Digital Twin Loading Simulation Unmanned Transfer System and Method for High-Speed ​​Rail Express Stations

By constructing a fully closed-loop operation system and digital twin technology, the problems of information collection relying on manual labor, low transfer efficiency, and numerous safety hazards at high-speed rail express stations have been solved. This has enabled balanced cargo weight and reasonable distribution of carriage loads, improving transfer efficiency and safety, and supporting unmanned and intelligent management throughout the entire process.

CN123116767APending Publication Date: 2026-11-17HUBEI REMOTE RAILWAY TECH +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610963460.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-30
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

High-speed rail express stations lack an integrated intelligent loading and unloading system. Cargo information collection relies on manual labor, resulting in low transfer efficiency, numerous safety hazards, a lack of real-time digital management and control, poor equipment versatility, and an inability to achieve fully unmanned closed-loop operation.

Method used

A fully closed-loop operation system was constructed, consisting of DWS full-scale perception, backend simulation, dynamic return flow of the circular line, robot balanced loading, AGV carriage loading, and real-time simulation of the twin. The minimum variance balanced loading optimization model and the load-volume dual-constraint judgment algorithm were adopted. Combined with the real-time center of gravity calculation of four-point weighing, a dynamic return flow mechanism for the circular conveyor line and a congestion prediction and speed regulation algorithm based on queue theory were designed. A three-dimensional virtual scene was constructed for real-time monitoring.

Benefits of technology

It achieves balanced cargo weight and reasonable load distribution in the carriages, avoiding the risk of overturning, improving transfer efficiency and safety, realizing unmanned operation and intelligent management throughout the entire process, and reducing reliance on manual labor and equipment modification costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116767A_ABST
    Figure CN123116767A_ABST
Patent Text Reader

Abstract

This invention discloses a digital twin-based unmanned transfer system and method for high-speed rail express stations. The system includes: a cargo receiving area, a cargo scanning and identification area, a circular conveyor line, an intensive operation area, a railway platform loading and unloading area, roller conveyors located in the sections, a back-end data analysis platform, and a digital twin system, arranged sequentially along the cargo transfer flow direction. The back-end data analysis platform is equipped with a loading simulation module, which executes a minimum variance-based balanced loading optimization model and generates cargo diversion and loading instructions by combining a load-volume dual-constraint judgment algorithm. The circular conveyor line is used to control the cyclical return of cargo and perform secondary identification and diversion when the target container is fully loaded, based on a fillability judgment function. The intensive operation area uses four-point weighing to calculate the center of gravity coordinates of the container in real time to achieve anti-tipping control. The digital twin system predicts congestion based on queue theory and dynamically adjusts the conveyor cycle. It can realize fully unmanned operation of high-speed rail express.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of high-speed rail express technology, and more specifically, relates to a digital twin loading simulation unmanned transfer system and method for high-speed rail express stations. Background Technology

[0002] With the marketization of high-speed rail freight, the volume of high-speed rail express freight has been growing rapidly year by year. This places higher demands on the timeliness, standardization, and safety of the entire process of cargo receiving, sorting, transshipment, and train loading at railway stations. Currently, most high-speed rail express stations in China have not yet established integrated intelligent loading and unloading systems. Cargo transshipment operations still rely primarily on manual labor, supplemented by traditional machinery, resulting in multiple technical shortcomings. First, cargo information collection relies on manual registration and lacks integrated automatic detection and identification equipment. Existing stations rely solely on manual barcode scanning and manual weighing and measurement of cargo dimensions, resulting in delayed data entry, large errors, and an inability to collect cargo weight, volume, and flow information in batches simultaneously, making it difficult to achieve pre-loading simulation planning. Secondly, there is no closed-loop automated production line for cargo sorting and transfer, resulting in low transfer efficiency and easy congestion. Existing stations mostly use forklifts for single-point transfer of goods, without continuous closed-loop conveyor channels, and goods to different destinations are mixed and stacked. During peak freight periods, problems such as cargo accumulation, missorting, and disconnection of the transfer link are very likely to occur, directly delaying the departure of high-speed rail freight trains. Third, the entire process of cargo containerization and loading relies on manual operation, resulting in high labor costs and significant safety hazards. When manually loading cargo into containers, it is impossible to accurately control the load of a single container and the internal counterweight, which can easily lead to container imbalance and tipping over during on-site transfer. During the train loading stage, containers are randomly placed by manual labor, resulting in uneven load distribution in the carriages and a risk of tipping over during high-speed train travel. At the same time, the high-intensity manual sorting and packing operations are labor-intensive, and the rate of cargo damage and loss remains high. Fourth, the station operation scheduling lacks real-time digital management and control methods. Traditional stations rely solely on manual on-site inspections and scheduling, lacking a full-station 3D visualization simulation monitoring platform, and are unable to synchronize real-time data on cargo transportation, equipment operation, and train loading across all dimensions; they cannot predict cargo flow congestion points in advance, and can only manually handle accumulation problems afterward, resulting in delayed scheduling adjustments; Fifth, the transfer equipment has poor versatility. Existing roller conveyors and conveyor systems are only compatible with single-size containers. When freight volume fluctuates or different types of containers such as bags and pallets are used as loading carriers, the conveyor equipment needs to be modified or replaced, resulting in weak flexibility and adaptability.

