A port operation digital sand table construction method and system based on panoramic three-dimensional modeling

Abstract

This invention relates to the field of digital sand table technology, providing a method and system for constructing a port operation digital sand table based on panoramic 3D modeling. The method includes constructing a panoramic 3D model of the port yard area, assigning a unique storage unit identifier to each 3D entity in the panoramic 3D model; configuring a data synchronization trigger mechanism to obtain multi-dimensional storage business data for the corresponding storage unit from the yard management platform; performing fusion inference on the multi-dimensional storage business data based on a preset uncertainty inference model, outputting a quantitative assessment value of the storage status and its confidence information; matching the corresponding 3D entity through the storage unit identifier, dynamically adjusting the visualization rendering attributes of the matched 3D entity according to the quantitative assessment value of the storage status and its confidence information, and rendering the dynamic changes of the yard storage status in real time in the panoramic 3D model. This invention achieves the perception and suppression of data quality uncertainty in the port operation digital sand table, improving the evaluation robustness of dynamic monitoring of storage status.

Description

Technical Field

[0001] This invention relates to the field of digital sand table technology, and more specifically, to a method and system for constructing a port operation digital sand table based on panoramic 3D modeling. Background Technology

[0002] In the process of digitalizing port operations, digital sand tables based on panoramic 3D modeling have become a key technical means to improve the port's refined scheduling and operation management capabilities. By constructing 3D models of elements such as storage yards and equipment, and combining them with real-time data-driven approaches, they enable dynamic and visual monitoring of ship loading and unloading, storage yard inventory, and equipment status. Their core value lies in the deep integration of discrete multi-source business data with 3D spatial scenes, providing managers with intuitive and real-time operational situation awareness capabilities.

[0003] However, existing digital sand table systems typically employ fixed-weighted averaging or simple rule-based fusion methods when performing data fusion. When conflicts arise between multiple data sources—including planned data from the yard management system, real-time data collected by IoT sensors, and verification data entered by tally personnel—due to sensor drift, transmission loss, or human error, existing solutions cannot effectively identify the degree of conflict or dynamically suppress the interference of abnormal data on the assessment results during the fusion process. This leads to easily contaminated storage status assessment results, a lack of quantitative representation of assessment reliability, and an inability to meet the stringent requirements of port production for the accuracy of monitoring data and the credibility of decision-making. Therefore, how to achieve the perception and suppression of data quality uncertainties in port operation digital sand tables, thereby improving the robustness of dynamic monitoring of storage status assessments, has become a challenge for the industry. Summary of the Invention

[0004] This invention provides a method and system for constructing a digital sand table for port operations based on panoramic 3D modeling. It can realize the perception and suppression of data quality uncertainty in the digital sand table for port operations, thereby improving the evaluation robustness of dynamic monitoring of storage status.

[0005] In a first aspect, the present invention provides a method for constructing a digital sand table for port operations based on panoramic 3D modeling, the method comprising the following steps: Construct a panoramic 3D model of the port yard area. The panoramic 3D model includes 3D entities of yard stacks and storage equipment. Assign a unique storage unit identifier to each 3D entity. Configure a data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the storage yard management platform according to a preset time period or operation event triggering method; Based on a preset uncertainty reasoning model, fusion reasoning is performed on the multidimensional heap storage business data to output a quantitative evaluation value of the heap storage status and its confidence information. The uncertainty reasoning model is used to handle the data quality uncertainty that exists in the process of collection, transmission and fusion of multidimensional heap storage business data. The corresponding three-dimensional entity is matched by the stacking unit identifier. The visualization rendering attributes of the matched three-dimensional entity are dynamically adjusted according to the quantitative evaluation value of the stacking status and its confidence information. The dynamic changes of the stacking status of the yard are rendered in real time in the panoramic three-dimensional model.

[0006] Furthermore, constructing a panoramic 3D model of the port yard area specifically includes: Obtain geospatial data and yard facility parameters for the port storage area; A base topography model of the storage yard is constructed based on the geospatial data and storage yard facility parameters; Obtain the three-dimensional geometric parameters of the stockpile positions and the structural parameters of the storage equipment; Based on the aforementioned yard base terrain model, a three-dimensional entity of the stacking site is constructed according to the aforementioned three-dimensional geometric parameters, and a three-dimensional entity of the equipment is constructed according to the aforementioned structural parameters; Each of the aforementioned stacking position 3D entities and equipment 3D entities is assigned a unique stacking unit identifier, and the stacking unit identifier is associated with the corresponding 3D entity for storage to form a panoramic 3D model.

[0007] Furthermore, a data synchronization triggering mechanism is configured. This mechanism, based on a preset time period or operational event triggering method, retrieves multi-dimensional storage business data for the corresponding storage unit from the yard management platform. Specifically, this includes: Configure a scheduled synchronization task to trigger data synchronization operations periodically at preset time intervals; Configure an event listening interface to capture the storage operation event in real time and trigger data synchronization operation when a storage operation event occurs on the yard management platform; In response to data synchronization, a data acquisition request is initiated to the yard management platform based on the storage unit identifier, and multi-dimensional storage business data is received from the yard management platform.

[0008] Furthermore, the multidimensional storage business data includes the attribute information of the stored items, the quantity information of the stored items, the warehousing time information of the stored items, the outbound plan information of the stored items, the turnover frequency information of the stored items, and the quality inspection information of the stored items.

[0009] Furthermore, based on a preset uncertainty reasoning model, fusion reasoning is performed on the multidimensional heap storage business data to output a quantitative evaluation value of the heap storage status and its confidence information, specifically including: The multidimensional stacked business data is grouped according to data source and data type, and then input into a preset uncertainty inference model. The uncertainty inference model includes a data quality assessment sub-model for different dimensions and a multi-source data fusion sub-model. The credibility weights of data in each dimension are determined through the data quality assessment sub-model. The data of each dimension and its confidence weight are input into the multi-source data fusion sub-model, and fusion reasoning based on evidence theory is performed to output the quantitative evaluation value of the heap status and its confidence information.

[0010] Furthermore, the data of each dimension and its confidence weight are input into the multi-source data fusion sub-model, and fusion reasoning based on evidence theory is performed to output the quantitative evaluation value of the heap state and its confidence information, specifically including: Identify the degree of evidence conflict between different dimensions of data in determining the stacking status, and trigger the corresponding fusion strategy based on the degree of evidence conflict; When there is a conflict of evidence, the weights of the data in each dimension participating in the fusion are adjusted according to the credibility weights of each dimension of data and the degree of conflict of evidence, so as to suppress the interference of abnormal data introduced by collection errors or transmission packet loss on the fusion results. Based on the corrected weights, the data of each dimension is fused and inferred, and the quantitative evaluation value of the heap status and its confidence information are output, so that the uncertainty inference model can maintain the ability to suppress the uncertainty of data quality throughout the entire chain of data collection, transmission and fusion.

