business characteristics. , Represents the number of resource types, and the revenue function is: , This refers to the number of participants in the material yard. Taking the smelting sector as an example, its resource demand vector may include multiple dimensions such as the demand for different quality ores, energy consumption, and equipment usage time. This leads to the introduction of resource equity indicators. The calculation method is as follows: ,in It is the first The average allocation of resources is calculated by surveying the needs of various departments in the material yard. Based on the actual needs of each department, the average allocation of each resource is determined. For example, after detailed research, it is determined that under the current production plan, the average allocation of a certain quality ore is x kilograms per ton. Under the constraint of meeting the basic needs of each department, an iterative algorithm is used to continuously adjust the resource allocation vector of each entity. During the adjustment process, the revenue function of each entity is taken into account. For example, the revenue of the smelting department is related to the steel output and quality, while the revenue of the mining department is related to the ore extraction volume and cost. Through multiple iterative calculations, the resource fairness index is maximized. At the same time, it takes into account the revenue functions of each subject. The optimal resource allocation scheme can be solved by using iterative algorithms. For example, during a certain production cycle, priority can be given to ensuring the supply of high-quality ore during the critical production period of the smelting department, and transportation vehicles and equipment resources can be allocated in a reasonable manner to ensure the coordinated and efficient operation of various departments and improve the production efficiency of the entire steel plant.

[0059] Dynamic optimization and regular application: By monitoring various data from the material yard in real time, relevant algorithms are used to calculate the causal correlation strength coefficient between material moisture content and equipment operating efficiency data. Assuming that N sets of material moisture content data are collected over a period of time... and equipment operating efficiency The data is The causal correlation strength coefficient is calculated using the following formula: ,in, and These are the average values ​​of humidity and equipment operating efficiency, respectively. The calculation results show... This indicates a significant correlation between material humidity (H) and equipment operating efficiency (E). For example, calculations show that when material humidity is high, the operating efficiency of ore conveying equipment decreases significantly, and for every 10% increase in humidity, equipment operating efficiency decreases by 15%. Based on this pattern, steel plants can take preventative measures, such as increasing ventilation equipment to reduce material humidity and optimizing equipment operating parameters, such as adjusting conveyor belt speed and tension to adapt to humidity changes. Simultaneously, resource allocation strategies can be dynamically adjusted based on these underlying patterns. For instance, during periods of high humidity, the use of drying equipment can be appropriately increased, and the number of transport vehicles deployed can be reduced to avoid resource waste due to decreased equipment efficiency. By continuously exploring and applying these patterns, the overall operation of the material yard can be continuously optimized, thereby improving the steel plant's production efficiency and economic benefits.

[0060] In summary, this intelligent material yard management implementation in a steel plant provides rich information for subsequent decision-making through comprehensive data collection covering material storage, equipment operation, and transportation vehicles. Advanced data quality processing and fusion technologies ensure data reliability and integrity. Game theory-based resource allocation decisions balance the interests of various departments and improve resource utilization efficiency. Regarding dynamic optimization and pattern application, by mining causal relationships between data, such as the impact of material moisture on equipment operating efficiency, proactive measures are taken to optimize operations. The entire process achieves intelligent management of the steel plant's material yard from data collection to operational optimization, effectively improving production efficiency, reducing costs, and ensuring production safety.

[0061] Example 2

[0062] Operation of intelligent coal storage yards at ports.

[0063] Data Acquisition: In the coal yard of the port, the layout of the material storage area fully considers the storage characteristics of coal. Temperature and humidity sensors are installed at different heights and positions of the coal stacks, focusing on monitoring the humidity in areas inside the stacks that are prone to spontaneous combustion. Sensors are also installed in the corners and vents of the warehouse to monitor the temperature and humidity environment inside the warehouse in real time and prevent coal spontaneous combustion accidents. Composition analyzers are installed near the coal conveyor belt and sampling points in the stacks to regularly test the ash and sulfur content quality indicators of the coal to ensure that the coal quality meets the requirements for sale and use. Liquid level sensors are installed inside the water storage tank used for coal dust suppression, installed vertically downwards, to accurately measure the water level in the tank and ensure the normal operation of dust suppression.

[0064] At the equipment operation site, vibration sensors are installed on the motor bearing housing and gearbox housing of the coal loading and unloading equipment to monitor the vibration status of the equipment and promptly detect potential equipment failures; speed sensors are installed near the rotating shaft of the equipment to accurately measure the operating speed of the equipment and ensure that the equipment operates within a safe and efficient speed range; load sensors are installed on the hook of the crane and the support structure of the conveyor belt to monitor the load on the equipment in real time and prevent the equipment from operating under overload.