[0003] While some existing stations have introduced simple sorting robots or single cargo scanning devices, these technologies only achieve partial automation of a single process. They cannot realize fully unmanned closed-loop operations of the entire process, from cargo entry identification to cyclic sorting, robot-based bulk packing, and platform AGV transfer to the train. Furthermore, they have not combined digital twin technology to achieve simultaneous virtual-real simulation, pre-loading simulation, and dynamic scheduling of the entire station. As a result, they cannot simultaneously address multiple industry pain points such as efficiency, loading safety, visual management and control, and flexible adaptation, thus hindering the large-scale and high-quality development of high-speed rail express. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, this invention provides a method and system for a digital twin-based unmanned transfer system for high-speed rail express stations. The core of this system lies in constructing a fully closed-loop operation system comprising "DWS full-scale perception - backend simulation - dynamic return flow on a circular line - robot balanced loading - AGV carriage loading - real-time simulation of the twin." By introducing a balanced loading optimization model based on minimum variance and a load-volume dual-constraint judgment algorithm, the system forces the minimization of weight variance among containers destined for the same destination while meeting the rigidity of railway load-bearing standards. Combined with real-time center of gravity calculation at four weighing points, it achieves anti-tipping control, completely eliminating the risk of uneven loading caused by manual loading. Simultaneously, by using a fillable judgment function to drive the circular conveyor line to execute a dynamic return flow strategy, and in conjunction with congestion prediction and speed adjustment algorithms based on queue theory, it achieves adaptive balance and forward-looking scheduling of the entire station's logistics cycle time. This effectively solves the problems of cargo accumulation and train delays during peak periods, significantly improving operational safety, throughput efficiency, and management intelligence.

[0005] To achieve the above objectives, one aspect of the present invention provides a digital twin loading simulation unmanned transfer system for high-speed rail express stations. The system comprises a cargo transfer area, a cargo scanning and identification area, a circular conveyor line, an intensive operation area, a roller conveyor, and a railway platform loading and unloading operation area, arranged sequentially along the cargo transfer flow direction. It also includes a backend data analysis platform for collecting multi-dimensional basic data of cargo and generating destination-specific pre-loading simulation schemes, and issuing equipment control commands. Furthermore, it is a digital twin system that interacts with the backend data analysis platform and field equipment to construct a three-dimensional virtual scene of the entire station process, real-time mapping of dynamic operating data of field equipment and cargo, simulation and dynamic deduction of cargo flow operation conditions within the station, prediction of congestion nodes, and output of transport rhythm scheduling adjustment commands to the backend data analysis platform. The cargo receiving area is used to receive cargo to be transferred; the cargo scanning and identification area is equipped with a DWS device, which is used to synchronously collect the external dimensions, weight, and freight destination information of a single shipment and transmit it to the back-end data analysis platform in real time. The backend data analysis platform is connected to the DWS detection equipment and is configured with a simulated loading module. The simulated loading module performs pre-loading simulation calculations based on the original cargo data to generate cargo diversion routing instructions and container loading instructions, and sends the two types of instructions to the control units corresponding to the circular conveyor line and the intensive operation area. The railway platform loading and unloading area is equipped with AGV transfer trolleys, which are remotely dispatched and controlled by the back-end data analysis platform. The AGV transfer trolleys receive the loading instructions issued by the back-end data analysis platform, transfer and buffer the fully loaded containers output from the roller conveyor, and transfer them to the inside of the carriage after the train enters the station. The containers are arranged according to the pre-loading simulation scheme to balance the overall load distribution of the carriage. The digital twin system establishes data interaction channels with the backend data analysis platform and the automated equipment in various areas of the station.

[0006] Furthermore, the circular conveyor line is a closed-loop circulating conveyor structure, used to receive goods output from the goods scanning and identification area, and according to the routing instructions issued by the back-end data analysis platform, to divert and transport the goods to the corresponding destination intensive operation area according to the pre-loading simulation scheme. When the backend data analysis platform determines that there is no available loading space in the destination intensive operation area corresponding to the goods, it controls the circular conveyor line to bypass the corresponding operation area and continue to transport the goods in a loop until the container at the destination makes room. Then, the DWS device will perform secondary identification and replan the route.

[0007] Furthermore, at least one group of intensive operation areas is set up, and each group of intensive operation areas corresponds to a single freight destination; Each intensive operation area is equipped with loading robots and automatic lifting transmission equipment; the automatic lifting transmission equipment supports and places containers and integrates a weighing and verification device; the loading robot receives loading instructions and loads the corresponding goods into the container; the weighing and verification device collects the total loaded weight of the container in real time, and when the verification determines that the total weight of the container reaches the railway's preset load-bearing threshold, it drives the automatic lifting transmission equipment to complete the container swapping operation of pushing the fully loaded container and replacing the empty container.