[0011] Furthermore, matching the corresponding three-dimensional entity through the stacking unit identifier specifically includes: Obtain the stacking unit identifier corresponding to the 3D entity to be updated and rendered, and match the stacking unit identifier with the stacking unit identifier of each 3D entity stored in the panoramic 3D model one by one. When a match is successful, the three-dimensional entity node corresponding to the heap unit identifier is located, and the visualization attribute configuration interface of the three-dimensional entity node is obtained.

[0012] Furthermore, dynamically adjusting the visualization rendering attributes of the matched 3D entity based on the quantitative evaluation value of the heap state and its confidence information specifically includes: The quantitative evaluation value of the stacking status is mapped to the first rendering channel, which is used to characterize the stacking occupancy degree of the stacked items and reflects the changing trend of the stacking status through color gradient. The confidence information is mapped to the second rendering channel, which is used to characterize the reliability of the heap state evaluation result and reflects the difference in confidence level through transparency gradient or texture density. The rendering parameters of the three-dimensional entity are generated based on the mapping results, wherein the color value of the three-dimensional entity is determined based on the mapping results of the first rendering channel, and the transparency or texture style of the three-dimensional entity is determined based on the mapping results of the second rendering channel. The rendering parameters are sent to the rendering engine to complete the adjustment of the rendering attributes for 3D entity visualization.

[0013] Furthermore, the real-time rendering of the dynamic changes in the storage status of the stockpile in the panoramic 3D model refers to the real-time refresh and display of the 3D entities with dynamically adjusted rendering attributes in the panoramic 3D model, so as to present the dynamic evolution process of changes in the storage height, storage area occupancy, and storage status confidence of the stockpile in a visual manner.

[0014] Secondly, the present invention provides a port operation digital sand table construction system based on panoramic 3D modeling, the system comprising: The model building module is used to build a panoramic 3D model of the port yard area. The panoramic 3D model includes 3D entities of yard stacks and storage equipment, and assigns a unique storage unit identifier to each 3D entity. The data synchronization module is used to configure the data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the storage yard management platform according to a preset time period or operation event triggering method. The fusion reasoning module is used to perform fusion reasoning on the multidimensional heap storage business data based on a preset uncertainty reasoning model, and output the heap storage status quantitative evaluation value and its confidence information. The uncertainty reasoning model is used to handle the data quality uncertainty that exists in the process of collection, transmission and fusion of multidimensional heap storage business data. The visualization rendering module is used to match the corresponding three-dimensional entity through the stacking unit identifier, dynamically adjust the visualization rendering attributes of the matched three-dimensional entity according to the quantitative evaluation value of the stacking status and its confidence information, and render and present the dynamic changes of the stacking status in the panoramic three-dimensional model in real time.

[0015] Thirdly, the present invention provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the above-described method for constructing a digital sand table for port operations based on panoramic 3D modeling.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for constructing a digital sand table for port operations based on panoramic 3D modeling.

[0017] The technical solutions provided by the embodiments disclosed in this invention have the following beneficial effects: This invention constructs a panoramic 3D model of the port yard area, assigns a unique storage unit identifier to each 3D entity, and establishes a precise correlation between 3D space and business data. It configures a dual-mode data synchronization triggering mechanism (timed and event-based) to ensure continuous updates of multi-source storage business data and real-time response during critical operations. Based on a preset uncertainty reasoning model, it performs fusion reasoning on multi-source data, triggers differentiated fusion strategies by identifying the degree of evidence conflict, and dynamically adjusts the weights of data participating in the fusion based on credibility weights and conflict levels. This suppresses abnormal data interference introduced by acquisition errors and transmission packet loss, and outputs a quantitative evaluation value of the storage status and its confidence information. Furthermore, by matching the storage unit identifier with the corresponding 3D entity, the quantitative evaluation value of the storage status is mapped to the first rendering channel to represent the storage occupancy level with color gradients, and the confidence information is mapped to the second rendering channel to represent the reliability of the evaluation result with transparency gradients or texture density. Dual-channel rendering parameters are generated and sent to the rendering engine to complete the dynamic adjustment of the 3D entity visualization rendering attributes.

[0018] In summary, this invention achieves the perception and suppression of data quality uncertainty in dynamic monitoring of heap status through the synergistic effect of panoramic modeling, dual-mode synchronization, conflict-aware fusion, and dual-channel visualization rendering. It transforms the traditional static fusion paradigm into a dynamic adaptive mechanism with conflict-aware capabilities, significantly improving the robustness of heap status assessment and the reliability of monitoring. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the working process of a port operation digital sand table construction system based on panoramic 3D modeling, according to some embodiments of the present invention. Figure 2 This is an exemplary flowchart of a method for constructing a digital sand table for port operations based on panoramic 3D modeling, according to some embodiments of the present invention. Figure 3 This is a structural schematic diagram of a port operation digital sand table construction system based on panoramic 3D modeling, according to some embodiments of the present invention; Figure 4 This is a schematic diagram of the structure of a computer device for implementing a method for constructing a digital sand table for port operations based on panoramic 3D modeling, as shown in some embodiments of the present invention. Detailed Implementation

[0020] To better understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] refer to Figure 1As shown in the figure, this is a flowchart illustrating the working process of the port operation digital sand table construction system based on panoramic 3D modeling in some embodiments of the present invention. The figure shows the working process of the port operation digital sand table construction system. In port operation scenario 110, physical elements such as yard stacks and storage equipment undergo continuous loading and unloading operations and changes in storage status. Their spatial location and status information are collected by data acquisition device 120. This data acquisition device 120 is used to acquire multi-dimensional storage business data, including planned data from the yard management system, real-time height and volume data from IoT sensors, and verification data entered by tally personnel, in order to meet the needs of multi-source data fusion. Then, the collected multi-dimensional storage business data is transmitted to the cloud server 130. This server performs fusion reasoning processing based on an uncertainty reasoning model. By identifying the degree of evidence conflict and dynamically correcting the data weights, it outputs a quantitative assessment value of the storage status and its confidence information. Finally, the visualization terminal 140 displays the processed storage status assessment results intuitively. Through a dual-channel visualization rendering method in the panoramic 3D model, which uses color gradients to reflect the storage occupancy level and transparency gradients or texture density to reflect the differences in confidence level, the entire workflow of the system is completed, thereby providing visualization assistance for port yard operation monitoring.