[0065] Regarding transport vehicles, GPS positioning devices are installed on the top or inside to track the vehicle's driving trajectory and location information in the port in real time, facilitating dispatch and management; speed sensors are installed near the wheels to measure the vehicle's speed and ensure transportation safety; weight sensors are installed on the chassis or bottom of the cargo box to accurately obtain vehicle load data and avoid overloading. In addition, satellite remote sensing image data is acquired regularly, and the data collected by various sensors and remote sensing image data are quickly transmitted to the control platform using 5G network, providing comprehensive data support for the operation and management of the port's coal yard.

[0066] Data quality processing: After receiving the data, the management platform sets a normal fluctuation range for the data based on industry standards for coal storage and transportation and the actual operational requirements of the port. For example, for coal moisture data, the normal range is set to... Based on this, data quality is judged using the data validity calculation formula, which is: If the coal moisture data collected at a certain moment is Data exceeding the normal range, with its validity below a set threshold, is considered substandard. For this type of abnormal data, an adaptive data interpolation repair formula is used for repair. The formula is as follows: By analyzing the changing trends of k normal data points before and after, the data change trend coefficient is calculated. This process yields the corrected data. Based on the port's coal loading and unloading operation plan, ship arrival times, and market demand factors, the data collection items and frequencies are determined. For example, during periods of concentrated ship arrivals for coal loading and unloading, the frequency of data collection on coal loading and unloading equipment operating parameters and transport vehicle data is increased. During coal storage, the focus is on collecting coal temperature, humidity, and quality data. Finally, the time format of data from different sources is standardized to ensure data consistency in the time dimension, providing a reliable foundation for subsequent data fusion and analysis.

[0067] Data Fusion and Monitoring: Data collected by IoT sensors undergoes format unification and noise reduction processing. Diverse data formats from different sensors are converted into a standard format recognizable by the platform. Different filtering methods are employed to address noise in the data. For high-frequency noise, a low-pass filter is used to remove high-frequency noise caused by electrical interference. For low-frequency interference, a high-pass filter is used to eliminate interference signals such as low-frequency vibrations from equipment. Basic statistical characteristics of the data are calculated, time-related features are extracted, and the time-domain data is converted to the frequency domain to extract frequency features. Based on the actual needs of port coal yard management, the dimensions of the micro-data feature vector are determined, and key features such as coal moisture content, composition, and equipment operating parameters are arranged in an orderly manner to construct the micro-data feature vector.

[0068] For satellite remote sensing image data, radiometric correction, geometric correction, and image enhancement operations are performed sequentially. Radiometric correction eliminates radiometric errors caused by sensor differences and atmospheric effects; geometric correction corrects geometric distortions in the image, ensuring accurate object positioning; image enhancement highlights information related to the coal yard and improves image clarity. Using image recognition and geographic information system (GIS) technologies, different land features in the remote sensing image are classified to accurately identify and locate specific targets such as coal piles, loading and unloading equipment, and transportation roads. The processed images are then registered with basic data in the GIS, such as port topography and waterway information, for spatial analysis. Multi-source geographic data are integrated to construct a port coal yard geographic information database. Features related to yard management, such as the area, volume, and distribution of coal piles, are extracted from this database and combined to form a macroscopic data feature vector. A multi-dimensional feature fusion mapping algorithm is then used to fuse the microscopic and macroscopic data feature vectors. Let the microscopic data feature vector be... The macroscopic data feature vector is The fusion weight matrix is It is The matrix, elements It is the first in the matrix Line number Column element, representing the first The weights of each feature are set based on the importance and relevance of the data, resulting in the fused feature vector. The calculation formula is: This enables comprehensive and real-time monitoring of coal yards in ports.

[0069] Resource allocation decision-making: Port coal yards involve multiple departments such as loading and unloading, storage, and transportation. These departments are interconnected yet competitive in their resource needs. A resource allocation model is established using game theory, assuming the yard has... Each participating entity, such as the loading and unloading department Warehousing Department Transportation Department Each entity's resource requirement vector is determined based on its business needs; the resource requirement vector is as follows: , Represents the number of resource types, and the revenue function is: , This refers to the number of participants in the material yard. For example, the resource demand vector for the loading and unloading department might include the usage time of loading and unloading equipment, energy consumption, and manpower allocation; the resource demand vector for the warehousing department might include warehouse storage space and ventilation equipment usage; and the resource demand vector for the transportation department might include the number of transport vehicles and routes. Resource equity indicators are then introduced. The calculation method is as follows: By conducting in-depth research into the actual needs of each department, the average allocation of each type of resource is determined. Under the constraint of meeting the basic needs of each department, an iterative algorithm is used to continuously adjust the resource allocation vector of each entity. During the adjustment process, the benefit functions of each entity are taken into account. For example, the revenue of the loading and unloading department is related to loading and unloading efficiency and business volume; the revenue of the warehousing department is related to storage costs and coal turnover rate; and the revenue of the transportation department is related to transportation efficiency and transportation volume. Through multiple iterative calculations, the resource fairness index is maximized. Develop optimal resource allocation plans. For example, during peak coal loading and unloading periods, rationally allocate loading and unloading equipment and transport vehicle resources, prioritize loading and unloading efficiency, and at the same time take into account the storage needs of the warehousing department to ensure the efficient operation of the port coal yard.