[0008] Furthermore, the simulated loading module employs a load-volume dual-constraint judgment algorithm to perform real-time feasibility verification on each piece of cargo to be loaded. Formula for calculating the load constraint of a container: in, Number the target container, such as A1, A2, B1, B2; The serial number of the goods to be loaded; The container has been filled. The collection consisting of all the goods inside; For container Total weight of cargo currently loaded; For the actual test of DWS device Weight of goods, in kg; The maximum rated load of a single container is limited by railway express transport regulations, in kg. Formula for calculating container volume constraints: in, For container The total volume occupied by the currently loaded cargo; For the actual test of DWS device Volume of a piece of cargo, in dm³; This represents the upper limit of the effective loading volume inside the container, in dm³. Comprehensive formula for determining whether cargo can be loaded: in, Assign identification tags to the cargo; Representative goods Can be loaded into container ; This indicates that the current container is underloaded or its volume is insufficient, and the goods need to be circulated back along the circular conveyor line to wait for an empty container. In addition to determining the feasibility of a single cargo, the simulated loading module is configured to: acquire the weight and volume data of each cargo to be loaded at the same destination; optimize the total weight variance of each container by minimizing the variance; and distribute the cargo to multiple containers at the destination by combining the rated load and volume constraints of each container, so that the loading weight of each container tends to be balanced to meet the train's load-bearing requirements. Furthermore, the module dynamically adjusts the cargo arrangement based on the multi-point weighing data at the bottom of the container to optimize the internal center of gravity distribution of the container and prevent overturning during on-site transfer or train operation. The objective optimization function is: The constraints are: in, This represents the total number of containers required for the current freight destination. The average total load of all containers at that destination; The container load variance; , For any two different numbered containers; Indicates any two different numbered containers and The intersection is empty; Gather all goods to the same destination The iterative solution model outputs the cargo allocation set for each container. The system generates a cargo loading sequence and robot grasping coordinate control parameters, which are then issued as global loading commands.

[0009] Furthermore, the automatic lifting transmission device is configured to descend gradually as the cargo accumulates in the container during loading to maintain a constant relative height between the loading robot and the top of the cargo; when the container is fully loaded, the automatic lifting transmission device rises to a height level with the roller conveyor, pushes the fully loaded container onto the roller conveyor, and then descends to position a new empty container for the next round of loading operations.

[0010] Furthermore, the digital twin system includes a 3D modeling module, a real-time data mapping module, and a simulation and deduction module; the real-time data mapping module is used to map the operating status of field equipment, cargo location, and transportation speed to the 3D model, and the simulation and deduction module is used to perform bottleneck prediction and operation cycle optimization based on the mapped data.

[0011] Furthermore, it also includes a monitoring and control room, which is equipped with a large visualization screen. The visualization screen is connected to the digital twin system and is used to display the daily cargo volume, the volume of cargo waiting to be transported, the equipment utilization rate, and the three-dimensional dynamic simulation screen. It also supports operators to issue manual intervention instructions based on the simulation results to adjust the operating parameters.

[0012] Furthermore, the DWS device integrates a volume measurement sensor, a weighing sensor, and a barcode / QR code recognition camera, and the DWS device is set along the conveying path of the circular conveyor line to perform multiple data collection and verification on the goods being conveyed in a circular manner.

[0013] A second aspect of the present invention provides a digital twin-based unmanned transfer method for high-speed rail express stations, implemented using the aforementioned digital twin-based unmanned transfer system for high-speed rail express stations, comprising: S1. Receive the goods, use DWS equipment to collect the weight, dimensions and destination information of the goods and upload them to the back-end data analysis platform; S2. The back-end data analysis platform performs automatic destination-specific loading simulation based on the collected information, calculates the optimal loading scheme for each container, and generates corresponding loading and routing instructions. S3. Control the circular conveyor line to transport goods according to the routing instructions. If the target intensive operation area is full, control the goods to circulate back on the circular conveyor line and wait for the next identification and diversion. S4. Control the loading robot to load the goods into the container according to the optimal loading scheme, and at the same time check the loading weight in real time through the weighing device. In response to the signal that the container is full, execute the operation of automatic container changing and conveying to the roller conveyor. S5. Control the AGV transfer trolley to transfer the fully loaded container on the roller conveyor to the train car and perform balanced loading within the car; S6. The entire process from steps S1 to S5 is monitored and simulated in real time using a digital twin system. The work rhythm is dynamically adjusted based on the simulation results to prevent congestion.

[0014] Furthermore, the destination-specific automatic simulation loading simulation described in step S2 includes: Collect statistics on the weight and volume of all goods awaiting loading to the same destination; Based on greedy algorithms or genetic algorithms, cargo combinations are allocated to multiple containers to minimize the total weight variance of each container and ensure that the weight of a single container does not exceed the limit. Generate the cargo loading sequence for each container and the corresponding robot grasping coordinates; Step S6, which involves dynamically adjusting the work cycle, includes: Monitor the dwell time of goods at each workstation using a digital twin system; When the time the goods are held up in any link exceeds the preset threshold, the system automatically generates a speed adjustment command to reduce the delivery speed of upstream goods or increase the processing priority of downstream equipment, and displays the adjusted status on the large visualization screen as an early warning.