[0022] refer to Figure 2 The figure is an exemplary flowchart of a method for constructing a digital sand table for port operations based on panoramic 3D modeling, according to some embodiments of the present invention. This method mainly includes the following steps: Step 101: Construct a panoramic 3D model of the port yard area. The panoramic 3D model includes 3D entities of yard stacks and storage equipment, and assigns a unique storage unit identifier to each 3D entity.

[0023] It should be noted that the embodiments of the present invention will be specifically introduced in conjunction with the operational digital sand table scenario in the construction of the port group's digital management and control platform. This scenario restores the spatial layout of the port area's yard, equipment, and operational environment through panoramic 3D modeling, integrates multi-source business data such as ship dynamics, equipment operating status, and yard inventory, and processes the data quality uncertainty during data collection, transmission, and fusion based on an uncertainty reasoning model, outputting a quantitative assessment value of the storage status and its confidence information. Then, through dual rendering channels, the storage occupancy level and the reliability of the assessment results are mapped respectively, realizing the dynamic adjustment of the 3D entity visualization rendering attributes, so as to help understand the practical application of the technical solution of the present invention in port yard operation monitoring.

[0024] In some embodiments, constructing a panoramic 3D model of a port yard area can be achieved using the following steps: Obtain geospatial data and yard facility parameters for the port storage area; A base topography model of the storage yard is constructed based on the geospatial data and storage yard facility parameters; Obtain the three-dimensional geometric parameters of the stockpile positions and the structural parameters of the storage equipment; Based on the aforementioned yard base terrain model, a three-dimensional entity of the stacking site is constructed according to the aforementioned three-dimensional geometric parameters, and a three-dimensional entity of the equipment is constructed according to the aforementioned structural parameters; Each of the aforementioned stacking position 3D entities and equipment 3D entities is assigned a unique stacking unit identifier, and the stacking unit identifier is associated with the corresponding 3D entity for storage to form a panoramic 3D model.

[0025] It should be noted that the yard base terrain model in this invention is a three-dimensional basic model used to reflect the topography and spatial layout of the port yard area and to serve as a reference for the positioning of stacking positions and equipment entities; the storage unit identifier is an identity code used to uniquely identify the three-dimensional entity of the yard stacking position or the three-dimensional entity of the storage equipment and to establish its association with the storage business data; the panoramic three-dimensional model is a three-dimensional scene model used to fully present the topography, stacking position and equipment spatial layout of the port yard area and to serve as a carrier for the visualization and monitoring of the storage status and the basis for data association.

[0026] In practical application, firstly, geospatial data and yard facility parameters of the port yard area are acquired. The geospatial data is collected from topographic map data, digital elevation model data, and satellite imagery data from surveying and mapping departments or geographic information system platforms. The yard facility parameters are extracted from the functional zoning boundaries, main road layout, and supporting facility location information of the port yard design drawings and engineering documents. The acquired geospatial data and yard facility parameters are used as the basic input data for constructing the yard base topographic model. Secondly, based on the topographic map data and digital elevation model data in the geospatial data, a topographic grid model reflecting the topographic undulations and surface morphology of the yard area is generated using 3D modeling software. Then, according to the functional zoning boundaries and road layout information in the yard facility parameters, the functional zoning areas and road network wireframes of the yard area are superimposed on the topographic grid model to form a yard base topographic model that includes the topographic base and facility layout. Then, the three-dimensional geometric parameters of the storage yard stacks and the structural parameters of the storage equipment are obtained. The three-dimensional geometric parameters of the storage yard stacks are obtained by extracting the length, width, and height dimensions, the top plane coordinates, and the slope angle of each stack based on the design dimensions and layout diagram of the stacks in the storage yard design drawings. The structural parameters of the storage equipment are obtained by extracting the length, width, and height dimensions of the main body of the equipment, the relative positional relationships of each component, and the coordinate positioning information of the equipment mounting base based on the product manual and the equipment's three-dimensional design drawings. The obtained three-dimensional geometric and structural parameters are used as the morphological basis for constructing the three-dimensional entities of the stacks and the equipment. Afterwards, the storage yard base terrain model is used as an empty... The spatial positioning reference is determined by using the coordinates of the top plane of the stack as recorded in the three-dimensional geometric parameters of the stack position in the storage yard. A three-dimensional entity of the stack position with a realistic geometric shape is generated according to the length, width, height and slope angle of the stack position. All the generated three-dimensional entities of the stack position are superimposed on the storage yard base terrain model according to their spatial position coordinates. At the same time, the spatial position of each piece of equipment is determined by using the coordinates of the installation base recorded in the structural parameters of the storage equipment. A three-dimensional entity of the equipment with a realistic shape structure is generated according to the length, width, height and relative positional relationship of the various components of the equipment. All the generated three-dimensional entities of the equipment are superimposed on the storage yard base terrain model according to their spatial position coordinates. Finally, after generating the 3D entity of each stack and the 3D entity of the equipment, a unique stacking unit identifier is generated for each 3D entity according to the preset coding rules. The stacking unit identifier is composed of the yard area code, the entity type code and the sequence number. The stacking unit identifier is used as an attribute field to be associated and bound with the corresponding 3D entity. All stacking unit 3D entities and equipment 3D entities after binding the stacking unit identifier are stored together with the yard base terrain model in the 3D scene file to obtain a panoramic 3D model for subsequent storage status visualization monitoring.

[0027] Step 102: Configure a data synchronization triggering mechanism. The data synchronization triggering mechanism obtains multi-dimensional storage business data of the corresponding storage unit from the storage yard management platform according to a preset time period or operation event triggering method.

[0028] It should be noted that the multidimensional storage business data in this invention specifically includes: the attribute information of the stored items, the quantity information of the stored items, the warehousing time information of the stored items, the outbound plan information of the stored items, the turnover frequency information of the stored items, and the quality inspection information of the stored items.

[0029] In some embodiments, configuring a data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the yard management platform according to a preset time period or operation event triggering method, can be implemented by the following steps: Configure a scheduled synchronization task to trigger data synchronization operations periodically at preset time intervals; Configure an event listening interface to capture the storage operation event in real time and trigger data synchronization operation when a storage operation event occurs on the yard management platform; In response to data synchronization, a data acquisition request is initiated to the yard management platform based on the storage unit identifier, and multi-dimensional storage business data is received from the yard management platform.