[0070] Dynamic optimization and application of patterns: Real-time monitoring of various data from the port coal yard, calculation of the causal correlation strength coefficient between data, the formula is as follows: For example, calculating the correlation strength between strong winds and coal loading and unloading efficiency and dust pollution levels, if the calculation results show... This indicates the existence of significant correlations. For example, it was found that coal loading and unloading efficiency decreases by 30% and dust pollution increases by 50% during windy weather. Based on this pattern, ports can adjust their operation plans in advance, stopping loading and unloading operations before strong winds arrive, or increasing investment in dust suppression equipment, such as turning on more spray dust suppression devices to reduce dust pollution. At the same time, resource allocation strategies can be dynamically adjusted based on these potential patterns, such as reducing the allocation of transport vehicles during windy weather to avoid wasting resources due to idle vehicles caused by reduced loading and unloading efficiency. By continuously exploring and applying these patterns, the operation and management of port coal yards can be continuously optimized, thereby improving the port's economic and environmental benefits.

[0071] In summary, the port coal intelligent yard operation implementation model utilizes IoT sensors, satellite remote sensing, and 5G technology to achieve real-time data collection and transmission. Data quality processing ensures data accuracy, data fusion monitoring provides a comprehensive understanding of the yard's status, and a resource allocation model based on game theory rationally coordinates the resource needs of loading / unloading, storage, and transportation departments, improving overall port operational efficiency. In terms of dynamic optimization, by analyzing the correlation between windy weather factors and coal loading / unloading efficiency and dust pollution, operational plans and resource allocation strategies are adjusted in advance. These measures not only improve the safety and efficiency of coal loading / unloading and storage but also reduce environmental pollution, achieving intelligent and green operation of the port coal yard.

[0072] Example 3:

[0073] Intelligent material yard management in open-pit mining

[0074] Data Acquisition: In open-pit mines, the planning of material storage areas is crucial. Temperature and humidity sensors are installed at different heights and corners of ore piles to monitor the temperature and humidity of the ore, preventing ore deterioration due to excessive humidity. Composition analyzers are installed near material conveyor belts and ore pile sampling points to detect ore composition in real time, ensuring that the quality of the mined ore meets the requirements for subsequent processing. At the operation site of mining equipment, vibration sensors are installed on key parts of excavators and crushers, such as bearing housings and gearbox housings, to monitor equipment vibration and detect potential faults in a timely manner. Speed ​​sensors are installed near the rotating shafts of the equipment to measure the operating speed of the equipment, ensuring that the equipment operates under optimal conditions. GPS positioning devices are installed on the top or inside of transport vehicles to track vehicle location in real time and optimize transportation routes. Speed ​​sensors are installed near the wheels to monitor driving speed and ensure transportation safety. Weight sensors are installed on the chassis or cargo box bottom to accurately obtain load data and avoid overloading. In addition, satellite remote sensing image data is acquired regularly, and 5G technology is used to quickly transmit various sensor data and remote sensing image data to the management and control platform.

[0075] Data quality processing: After receiving the data, the management platform assesses the data quality according to pre-set standards. For abnormal data, such as ore moisture data exceeding the normal range at a certain time, the platform uses adaptive methods to repair it. Based on the mining operation plan and equipment maintenance needs, the platform determines the data collection items and frequency. During peak mining periods, the collection frequency of mining equipment operating parameters and transport vehicle data is increased; during equipment maintenance, the focus is on collecting operating data of relevant equipment. At the same time, the time format of data from different sources is standardized to facilitate subsequent analysis.