[0015] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: (1) This invention breaks through the limitations of traditional manual loading, which relies solely on experience and is highly random. By introducing a balanced loading optimization model based on minimum variance, it forces the weight of each container to be consistent to the same destination, under the premise of satisfying the dual hard constraints of single container load and volume. At the same time, combined with a four-point weighing real-time center of gravity calculation algorithm, it dynamically monitors and corrects the internal center of gravity shift of the container. This "dual guarantee" mechanism eliminates the risk of overturning in the yard and train rollover caused by single container overweight, axle load imbalance of the whole vehicle, or container center of gravity shift, significantly improving the safety level of high-speed rail express.

[0016] (2) To address the pain point of congestion during peak periods, this invention designs a unique dynamic return flow mechanism for the circular conveyor line. By introducing a filling determination function, the system can determine the capacity status of the downstream intensive operation area in real time. When the target container is fully loaded, the system automatically controls the goods to circulate and wait on the circular line and triggers secondary identification and diversion, rather than blindly piling them up. Combined with a congestion prediction and speed regulation algorithm based on queue theory, the digital twin system can proactively predict bottleneck nodes and automatically issue speed reduction commands, achieving adaptive balance of the logistics cycle time of the entire station, effectively eliminating coupling blockages between processes, and ensuring the punctuality rate of high-speed train departures.

[0017] (3) This invention changes the traditional "black box" operation mode of the station, which is "invisible, intangible, and uncontrollable". By constructing a three-dimensional virtual model that is mapped to the physical station in real time, the system realizes the digital presentation of all elements such as cargo location, equipment status, and environmental parameters. The system can not only reproduce the current working conditions, but also predict the congestion situation and equipment load in the future based on simulation and deduction. It supports seamless switching between manual intervention and automatic control, greatly reduces the reliance on manual inspection on site, and improves the speed and accuracy of emergency response.

[0018] (4) This invention solves the problems of high equipment specialization and high replacement cost in the prior art by configuring universal roller conveyors and modular lifting transmission equipment that are compatible with various specifications of containers such as baggage and pallets. No matter how the freight volume fluctuates or how the container is replaced, the underlying logic of the system and the hardware interface remain consistent. The operating mode can be quickly switched without stopping work for modification, which effectively supports the large-scale and diversified development needs of high-speed rail express business.

[0019] (5) This invention automatically collects all dimensions of cargo data, such as weight, volume, flow direction and loading time, through the linkage between DWS device and back-end platform, forming an unalterable freight archive. These data not only serve the current loading simulation, but also provide high-value data assets for long-term station layout optimization, transportation resource allocation and cost accounting, helping high-speed rail express to upgrade from a labor-intensive to a technology-intensive industry. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a digital twin loading simulation unmanned transfer system for a high-speed rail express station, according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating an unmanned transfer method using digital twin loading simulation at a high-speed rail express station, according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0022] like Figure 1 As shown, one aspect of the present invention provides a digital twin loading simulation unmanned transfer system for high-speed rail express stations, including a cargo receiving area 1, a cargo scanning and identification area 2 equipped with DWS equipment, a circular conveyor line 3, intensive operation areas corresponding to different destinations 4, railway platform loading and unloading operation areas 5, and a general roller conveyor 6 located in the section, arranged sequentially along the cargo transfer flow direction; it also includes a back-end data analysis platform and a digital twin system; the back-end data analysis platform is used to collect multi-dimensional basic data of cargo and generate pre-loading simulation schemes for different destinations, and issue equipment control commands; the digital twin system is used to construct a three-dimensional virtual scene of the station, map on-site equipment and cargo operation data in real time, dynamically deduce the cargo flow status, and output scheduling adjustment commands; the back-end data analysis platform is equipped with a simulated loading simulation system. The true module is configured to execute a minimum variance-based balanced loading optimization model, combined with a load-volume dual-constraint judgment algorithm, to generate cargo diversion and loading instructions; the circular conveyor line 3 is configured to control cargo circulation and secondary diversion when the target container is fully loaded, based on a fillable judgment function; the intensive operation area 4 calculates the center of gravity coordinates of the container in real time through four-point weighing to achieve anti-tipping control; the digital twin system predicts congestion based on queue theory and dynamically adjusts the conveying rhythm; this invention can solve the problems of uneven manual loading, easy congestion during peak periods, and lack of visual management in existing high-speed rail express stations, and achieves balanced container weight, compliant carriage axle load, adaptive scheduling of logistics throughout the station, and unmanned operation throughout the entire process, significantly improving the safety, punctuality, and intelligence level of high-speed rail express.

[0023] Furthermore, the DWS device integrates a volume measurement sensor, a weighing sensor, and a barcode / QR code recognition camera to simultaneously collect cargo data such as the dimensions, weight, and destination information of a single shipment and transmit it unidirectionally to the backend data analysis platform in real time. The DWS device is set along the conveying path of the circular conveyor line 3 to perform multiple data collection and verification on the cyclically conveyed cargo.

[0024] Furthermore, the circular conveyor line 3 is a closed-loop circulating conveyor structure, used to receive goods output from the goods scanning and identification area 2, and to divert and transport the goods to the corresponding destination intensive operation area 4 according to the pre-loading simulation scheme. When the back-end data analysis platform determines that the destination intensive operation area 4 corresponding to the goods currently has no available loading space, it controls the circular conveyor line 3 to bypass the corresponding operation area and continue to circulate and transport the goods until the container at the destination makes room. Then, the DWS device performs secondary identification and replans the route. The circular conveyor line is configured with multiple sets of DWS identification points along the conveying path, and the goods information can be verified and diverted twice during the goods circulation process.