[0030] In practical applications, firstly, when configuring the data synchronization trigger mechanism, a timed synchronization task is configured. The specific implementation method is to set up a timed scheduler in the data synchronization service, configure the timed trigger rules according to the preset time interval parameters, bind the timed trigger rules with the data synchronization operation, so that the scheduler automatically calls the data synchronization operation when each time interval arrives, forming a data synchronization mechanism that is executed periodically according to the preset time interval, and using the timed synchronization task as the basic trigger method to ensure continuous data updates. Secondly, an event listening interface is configured. When a storage operation event occurs on the yard management platform, the operation event is captured in real time and a data synchronization operation is triggered. Specifically, an event subscription channel is established between the yard management platform and the data synchronization service. A listening interface is registered in the event subscription channel. When a storage operation event such as storage material entry, exit, transfer, or inventory occurs on the yard management platform, the yard management platform pushes the operation event message to the event subscription channel. The listening interface captures the pushed operation event message in real time and parses the event type. According to the event type, the corresponding data synchronization operation is called to form a real-time data synchronization mechanism driven by operation events. The event listening interface is used as a supplementary triggering method to ensure the real-time updating of key operation data. Then, in response to data synchronization triggering, a data acquisition request is initiated to the yard management platform based on the storage unit identifier, and the multidimensional storage business data returned by the yard management platform is received. Specifically, when a timed synchronization task is triggered or an event listening interface triggers a data synchronization operation, the data synchronization service obtains the storage unit identifier corresponding to the storage unit to be synchronized, constructs a data acquisition request based on the storage unit identifier, and sends the data acquisition request to the data service interface of the yard management platform. The yard management platform queries the corresponding multidimensional storage business data from the business database based on the storage unit identifier in the data acquisition request and returns it. The data synchronization service receives the returned multidimensional storage business data and temporarily stores it in the data cache area, using the received multidimensional storage business data as the input data source for the uncertain reasoning model.

[0031] In other embodiments, a data synchronization triggering mechanism is configured. This mechanism obtains multi-dimensional storage business data of the corresponding storage unit from the yard management platform according to a preset time period or a job event triggering method. The priority of job event triggering is higher than that of time period triggering. In specific applications, when both timed synchronization tasks and event listening interfaces are configured, when the event listening interface captures a storage job event, the system first determines whether there is a timed synchronization task waiting to be executed. If so, the timed synchronization task is suspended or canceled, and the data synchronization operation triggered by the job event is executed first. The scheduling of the timed synchronization task is resumed after the data synchronization operation triggered by the job event is completed, so as to ensure the real-time data of critical job moments.

[0032] Step 103: Based on the preset uncertainty reasoning model, perform fusion reasoning on the multidimensional heap storage business data, and output the heap storage status quantitative evaluation value and its confidence information. The uncertainty reasoning model is used to handle the data quality uncertainty that exists in the process of collection, transmission and fusion of multidimensional heap storage business data.

[0033] It is important to note that in port yard operation monitoring scenarios, accurate perception of storage status relies on the fusion of multi-source data, including planned data from the yard management system, real-time height and volume data collected by IoT sensors, verification data entered by tally personnel, and equipment operating status data. However, due to sensor measurement drift, data transmission loss and delay, subjective bias in manual data entry, and frequent logical conflicts between different data sources in determining storage status, a single data source cannot accurately reflect the storage status. Existing technologies for multi-source data fusion typically employ fixed-weighted averaging or simple rule-based fusion methods. When evidence conflicts arise between different data sources due to acquisition errors, transmission loss, or sensor drift, existing solutions cannot effectively identify the degree of conflict, nor can they dynamically suppress interference from abnormal data during the fusion process. This makes the fusion results susceptible to contamination by abnormal data, resulting in insufficient reliability and robustness in storage status assessment, and failing to meet the requirements of refined management for quantifiable and traceable assessment results.

[0034] In view of this, the present invention quantifies the degree of evidence conflict between data of different dimensions by introducing a conflict detection mechanism, and triggers differentiated fusion strategies based on the degree of conflict. When evidence conflict exists, the weights are dynamically adjusted based on the credibility weights and conflict degree of each dimension of data. This effectively reduces or even isolates the weights of data sources with excessive deviation or low credibility during fusion, transforming the traditional static fusion paradigm into a dynamic adaptive fusion mechanism with conflict perception capabilities. This solves the problem of fusion result distortion caused by data quality uncertainty during the collection, transmission and fusion of multi-source data. It achieves active suppression of abnormal data and quantitative controllability of fusion reliability, significantly improving the accuracy and robustness of heap status assessment.

[0035] In some embodiments, the following steps are used to perform fusion inference on the multidimensional heap storage business data based on a preset uncertainty inference model, and output the quantitative evaluation value of the heap storage status and its confidence information: The multidimensional stacked business data is grouped according to data source and data type, and then input into a preset uncertainty inference model. The uncertainty inference model includes a data quality assessment sub-model for different dimensions and a multi-source data fusion sub-model. The credibility weights of data in each dimension are determined through the data quality assessment sub-model. The data of each dimension and its confidence weight are input into the multi-source data fusion sub-model, and fusion reasoning based on evidence theory is performed to output the quantitative evaluation value of the heap status and its confidence information.

[0036] It should be noted that the data quality assessment sub-model in this invention is an assessment model used to quantitatively evaluate the accuracy, completeness, timeliness, and consistency of data sources in various dimensions and output credibility weights; credibility weights are quantitative coefficients used to characterize the reliability and reference value of data in each dimension in fusion inference; the heap status quantitative assessment value is a comprehensive assessment value used to quantitatively describe the current heap quantity, heap height, or heap occupancy level of the heap unit; confidence information is an uncertainty measure index used to characterize the reliability of the heap status quantitative assessment results.

[0037] In practical applications, after receiving multidimensional storage business data, the data synchronization service first parses the data source identifier and data type identifier carried in the data. The data source identifier is used to distinguish that the data comes from different channels such as the storage yard management system, IoT platform, handheld terminal input, or manual entry. The data type identifier is used to distinguish that the data belongs to different categories such as quantity information, weight information, volume information, height information, or status information of the stored items. After parsing, the data is grouped according to the two dimensions of data source and data type. Data from the same source and of the same type are grouped together. The grouped data is then fed into the uncertainty inference model as input. The uncertainty inference model contains data quality assessment sub-models for different dimensions. This model is specifically designed to handle data from specific sources or of specific types. For example, in the monitoring scenario of iron ore storage units in a port yard, the planned storage weight data of the storage unit is obtained from the yard management system, the real-time storage height data collected by the radar level gauge and the real-time storage volume data collected by the laser scanner are obtained from the IoT platform, and the manually verified storage quantity data entered by the on-site tally clerk is obtained from the handheld terminal. After grouping according to the data source and data type, the planned storage weight data is input into the quality assessment sub-model that specifically processes management system data, the radar level gauge height data and the laser scanner volume data are input into the quality assessment sub-model that specifically processes IoT sensor data, and the manually verified quantity data is input into the quality assessment sub-model that specifically processes manually entered data. Secondly, the credibility weights of each dimension of data are determined through a data quality assessment sub-model. Specifically, each sub-model assesses the data across four dimensions—accuracy, completeness, timeliness, and consistency—based on the characteristics of the data dimension it processes. Accuracy is assessed based on the historical accuracy of the data source, sensor calibration status, or manual data entry and verification records. Completeness is assessed based on the completeness of data fields and the presence of missing values. Timeliness is assessed based on the interval between the data collection time and the current time; a longer interval indicates lower timeliness. Consistency is assessed based on the logical matching degree between the data and other dimensions of data within the same storage unit; significant contradictions reduce consistency. The data quality assessment sub-model then weights and synthesizes the assessment results of the four dimensions, outputting a credibility weight value between 0 and 1. A higher value indicates higher reference value for that dimension of data in subsequent fusion inference.