[0076] Data fusion and monitoring: Data collected by IoT sensors undergoes format unification and noise reduction processing, converting data of different formats into a platform-recognizable standard format. After removing high-frequency noise and low-frequency interference, basic statistical characteristics are calculated, time-related features are extracted, and time-domain data is converted to the frequency domain to extract frequency features. Based on the actual needs of open-pit mine management, the dimensions of micro-data feature vectors are determined, and key features such as ore moisture, composition, and equipment operating parameters are arranged in an orderly manner to construct micro-data feature vectors. For satellite remote sensing image data, radiometric correction, geometric correction, and image enhancement operations are performed to improve image quality. Like image recognition and geographic information system (GIS) technologies, different features in remote sensing images are classified to accurately identify and locate targets such as ore piles, mining equipment, and transportation roads. The processed images are then registered with basic data in the GIS to conduct spatial analysis, integrate multi-source geographic data, and construct a mining site geographic information database. Features related to material yard management, such as the area, volume, and distribution location of ore piles, are extracted from this database and combined to form macroscopic data feature vectors. Through a multi-dimensional feature fusion mapping algorithm, microscopic and macroscopic data feature vectors are fused to provide richer and more accurate information for comprehensive monitoring of the mining site.

[0077] Resource Allocation Decisions: Open-pit mines involve multiple sectors, including mining, transportation, and processing. These sectors have differing and competitive resource needs. A game theory model is used to establish a resource allocation model. Assume there are multiple participating entities in the mine, such as mining, transportation, and processing. Each entity's resource demand vector is determined based on its business characteristics. For example, the mining sector's resource demand vector might include the usage time of mining equipment, energy consumption, and manpower allocation; the transportation sector's resource demand vector includes the number of transport vehicles and routes; and the processing sector's resource demand vector includes the usage time of ore processing equipment and energy consumption. A resource fairness index is introduced, and the actual needs of each sector are investigated. The goal is to determine the average allocation of each resource. Under the constraint of meeting the basic needs of each sector, an iterative algorithm is used to continuously adjust the resource allocation vector of each entity, taking into account the revenue function of each entity. For example, the revenue of the mining sector is related to the amount and cost of ore mining, the revenue of the transportation sector is related to the transportation efficiency and the amount of transportation, and the revenue of the processing sector is related to the quality and output of processed products. Through multiple iterative calculations, the resource fairness index is maximized, and the optimal resource allocation scheme is formulated. For example, during the peak period of ore mining, mining equipment and transportation vehicle resources are rationally allocated to prioritize mining efficiency while taking into account the raw material supply needs of the processing sector, thus ensuring the efficient operation of the entire mining site.

[0078] Dynamic optimization and pattern application: Real-time monitoring of various data in open-pit mines, analysis of causal relationships between data to uncover potential patterns. For example, long-term monitoring has revealed that the operating efficiency of mining equipment decreases and the difficulty of ore transportation roads increases during rainy weather. Based on this pattern, resource allocation strategies are adjusted in advance before rainy weather, reducing investment in mining equipment, increasing anti-skid measures for transport vehicles, and rationally allocating personnel. At the same time, based on these potential patterns, the overall operation of the mine is continuously optimized to improve the mine's production efficiency and economic benefits, and ensure safe production.

[0079] In summary, the information integration and intelligent decision-making methods of the intelligent material yard integrated management and control platform, leveraging IoT sensors, satellite remote sensing, and 5G technology, achieve real-time multi-dimensional data collection and transmission. Through data quality processing, fusion, and monitoring, the accuracy and reliability of data are improved. Game theory is used to establish a resource allocation model, balancing the interests of various departments and optimizing resource allocation. By mining data patterns and dynamically adjusting strategies, continuous optimization of material yard operations is achieved. In applications in steel plants, port coal yards, and open-pit mines, this method effectively improves production efficiency, reduces costs, ensures safety, and reduces pollution, promoting the development of material yard management towards intelligence and refinement. It has significant application value and promotional significance.