[0025] Furthermore, at least one set of intensive operation areas 4 is set up, with each set of intensive operation areas 4 corresponding to a single freight destination; each intensive operation area 4 is equipped with a loading robot and an automatic lifting transmission device; the loading robot can completely replace manual labor to complete the cargo sorting and loading operations, realizing automated container loading in the intensive operation area without human intervention; the container is placed on the automatic lifting transmission device, which integrates a weighing and verification device; the loading robot receives the pre-loading simulation results sent by the back-end data analysis platform, and loads the cargo into the container according to the destination based on the feedback of the simulation results; during the loading operation, the weighing and verification device collects the total weight of the container after loading in real time and uploads the data. The total weight data of the containers is sent to the backend data analysis platform to verify whether the total load of the containers meets the railway transportation load limit standards. If the total weight of the containers exceeds the railway load requirements, the system triggers an alarm and suspends the loading robot's loading operation. After the containers are fully loaded, the automatic lifting transmission equipment raises the containers to the same height as roller conveyor 6 and pushes them onto roller conveyor 6. Then, the containers automatically descend and automatically switch to the work station for uninterrupted continuous container loading operations. After the cargo loading is completed based on the pre-loading simulation calculations guided by the backend data analysis platform, the total weight of each container tends to be consistent, and the total load of all containers meets the railway transportation load specifications. Moreover, the weight distribution of the cargo inside the containers is uniform, which can prevent the containers from tipping over due to center of gravity shift or counterweight imbalance.

[0026] Furthermore, to prevent the container from tipping over due to a shift in the center of gravity during on-site transfer or train travel, in this embodiment, the weighing device of the automatic lifting transmission equipment adopts a four-point pressure sensor array. The system calculates the projected coordinates of the center of gravity of the container's bottom surface in real time, using the following formula: in, The calculated lateral and longitudinal offsets of the center of gravity in the coordinate system of the container bottom surface; These are the length and width of the bottom surface of the container, respectively. Real-time readings from the four pivot sensors; This represents the current total weight of the container; System preset safety factor (Usually taken as 0.1~0.2), when or If the system detects an off-center loading anomaly, it will pause loading the cargo and send an alarm command to the digital twin platform, prompting the user to adjust the cargo placement or fine-tune the cargo position via a robot.

[0027] Furthermore, the roller conveyor 6 is used to transfer fully loaded containers to the railway platform loading and unloading area 5 for different destinations; the roller conveyor is compatible with various specifications of containers such as baggage and pallets, and can be used to transfer different loading carriers without replacing the roller conveyor body.

[0028] Furthermore, the railway platform loading and unloading area 5 is equipped with an AGV transfer trolley remotely controlled by a back-end data analysis platform. The AGV transfer trolley is used to buffer and transfer the containers output from the roller conveyor 6 into the train carriages, and to arrange the containers according to the pre-loading simulation scheme to balance the load on the carriages. The AGV transfer trolley is used to perform a two-stage transfer process: before the train enters the station, the containers are buffered in a designated area of ​​the platform; after the train enters the station, the buffered containers are transferred into the carriages, and the placement of the containers is adjusted according to the pre-loading simulation scheme to balance the weight distribution of the carriages and avoid the risk of train rollover.

[0029] Furthermore, the backend data analysis platform is equipped with a simulated loading module. This module generates a pre-loading simulation plan for each destination based on the cargo data uploaded by the DWS device, and distributes the simulation plan to the control units of the circular conveyor line 3 and the intensive operation area 4. The simulated loading module pre-calculates the remaining capacity and load threshold of each container at a single destination. When the cargo to be transported cannot be loaded into the container at the current workstation, it determines that the cargo will flow back to the circular conveyor line to wait for the next empty container. The backend data analysis platform synchronously stores and archives all cargo dimensions, weights, flow directions, container loading, and AGV carriage layout data for the entire process, forming a complete freight data ledger for traceability and station operation optimization.