[0038] Then, the data of each dimension and its credibility weights are input into the multi-source data fusion sub-model, and fusion inference based on evidence theory is performed to output the quantitative evaluation value of the stockpile status and its confidence information. The specific implementation is as follows: After receiving the data of each dimension and its corresponding credibility weights, the multi-source data fusion sub-model first converts the data of each dimension into a unified stockpile quantity representation form according to the preset conversion rules. For example, the stockpile height combined with the bottom area of ​​the stack is converted into the stockpile volume, and the stockpile volume combined with the density of the stockpile is converted into the stockpile weight, so that all dimension data are comparable under the same dimension. Then, according to evidence theory, each dimension data is treated as an independent source of evidence, and its credibility weight is used as the basis for the basic probability allocation of the source of evidence. Multiple sources of evidence are synthesized, and the comprehensive confidence of each candidate stockpile quantity level or stockpile quantity value is calculated. The stockpile quantity value with the highest comprehensive confidence is selected as the quantitative evaluation value of the stockpile status, and the confidence corresponding to the evaluation value is output as the confidence information. This is only a brief explanation here, and a detailed explanation will be given below.

[0039] Preferably, in some embodiments, the data of each dimension and its confidence weight are input into the multi-source data fusion sub-model, fusion inference based on evidence theory is performed, and the heap state quantitative evaluation value and its confidence information are output, which is achieved by the following steps: Identify the degree of evidence conflict between different dimensions of data in determining the stacking status, and trigger the corresponding fusion strategy based on the degree of evidence conflict; When there is a conflict of evidence, the weights of the data in each dimension participating in the fusion are adjusted according to the credibility weights of each dimension of data and the degree of conflict of evidence, so as to suppress the interference of abnormal data introduced by collection errors or transmission packet loss on the fusion results. Based on the corrected weights, the data of each dimension is fused and inferred, and the quantitative evaluation value of the heap status and its confidence information are output, so that the uncertainty inference model can maintain the ability to suppress the uncertainty of data quality throughout the entire chain of data collection, transmission and fusion.

[0040] It should be noted that the degree of evidence conflict in this invention is a consistency metric used to quantify the degree of logical contradiction in the determination of the stacking state of data from different dimensions; the abnormal data in this invention is used to characterize distorted data that deviates significantly from the actual stacking state due to acquisition errors or transmission packet loss.

[0041] In practical application, firstly, the degree of evidence conflict between different dimensions of data in determining the stacking state is identified, and a corresponding fusion strategy is triggered based on the degree of evidence conflict. Specifically, after receiving data from each dimension, the multi-source data fusion sub-model first makes a preliminary judgment on the stacking state reflected by each dimension, calculates the degree of difference between the judgment results of each dimension, and quantifies the degree of evidence conflict as the degree of evidence conflict. Specifically, the calculation of the degree of evidence conflict can adopt an evaluation method based on data deviation rate. The data with the highest credibility weight in each dimension is used as a reference benchmark, and the relative deviation of other dimension data from this benchmark data in terms of stacking quantity is calculated. The weighted average of all deviation values ​​is taken as the degree of evidence conflict. When the degree of evidence conflict is lower than a first preset threshold, it is judged as a low-conflict state, triggering a direct weighted fusion strategy, i.e., direct fusion without weight correction; when the degree of evidence conflict is between the first preset threshold and the second preset threshold, it is considered a low-conflict state. When the threshold is between a certain threshold, it is determined to be a medium conflict state, triggering a weight correction fusion strategy, which means that the weights of the data in each dimension are moderately adjusted before fusion. When the degree of evidence conflict is higher than the second preset threshold, it is determined to be a high conflict state, triggering an anomaly isolation fusion strategy, which means that the data source with the highest degree of conflict is isolated or significantly reduced in weight before fusion. In this way, the degree of evidence conflict is used as the decision basis for fusion strategy selection, realizing adaptive switching of fusion strategy. It should be further noted that the second preset threshold in this invention is greater than the first preset threshold. The first and second preset thresholds can be set by statistical analysis based on the distribution of evidence conflict in historical data fusion records, or they can be configured differently according to the level of business importance and the type of accumulated items. They can also be used as system parameters and adjusted by the administrator according to the fault tolerance requirements and monitoring sensitivity of the actual application scenario, so as to balance the tolerance of fusion inference to normal data fluctuations and the sensitivity to abnormal data. Secondly, when evidence conflicts exist, the weights of the data in each dimension participating in the fusion are adjusted based on the credibility weights of each dimension and the degree of evidence conflict. This is to suppress the interference of abnormal data introduced by collection errors or transmission packet loss on the fusion results. Specifically, when the degree of evidence conflict reaches a medium or high conflict state, the multi-source data fusion sub-model initiates a weight adjustment mechanism. The weight adjustment mechanism first obtains the initial credibility weights of each dimension data, and then calculates the deviation correction coefficient based on the degree of deviation between each dimension data and the reference benchmark data. Specifically, it can be calculated using an exponential decay method based on the deviation rate: first, the absolute value of the relative deviation between each dimension data and the reference benchmark data is calculated, and this absolute value of the relative deviation is used as an input parameter and substituted into a preset exponential decay function. The deviation correction coefficient between 0 and 1 is output through the exponential decay function, so that the deviation correction coefficient approaches zero exponentially when the relative deviation is larger, thereby effectively suppressing excessively large deviations. The deviation correction coefficient ranges from 0 to 1, and the larger the deviation, the smaller the correction coefficient. At the same time, an additional penalty factor is added to the dimension data with a low credibility weight, and the penalty factor is usually set to 0. 3. This value is determined by balancing the contribution of low-reliability data sources in the fusion process with the need to retain their effective information. It can effectively suppress the interference of abnormal data while avoiding the complete exclusion of valuable information due to excessive punishment. In practical applications, the punishment factor can be adjusted according to the sensitivity of the business scenario to data reliability. For scenarios with high fault tolerance requirements, 0.2 can be used to further reduce the contribution of low-reliability data sources. For scenarios that need to retain more data samples, 0.5 can be used to moderately retain the reference value of low-reliability data sources. Furthermore, the initial reliability weight is multiplied by the deviation correction coefficient and then by the punishment factor to obtain the corrected weight. For data whose deviation exceeds the preset abnormal threshold, its correction weight is directly set to 0 to achieve complete isolation of abnormal data. After the weight correction is completed, the correction weights of all dimensions of data are normalized so that the sum of the correction weights of each dimension of data is 1. The normalized correction weights are used as the basis for subsequent fusion inference. Through the above methods, the weights of abnormal data with low reliability weights and large deviations from other data sources are effectively reduced or completely removed during the fusion process, suppressing the interference of collection errors and transmission packet loss on the fusion results. Then, based on the corrected weights, the data of each dimension is fused and inferred, outputting a quantitative evaluation value of the heap state and its confidence information. This enables the uncertainty inference model to maintain its ability to suppress data quality uncertainty throughout the entire chain of data acquisition, transmission, and fusion. Specifically, the multi-source data fusion sub-model obtains the normalized corrected weights of each dimension data after weight correction, and calculates the weighted sum of each dimension data according to its corrected weight to obtain the weighted average quantitative evaluation value of the heap state. At the same time, the multi-source data fusion sub-model calculates the uncertainty measure of the fused evaluation value. This uncertainty measure comprehensively considers the uniformity of the distribution of the corrected weights of each dimension data and the dispersion between each dimension data. When the corrected weights are concentrated and the data of each dimension are relatively uniform, the uncertainty measure is calculated. When the dispersion is low, the confidence information value is higher, and vice versa. The confidence information is calculated using an information entropy-based evaluation method. The information entropy is calculated by adjusting the weights of the data in each dimension, and the entropy value is converted into a confidence index between 0 and 1. The calculated heap state quantitative evaluation value and confidence information are used as the final output of the uncertainty inference model. Through the above mechanism, the uncertainty inference model evaluates the data quality of each dimension through the data quality evaluation sub-model in the data acquisition stage, prioritizes the processing of job event data through the event listening interface in the data transmission stage, and suppresses abnormal data interference through conflict detection and weight adjustment in the data fusion stage. Thus, it maintains an effective ability to suppress data quality uncertainty throughout the entire chain of data acquisition, transmission and fusion.