[0080] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An information integration and intelligent decision method of an intelligent stockyard integrated management and control platform, characterized in that, The specific steps of the decision-making method are: S100, data collection: deploying Internet of Things sensors in the material storage area of the stockyard, the equipment operation site, and the transport vehicle, collecting real-time material status, equipment operation parameters, and transportation information, regularly obtaining satellite remote sensing image data through the data interface of a satellite remote sensing data provider, and transmitting these data to the management and control platform using 5G technology; S200, data quality processing: the data quality is judged by using data validity in the management and control platform, and there are data , the data validity is , and the calculation formula is: , wherein and are the normal data fluctuation range boundary values set according to the equipment characteristics, is the total number of collected data, , the data quality is qualified, otherwise the data quality is unqualified, and are the set data quality judgment thresholds, for the data with poor quality, the adaptive data interpolation repair formula is used for repair, and the project and frequency of data collection are determined according to the collection plan and business needs, and the time formats of different source data are unified; S300, data fusion and monitoring: formatting and noise reduction processing of the data collected by the Internet of Things sensors, constructing a micro data feature vector, processing the satellite remote sensing image data using image recognition and geographic information system technology to construct a macro data feature vector, and performing fusion through a multi-dimensional feature fusion mapping algorithm; S400, resource allocation decision: establishing a resource allocation model using game theory based on the requirements of each department of the stockyard, introducing a resource fairness index to quantify fairness, adjusting the resource allocation scheme using an iterative algorithm, and formulating an optimal resource allocation strategy; S500, dynamic optimization and rule application: real-time monitoring of various data of the stockyard, analysis of the causal relationship between the data to mine potential rules, dynamic adjustment of the resource allocation strategy based on the potential rules, and optimization of the overall operation of the stockyard according to the potential rules. 2.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The sensors used in each location in the S100, data collection, are: Material storage area: Temperature and humidity sensor: installed at different heights of the material stack, corners of the warehouse, and ventilation openings; Composition analyzer: installed on the material conveyor belt or near the sampling point of the material stack; Liquid level sensor: installed inside the liquid storage tank, vertically from the top of the tank; Equipment operation site: Vibration sensor: installed on the bearing seat of the motor and the housing of the gear box; Speed sensor: Installation location: installed near the rotating shaft of the equipment; Load sensor: installed on the hook of the crane and the support structure of the conveyor belt; Transport vehicle: GPS positioning device: installed on the top or inside of the vehicle; Speed sensor: installed near the wheels of the vehicle; Weight sensor: installed on the chassis or the bottom of the cargo box of the vehicle. 3.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The S200, the abnormal data in the data quality processing is repaired by using the adaptive data interpolation repair formula, assuming that the existing abnormal data point is , the adjacent normal data point is , , the data change trend coefficient is , the repaired data is , and the calculation formula is: , wherein is the repaired data, is the original abnormal data point, the data change trend coefficient is calculated by the average of the slopes of the front and rear normal data points. 4.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The specific steps for constructing the micro data feature vector in S300, data fusion and monitoring, are: (1) Determine the data standard format, parse the raw data of different sensors, and convert them to a unified format; (2) Identify noise in the data, filter high-frequency noise using a low-pass filter, and process low-frequency interference using a high-pass filter; (3) Calculate the basic statistical features of the data, extract time-related features, convert time-domain data to frequency domain, and extract frequency features; (4) Determine the dimensions based on management requirements, arrange the extracted features in order to construct the micro data feature vector. 5.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The specific steps for constructing the macro data feature vector in S300, data fusion and monitoring, are: (1) Perform radiation correction, geometric correction, and image enhancement operations on the satellite remote sensing image data; (2) Classify different ground objects in the remote sensing image, and accurately identify and locate specific targets in the stockyard; (3) Register the processed image with the basic data in the geographic information system, conduct spatial analysis, integrate multi-source geographic data, and construct a stockyard geographic information database; (4) Screen the features related to stockpile management and combine them to form a macro data feature vector. 6.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The S300, data fusion and monitoring through multi-dimensional feature fusion mapping algorithm micro data feature vector and macro data feature vector fusion, set micro data feature vector is , macro data feature vector is , fusion weight matrix is , a The element is the first Row first Column element, indicating the first Weight of the feature, according to the importance and correlation of data setting, the feature vector The calculation formula is: . 7.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The S400 process utilizes game theory to establish a resource allocation model for intelligent material yard resource allocation, assuming the material yard contains... Participating entities The resource demand vector for each entity is: , Represents the number of resource types, and the revenue function is: , It refers to the number of participants in the material yard, introducing a resource equity index. The calculation method is as follows: ,in It is the first The average allocation of various resources is calculated by surveying the needs of each department in the material yard, and the resource allocation vectors of each entity are continuously adjusted. To maximize the resource equity index while meeting the basic needs of all stakeholders. At the same time, it takes into account the revenue functions of each subject. The optimal resource allocation scheme is solved by using an iterative algorithm. 8.The information integration and intelligent decision method of the intelligent stockyard integrated management and control platform according to claim 1, characterized in that, The S500, dynamic optimization and regular application of the calculation of the strength coefficient of causal correlation, the material humidity is and the equipment operation efficiency is , the group data collected in a period of time is , the group data is , , the group number of collected data, the strength coefficient of causal correlation The calculation formula is: , wherein and are the average values of humidity and equipment operation efficiency, , indicating that there is a significant correlation between the material humidity and the equipment operation efficiency , and The greater the causal correlation is, the stronger the causal correlation is, , it is considered that there is no significant correlation between the two, wherein is the threshold value of the strength of causal correlation determined by statistical significance test.

Images