[0030] Furthermore, to achieve precise matching between cargo and containers, the simulated loading module employs a load-volume dual-constraint judgment algorithm to perform real-time feasibility verification for each piece of cargo to be loaded. Specifically, it simultaneously verifies two hard indicators: the container's maximum load capacity and the maximum effective loading volume. Only when both constraints are met can the current cargo be allocated to the corresponding container. The calculation formula is as follows: Formula for calculating the load constraint of a container: in, Number the target container, such as A1, A2, B1, B2; The serial number of the goods to be loaded; The container has been filled. The collection consisting of all the goods inside; For container Total weight of cargo currently loaded; For the actual test of DWS device Weight of goods, in kg; The maximum rated load of a single container is limited by railway express transport regulations, in kg; this formula is used to check whether the total weight after loading exceeds the railway transport load limit. Formula for calculating container volume constraints: in, For container The total volume occupied by the currently loaded cargo; For the actual test of DWS device Volume of a piece of cargo, in dm³; This is the upper limit of the effective loading volume inside the container, in dm³; this formula is used to prevent cargo from being unable to close or being damaged by compression due to exceeding the physical size limit. The comprehensive judgment formula can be filled in for the goods: in, Assigning markings to cargo to indicate its contents Can it be loaded into the container? ; Representative goods Can be loaded into container ; This indicates that the current container is underloaded or its volume is insufficient, and the goods need to be circulated back along the circular conveyor line to wait for an empty container. The calculation logic of the simulated loading module is as follows: For all containers at the same freight destination, the above formula is substituted into the calculation, and priority is given to selecting... Furthermore, the container with the smallest remaining load margin is allocated cargo to ensure a uniform loading capacity across all containers; if the calculation results for all containers at the destination are... The platform sends a return control command to the circular conveyor line, and the goods are transported in a cycle. After the fully loaded container is pushed to the roller conveyor and the empty container is replaced, the calculation is recalculated and determined.

[0031] Furthermore, based on the feasibility assessment of individual cargo, in order to meet the requirements of train operation stability, the backend data analysis platform further adopts a greedy algorithm or genetic algorithm to globally allocate cargo to the same destination with the goal of minimizing the load variance of each container; the simulated loading module, based on the feasibility assessment of individual cargo, is also configured to: acquire the weight and volume data of each cargo to be loaded at the same destination, with the goal of minimizing the total weight variance of each container, and combine the rated load and volume constraints of a single container to allocate cargo to multiple containers at the destination, so that the loading weight of each container tends to be balanced to meet the train's load-bearing requirements; and dynamically adjust the cargo arrangement based on the multi-point weighing data at the bottom of the container to optimize the internal center of gravity distribution of the container and prevent overturning during in-yard transfer or train operation. The objective optimization function is: The constraints are: in, This represents the total number of containers required for the current freight destination. The average total load capacity of all containers at this destination (unit: kg). The variance of the container load is minimized to make the total weight of each container at the same destination more consistent, thereby avoiding safety hazards caused by uneven axle load in the train. , For any two different numbered containers; Indicates any two different numbered containers and The intersection of these two containers is empty, meaning that a single item can only be assigned to one container and cannot be assigned repeatedly. Logistics process: Gathering all goods destined for the same destination The iterative solution model outputs the cargo allocation set for each container. The system generates a cargo loading sequence and robot grasping coordinate control parameters, which are then issued as global loading commands.

[0032] Furthermore, the digital twin system is communicatively connected to the backend data analysis platform and equipment in various areas of the station. The digital twin system includes a 3D model construction module, a real-time data mapping module, and a simulation module. The 3D model construction module is used to construct a 3D scene model of the entire station process. The real-time data mapping module collects real-time operating data of station equipment and goods and maps it to the 3D scene model. The simulation module dynamically simulates the operation status of the cargo flow based on real-time data and outputs scheduling adjustment instructions. During operation, the digital twin system constructs a 3D virtual scene of the entire loading and transfer process of the station based on the real-time operating data of the station's on-site equipment and goods, and simultaneously conducts real-time working condition simulation, generating a 3D visualized dynamic simulation screen to achieve real-time monitoring of the cargo transfer status of the entire station. The simulation module of the digital twin system monitors the cargo congestion at each workstation, uses a congestion prediction and speed adjustment algorithm based on queue theory to predict cargo flow accumulation and congestion nodes, and automatically sends conveying speed adjustment instructions to the backend data analysis platform, linking the circular conveyor line 3 to adjust the cargo conveying rhythm.

[0033] The queue length prediction model expression in the congestion prediction and speed regulation algorithm based on queue theory is as follows: in, for The queue length at a certain workstation (such as the scanning and identification area or sorting port) at any given time; For the real-time arrival rate of goods; This refers to the processing rate (i.e., the equipment throughput capacity) of that workstation. This is for simulating step size; When predicting queue length Exceeding the preset security threshold When the system detects that congestion is imminent at a workstation, the digital twin system sends a speed-reduction command to the backend data analysis platform to adjust the operating parameters of the upstream equipment. in, This refers to the current operating speed of the upstream circular conveyor line or connecting area; The adjusted target speed; The deceleration coefficient (usually taken as a value) ); By reducing the speed of upstream cargo flow, a dynamic balance of the logistics rhythm of the entire station can be achieved, avoiding system paralysis caused by local accumulation.

[0034] Furthermore, the system also includes a station monitoring room; the station monitoring room is equipped with a large visualization screen, which is connected to the digital twin system to display real-time operational indicators such as the station's daily transport volume, the inventory of goods awaiting transfer, and the real-time transport rate of goods, allowing on-site operators to manually intervene and adjust on-site operation parameters based on the station's real-time operating conditions.