[0042] Step 104: Match the corresponding three-dimensional entity through the stacking unit identifier, dynamically adjust the visualization rendering attributes of the matched three-dimensional entity according to the stacking status quantitative evaluation value and its confidence information, and render and present the dynamic changes of the stacking status in the panoramic three-dimensional model in real time.

[0043] In some embodiments, matching the corresponding three-dimensional entity through the stacking unit identifier specifically includes: Obtain the stacking unit identifier corresponding to the 3D entity to be updated and rendered, and match the stacking unit identifier with the stacking unit identifier of each 3D entity stored in the panoramic 3D model one by one. When a match is successful, the three-dimensional entity node corresponding to the heap unit identifier is located, and the visualization attribute configuration interface of the three-dimensional entity node is obtained.

[0044] In practical application, firstly, after the uncertainty inference model outputs the quantitative evaluation value of the heap state and its confidence information, the visualization rendering module obtains the corresponding heap unit identifier from the data association mapping table as the query key value based on the business attributes of the target heap unit. It then traverses all 3D entity nodes in the panoramic 3D model, compares the heap unit identifier stored in each node with the query key value, until a 3D entity node with a completely matching identifier is found and identified as the target 3D entity node. Then, it records the memory address of the target 3D entity node to complete the location, and reads the call handle of the visualization attribute configuration interface from the node attributes. This interface encapsulates methods for setting color values, setting transparency, setting texture styles, and applying rendering parameters, establishing an operation channel for dynamically adjusting the 3D entity rendering attributes based on the quantitative evaluation value of the heap state and the confidence information.

[0045] In some embodiments, dynamically adjusting the visualization rendering attributes of the matched 3D entity based on the heap state quantification evaluation value and its confidence information specifically includes: The quantitative evaluation value of the stacking status is mapped to the first rendering channel, which is used to characterize the stacking occupancy degree of the stacked items and reflects the changing trend of the stacking status through color gradient. The confidence information is mapped to the second rendering channel, which is used to characterize the reliability of the heap state evaluation result and reflects the difference in confidence level through transparency gradient or texture density. The rendering parameters of the three-dimensional entity are generated based on the mapping results, wherein the color value of the three-dimensional entity is determined based on the mapping results of the first rendering channel, and the transparency or texture style of the three-dimensional entity is determined based on the mapping results of the second rendering channel. The rendering parameters are sent to the rendering engine to complete the adjustment of the rendering attributes for 3D entity visualization.

[0046] It should be noted that the first rendering channel in this invention is a visualization mapping link used to convert the heap state quantization evaluation value into a three-dimensional entity color value expression; the second rendering channel is a visualization mapping link used to convert confidence information into a three-dimensional entity transparency or texture style expression.

[0047] In practical application, firstly, the quantitative evaluation value of the stacking status is mapped to the first rendering channel. The first rendering channel is used to characterize the stacking occupancy level of the stacked items. The changing trend of the stacking status is reflected by color gradient. Specifically, the visualization rendering module obtains the quantitative evaluation value of the stacking status output by the uncertainty inference model. The evaluation value is expressed as a percentage of the current stacking quantity of the stacking unit to the designed stacking capacity, and the value range is 0 to 100%. The visualization rendering module converts the quantitative evaluation value of the stacking status into the corresponding color value according to the preset color mapping table. The color mapping table defines the color transition sequence corresponding to the stacking occupancy level from low to high. When the stacking occupancy level is low, it corresponds to the green series. When the stacking occupancy level is medium, it corresponds to the yellow series. When the stacking occupancy level is high, it corresponds to the red series. The evaluation value between the mapping points is calculated by linear interpolation to obtain the intermediate color value. Secondly, the confidence information is mapped to the second rendering channel, which is used to characterize the reliability of the heap state evaluation result. The difference in confidence level is reflected by the transparency gradient or texture density. Specifically, the visualization rendering module obtains the confidence information output by the uncertainty inference model. The confidence information is represented by a value between 0 and 1, indicating the reliability of the heap state quantitative evaluation result. The higher the value, the more reliable the evaluation result. The visualization rendering module generates the corresponding transparency value based on the confidence information and sets the confidence information and transparency value to a negative correlation. The higher the confidence, the lower the transparency value, making the 3D entity clearer. The lower the confidence, the higher the transparency value, making the 3D entity more semi-transparent. When using the texture density expression method, the visualization rendering module generates the corresponding texture density parameter based on the confidence information. The higher the confidence, the denser the texture. The lower the confidence, the sparser the texture. Then, the rendering parameters of the three-dimensional entity are generated based on the mapping results. The color value of the three-dimensional entity is determined based on the mapping results of the first rendering channel, and the transparency or texture style of the three-dimensional entity is determined based on the mapping results of the second rendering channel. Specifically, the visualization rendering module uses the color value mapped from the first rendering channel as the basic color attribute of the three-dimensional entity, and the transparency value or texture density parameter mapped from the second rendering channel as the material attribute of the three-dimensional entity. The color value and transparency value or the color value and texture style are combined to form a complete set of rendering parameters. This set of rendering parameters contains all the attribute configuration information required for the three-dimensional entity to display its appearance in the rendering engine. Finally, the rendering parameters are sent to the rendering engine to complete the adjustment of the visualization rendering attributes of the 3D entity. Specifically, the visualization rendering module passes the generated rendering parameters to the rendering engine through the pre-acquired visualization attribute configuration interface of the 3D entity node. After receiving the rendering parameters, the rendering engine recalculates the pixel color and transparency of the 3D entity according to the color value, transparency value or texture style in the rendering parameters, updates the display appearance of the 3D entity in the 3D scene, and outputs the updated 3D scene to the display device. In this way, the rendering parameters are used as input instructions for the rendering engine to adjust the appearance of the 3D entity, and the dynamic adjustment of the visualization rendering attributes of the 3D entity is completed.