[0035] Furthermore, the goods to be transferred in this invention are bagged or boxed goods; the goods flow from the goods receiving area 1 into the goods scanning and identification area 2 to complete the goods information collection process; after the goods enter the DWS equipment detection station, the external dimensions are measured, the weight is weighed, and the freight destination barcode is identified simultaneously; the DWS equipment uploads the collected raw data such as the weight, external dimensions, and freight destination to the back-end data analysis platform in real time, and the back-end data analysis platform completes the data storage, classification, and processing, and performs pre-loading simulation calculations for the goods to different destinations based on the collected data; the back-end data analysis platform sends the pre-loading simulation calculation results to the identification and control unit of the circular conveyor line 3, thereby realizing the accurate diversion of goods and guiding the goods to the intensive operation area 4 of the corresponding freight destination.

[0036] like Figure 2 As shown, a second aspect of the present invention provides a digital twin-based unmanned transfer method for high-speed rail express stations, implemented using the aforementioned digital twin-based unmanned transfer system for high-speed rail express stations, comprising the following steps: S1. Cargo Inbound Information Collection: Bagged / boxed goods are sent from cargo receiving area 1 to cargo scanning and identification area 2. The external dimensions, weight, and freight destination information of each piece of cargo are collected synchronously through DWS equipment, and all original cargo data are uploaded to the back-end data analysis platform in real time. S2. Destination-specific preloading simulation calculation: The back-end data analysis platform stores, classifies, and calculates the received cargo data, and conducts container preloading simulation for different freight destinations; the simulation calculates the rated load and remaining capacity of the containers at each destination, generates cargo diversion, loading and distribution control schemes, and sends the simulation schemes to the control units of the circular conveyor line 3 and the intensive operation area 4. S3, Circular Closed-Loop Circular Diversion Conveying: Goods enter the circular conveyor line 3 for cyclic conveying; during the conveying process, the pre-loading simulation scheme is continuously matched to determine the remaining loading capacity of the corresponding intensive workstation container: if the current destination container has a capacity reserve, the goods are controlled to be conveyed to the corresponding intensive operation area 4; if the current container is full and has no capacity space, the goods are controlled to circulate back along the circular conveyor line 3, pass through the DWS equipment for information verification a second time, and wait for an empty container to arrive before diverting; S4. Robotic Automatic Loading and Load Verification: The diverted goods are loaded by the loading robot into the container on the automatic lifting transmission equipment according to the pre-loading simulation scheme. The weighing and verification device on the automatic lifting transmission equipment collects the total weight of the container after loading in real time and verifies whether the total load meets the railway transportation load-bearing standards. When the container reaches the rated loading threshold and is determined to be fully loaded, the automatic lifting transmission equipment raises the container to the same height as the roller conveyor and pushes the fully loaded container to the roller conveyor surface; then the equipment resets and lowers, positioning the empty container at the filling station, and the filling operation continues in a cycle. Roller conveyor 6 receives fully loaded containers and transports them to the railway platform loading and unloading area 5 according to the freight destination. S5. Platform AGV buffer and train balanced loading: The back-end data analysis platform remotely dispatches AGV transfer trolleys to transfer the containers output from roller conveyor 6 to the platform buffer area for temporary storage; after the freight train enters the station, the AGV transfer trolleys transfer the buffer containers to the inside of the train carriages; the AGV transfer trolleys arrange the containers in the carriages according to the pre-loaded simulation loading instructions to balance the overall load distribution of the carriages and prevent the train from overturning. S6. Steps for digital twin virtual-real synchronous simulation and dynamic scheduling: The digital twin system collects real-time operational data of equipment and goods throughout the entire station area, constructs a three-dimensional virtual scene of station loading and transfer, and completes real-time data mapping between the physical site and the virtual model; The digital twin system simulates and extrapolates the operational conditions of cargo flow in the field in real time. It adopts a congestion prediction and speed adjustment algorithm based on queue theory to predict cargo accumulation and transportation congestion nodes, and automatically sends transportation speed adjustment instructions to the back-end data analysis platform to dynamically optimize the cargo flow transportation rhythm. The S7 and the monitoring room's large visual screen synchronously load digital twin simulation images, displaying real-time operational data such as daily transport volume, waiting volume, and transport rate, and supporting manual intervention by operators to adjust operational parameters.

[0037] Furthermore, the destination-specific automatic simulation loading simulation described in step S2 includes: Collect statistics on the weight and volume of all goods awaiting loading to the same destination; Based on greedy algorithms or genetic algorithms, cargo combinations are allocated to multiple containers to minimize the total weight variance of each container and ensure that the weight of a single container does not exceed the limit. Generate the cargo loading sequence for each container and the corresponding robot grasping coordinates; Specifically, it includes the following steps: Cargo and Container Modeling Definition: Let set This refers to all goods awaiting loading to the same destination, among which... The attributes of a shipment are defined as follows: In the formula: :No. The weight of the goods, in kg; :No. Volume of outer packaging of goods, unit ; The three-dimensional coordinates of the cargo within the container are generated by a subsequent layout algorithm; let the set be... This represents the set of containers required for the destination, where The state of a container is defined as follows: In the formula: Allocated to container A set of goods indexes; : container The current total load weight; : container The current total volume of the load.