[0048] In some embodiments, real-time rendering of the dynamic changes in the storage status of the stockpile in the panoramic 3D model refers to refreshing and displaying the 3D entity with dynamically adjusted rendering attributes in the panoramic 3D model in real time, so as to present the dynamic evolution process of changes in the storage height, storage area occupancy, and storage status confidence of the stockpile in a visual manner.

[0049] On the other hand, in some embodiments, the present invention provides a port operation digital sand table construction system based on panoramic 3D modeling, referencing Figure 3 The figure is a schematic diagram of a port operation digital sand table construction system based on panoramic 3D modeling according to some embodiments of the present invention. The port operation digital sand table construction system based on panoramic 3D modeling includes: a model construction module 301, a data synchronization module 302, a fusion inference module 303, and a visualization rendering module 304, which are described below: The model building module 301 is used to build a panoramic three-dimensional model of the port yard area. The panoramic three-dimensional model includes three-dimensional entities of yard stacks and storage equipment, and assigns a unique storage unit identifier to each three-dimensional entity. Data synchronization module 302 is used to configure a data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the storage yard management platform according to a preset time period or operation event triggering method. The fusion reasoning module 303 is used to perform fusion reasoning on the multidimensional heap storage business data based on a preset uncertainty reasoning model, and output the heap storage status quantitative evaluation value and its confidence information. The uncertainty reasoning model is used to handle the data quality uncertainty that exists in the process of collection, transmission and fusion of multidimensional heap storage business data. The visualization rendering module 304 is used to match the corresponding three-dimensional entity through the stacking unit identifier, dynamically adjust the visualization rendering attributes of the matched three-dimensional entity according to the stacking state quantitative evaluation value and its confidence information, and render and present the dynamic changes of the stacking state in the panoramic three-dimensional model in real time.

[0050] In addition, the present invention also provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the above-described method for constructing a digital sand table for port operations based on panoramic 3D modeling.

[0051] In some embodiments, reference Figure 4 The figure is a schematic diagram of the structure of a computer device for implementing a method for constructing a digital sand table for port operations based on panoramic 3D modeling, according to some embodiments of the present invention. The method for constructing a digital sand table for port operations based on panoramic 3D modeling in the above embodiments can be achieved through… Figure 4 The computer device shown is used to implement this, and the computer device 400 includes at least one processor 401, a communication bus 402, a memory 403, and at least one communication interface 404.

[0052] Processor 401 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0053] The communication bus 402 can be used to transmit information between the aforementioned components.

[0054] The memory 403 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 403 may exist independently and be connected to the processor 401 via the communication bus 402. The memory 403 may also be integrated with the processor 401.

[0055] The memory 403 stores program code for executing the present invention, and its execution is controlled by the processor 401. The processor 401 executes the program code stored in the memory 403. The program code may include one or more software modules. In the above embodiments, the port operation digital sand table construction method based on panoramic 3D modeling can be implemented by the processor 401 and one or more software modules in the program code in the memory 403.

[0056] Communication interface 404 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0057] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0058] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This embodiment of the invention does not limit the type of computer device.

[0059] In addition, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for constructing a digital sand table for port operations based on panoramic 3D modeling.

[0060] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0061] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for constructing a digital sand table for port operations based on panoramic 3D modeling, characterized in that, The method includes the following steps: Construct a panoramic 3D model of the port yard area. The panoramic 3D model includes 3D entities of yard stacks and storage equipment. Assign a unique storage unit identifier to each 3D entity. Configure a data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the storage yard management platform according to a preset time period or operation event triggering method; Based on a preset uncertainty reasoning model, fusion reasoning is performed on the multidimensional heap storage business data to output a quantitative evaluation value of the heap storage status and its confidence information. The uncertainty reasoning model is used to handle the data quality uncertainty that exists in the process of collection, transmission and fusion of multidimensional heap storage business data. The corresponding three-dimensional entity is matched by the stacking unit identifier. The visualization rendering attributes of the matched three-dimensional entity are dynamically adjusted according to the quantitative evaluation value of the stacking status and its confidence information. The dynamic changes of the stacking status of the yard are rendered in real time in the panoramic three-dimensional model. Specifically, based on a preset uncertainty reasoning model, fusion reasoning is performed on the multidimensional heap storage business data to output a quantitative evaluation value of the heap storage status and its confidence information, including: The multidimensional stacked business data is grouped according to data source and data type, and then input into a preset uncertainty inference model. The uncertainty inference model includes a data quality assessment sub-model for different dimensions and a multi-source data fusion sub-model. The credibility weights of data in each dimension are determined through the data quality assessment sub-model. The data of each dimension and its confidence weight are input into the multi-source data fusion sub-model, and fusion reasoning based on evidence theory is performed to output the quantitative evaluation value of the heap status and its confidence information. Specifically, the process of inputting data from each dimension and its confidence weights into the multi-source data fusion sub-model, performing fusion inference based on evidence theory, and outputting a quantitative evaluation value of the heap state and its confidence information includes: The system identifies the degree of evidence conflict between different dimensions of data in determining their stacking status and triggers corresponding fusion strategies based on this degree of conflict. Specifically, the degree of evidence conflict is calculated using a data deviation rate-based evaluation method. The data with the highest credibility weight in each dimension is used as a reference benchmark. The relative deviations of other dimensions from this benchmark in terms of stacking quantity are calculated, and the weighted average of all deviations is taken as the degree of evidence conflict. When the degree of evidence conflict is below a first preset threshold, it is considered a low-conflict state, triggering a direct weighted fusion strategy (i.e., fusion without weight adjustment). When the degree of evidence conflict is between the first and second preset thresholds, it is considered a medium-conflict state, triggering a weight adjustment fusion strategy (i.e., fusion after appropriate adjustment of the weights of each dimension). When the degree of evidence conflict is above the second preset threshold, it is considered a high-conflict state, triggering an anomaly isolation fusion strategy (i.e., fusion after isolating or significantly reducing the weights of the data source with the highest degree of conflict). When conflicting evidence exists, the weights of the data in each dimension participating in the fusion are adjusted based on the credibility weights of each dimension and the degree of conflict, in order to suppress the interference of abnormal data introduced by collection errors or transmission packet loss on the fusion result. Specifically, when the degree of conflict reaches a medium or high level, the multi-source data fusion sub-model initiates a weight adjustment mechanism. This mechanism first obtains the initial credibility weights of each dimension, and then adjusts the weights based on the data in each dimension and the reference baseline data. The deviation correction coefficient is calculated based on the degree of deviation between the data and the reference data. Specifically, it is calculated using an exponential decay method based on the deviation rate: First, the absolute value of the relative deviation between each dimension data and the reference baseline data is calculated. This absolute value of the relative deviation is then used as an input parameter and substituted into a preset exponential decay function. The exponential decay function outputs a deviation correction coefficient between 0 and 1, so that the deviation correction coefficient approaches zero exponentially when the relative deviation is larger, thereby effectively suppressing excessively large deviations. The deviation correction coefficient ranges from 0 to 1, and the larger the deviation, the smaller the correction coefficient. At the same time, an additional penalty factor is added to the dimension data with a lower confidence weight. Based on the corrected weights, the data of each dimension is fused and inferred, and the quantitative evaluation value of the heap status and its confidence information are output, so that the uncertainty inference model can maintain the ability to suppress the uncertainty of data quality throughout the entire chain of data collection, transmission and fusion.

2. The method according to claim 1, characterized in that, Constructing a panoramic 3D model of the port yard area specifically includes: Obtain geospatial data and yard facility parameters for the port storage area; A base topography model of the storage yard is constructed based on the geospatial data and storage yard facility parameters; Obtain the three-dimensional geometric parameters of the stockpile positions and the structural parameters of the storage equipment; Based on the aforementioned yard base terrain model, a three-dimensional entity of the stacking site is constructed according to the aforementioned three-dimensional geometric parameters, and a three-dimensional entity of the equipment is constructed according to the aforementioned structural parameters; Each of the aforementioned stacking position 3D entities and equipment 3D entities is assigned a unique stacking unit identifier, and the stacking unit identifier is associated with the corresponding 3D entity for storage to form a panoramic 3D model.

3. The method according to claim 1, characterized in that, Configure a data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the yard management platform according to a preset time period or operation event triggering method. Specifically, this includes: Configure a scheduled synchronization task to trigger data synchronization operations periodically at preset time intervals; Configure an event listening interface to capture the storage operation event in real time and trigger data synchronization operation when a storage operation event occurs on the yard management platform; In response to data synchronization, a data acquisition request is initiated to the yard management platform based on the storage unit identifier, and multi-dimensional storage business data is received from the yard management platform.

4. The method according to claim 1, characterized in that, The multidimensional storage business data includes the attribute information of the stored items, the quantity information of the stored items, the warehousing time information of the stored items, the outbound plan information of the stored items, the turnover frequency information of the stored items, and the quality inspection information of the stored items.

5. The method according to claim 1, characterized in that, Matching the corresponding 3D entity through the stacking unit identifier specifically includes: Obtain the stacking unit identifier corresponding to the 3D entity to be updated and rendered, and match the stacking unit identifier with the stacking unit identifier of each 3D entity stored in the panoramic 3D model one by one. When a match is successful, the three-dimensional entity node corresponding to the heap unit identifier is located, and the visualization attribute configuration interface of the three-dimensional entity node is obtained.

6. The method according to claim 1, characterized in that, The dynamic adjustment of the visualization rendering attributes of the matched 3D entity based on the quantitative evaluation value of the heap state and its confidence information specifically includes: The quantitative evaluation value of the stacking status is mapped to the first rendering channel, which is used to characterize the stacking occupancy degree of the stacked items and reflects the changing trend of the stacking status through color gradient. The confidence information is mapped to the second rendering channel, which is used to characterize the reliability of the heap state evaluation result and reflects the difference in confidence level through transparency gradient or texture density. The rendering parameters of the three-dimensional entity are generated based on the mapping results, wherein the color value of the three-dimensional entity is determined based on the mapping results of the first rendering channel, and the transparency or texture style of the three-dimensional entity is determined based on the mapping results of the second rendering channel. The rendering parameters are sent to the rendering engine to complete the adjustment of the rendering attributes for 3D entity visualization.

7. The method according to claim 1, characterized in that, Real-time rendering of the dynamic changes in the storage status of the stockpile in the panoramic 3D model means that the 3D entities with dynamically adjusted rendering attributes are refreshed and displayed in the panoramic 3D model in real time, so as to present the dynamic evolution of the changes in the storage height, storage area occupancy and storage status confidence of the stockpile in a visual manner.

8. A port operation digital sand table construction system based on panoramic 3D modeling, wherein the digital sand table is constructed using the method described in any one of claims 1 to 7, characterized in that, The system includes: The model building module is used to build a panoramic 3D model of the port yard area. The panoramic 3D model includes 3D entities of yard stacks and storage equipment, and assigns a unique storage unit identifier to each 3D entity. The data synchronization module is used to configure the data synchronization triggering mechanism, which obtains multi-dimensional storage business data of the corresponding storage unit from the storage yard management platform according to a preset time period or operation event triggering method. The fusion reasoning module is used to perform fusion reasoning on the multidimensional heap storage business data based on a preset uncertainty reasoning model, and output the heap storage status quantitative evaluation value and its confidence information. The uncertainty reasoning model is used to handle the data quality uncertainty that exists in the process of collection, transmission and fusion of multidimensional heap storage business data. The visualization rendering module is used to match the corresponding three-dimensional entity through the stacking unit identifier, dynamically adjust the visualization rendering attributes of the matched three-dimensional entity according to the quantitative evaluation value of the stacking status and its confidence information, and render and present the dynamic changes of the stacking status in the panoramic three-dimensional model in real time.

Images