[0038] Single-container hard constraint: for any container The carrier loading process must meet the following dual constraints: In the formula: The maximum permissible load of a single container as specified in railway industry standards; : Maximum effective loading volume inside the container; Global optimization objective function (minimum variance model): The core optimization objective of this step is to minimize the weight difference between containers at the same destination to improve the axle load balance of the train. The objective function is expressed as: In the formula: : Total weight variance of the container assembly; : The average total weight of all containers at this destination; Goods mutual exclusion and uniqueness constraints: To ensure that each item is allocated only once, the following must be satisfied: In the formula: and These represent any two different containers, with their intersection being empty, meaning that no cargo can be loaded into two containers at the same time; Greedy allocation decision logic (online implementation strategy): In actual assembly line operations, the system uses a greedy strategy based on residual margin for real-time allocation, and the decision function is as follows: when When loading, prioritize the container with the smallest remaining load margin, that is: In the formula For current goods The optimal target container number; After solving the above model, the system outputs the following control parameters: Goods allocation set: To determine the container to which each item belongs; Loading sequence: for each Generate an ordered list This determines the order in which the robot picks up the goods; Grab coordinates: The target pose of the robot end effector is calculated based on the 3D bin packing algorithm (3D-BPP).

[0039] Furthermore, in step S4, during the loading process, the system collects the pressure values ​​at four points on the bottom of the container in real time. Calculate the current centroid coordinates using the centroid formula. If the center of gravity shifts or Exceeding the preset safety factor Multiply by half the bottom dimension, the system determines that the load is abnormally off-center, suspends loading and prompts for adjustment or fine-tuning of the cargo position via robot; Furthermore, the dynamic adjustment of the work cycle in step S6 specifically includes: Monitor the dwell time of goods at each workstation using a digital twin system; When the time the goods are held up in any link exceeds the preset threshold, the system automatically generates a speed adjustment command to reduce the delivery speed of upstream goods or increase the processing priority of downstream equipment, and displays the adjusted status on the large visualization screen as an early warning.

[0040] The embodiment of the cargo diversion and container loading operation logic of the present invention is described below: The circular conveyor line 3 sequentially transports four pieces of goods to be transferred: ①, ②, ③, and ④. The goods correspond to two destinations, A and B. Containers A1 and A2 are set up for destination A, and containers B1 and B2 are set up for destination B. Goods ①, ②, ③, and ④ pass through the circular conveyor line 3 in sequence, and the DWS detection unit completes information collection. The equipment identifies the following: Goods ① destination A, weight 10kg, volume 50dm³; Goods ② destination B, weight 7kg, volume 13dm³; Goods ③ destination B, weight 1kg, volume 5dm³; Goods ④ destination A, weight 3kg, volume 7dm³. Goods ① are transported to the corresponding intensive operation station via the circular conveyor line 3. The loading robot loads goods ① into container A1 according to the pre-loading simulation instructions. After loading is completed, the automatic lifting transmission equipment determines that container A1 has reached the rated loading threshold, drives container A1 to be raised to the same height as the roller conveyor 6, and pushes container A1 to be completely transferred to the conveying surface of roller conveyor 6. After the push is completed, the automatic lifting transmission equipment resets and descends, positioning the empty container A2 at the loading station to continue to undertake subsequent goods loading operations. When cargo ② first passes through the DWS detection unit, the control system determines that the current station container B1 has only 1kg of rated capacity and cannot accommodate cargo ②. The control system then controls cargo ② to continue to be transported in a loop with the circular conveyor line 3 and will not enter the B destination loading station for the time being. It will be loaded after the empty container is switched. After cargo ③ is transported to the centralized work station at destination B, the loading robot loads cargo ③ into container B1; after loading is completed, the automatic lifting transmission equipment determines that container B1 is full, lifts and pushes container B1 to roller conveyor 6, and then resets and lowers, placing the empty container B2 in the loading station to wait for the next task. After goods ④ are transported to the intensive work station at destination A, the loading robot loads goods ④ into the empty container A2; the goods ② transported in a cycle are checked a second time by the DWS detection unit. At this time, the control system determines that container B2 has the remaining loading space to accommodate goods ②, and controls goods ② to be transported to the loading work station at destination B, where the loading robot loads goods ② into container B2. The fully loaded containers, once filled, are transported via roller conveyor 6 to the railway platform loading and unloading area 5 according to their freight destinations. The railway platform loading and unloading area 5 is equipped with AGV transfer trolleys remotely dispatched by a backend data analysis platform. The AGV transfer trolleys first transfer the fully loaded containers output from roller conveyor 6 to the platform buffer area for temporary storage. After the freight train enters the platform, the AGV transfer trolleys transfer the buffer containers to the inside of the train carriages and adjust the placement of each container in the carriages according to the loading instructions issued by the backend data analysis platform. This balances the overall load distribution in the carriages, ensures smooth train operation, and avoids the safety hazard of overturning during high-speed train operation.

[0041] This invention automatically collects multi-dimensional cargo data using DWS equipment and synchronizes it to a backend data analysis platform. This enables unified data collection and destination-specific pre-loading simulation, providing a data foundation for intelligent end-to-end processing. The backend data analysis platform feeds back the simulation results to the loading robot, allowing it to operate according to an optimized sequence. Combined with real-time four-point weighing verification and center of gravity calculation, it ensures balanced container load and reasonable loading, meeting the requirements for smooth roller conveyor operation and ensuring compliance with train carriage axle load regulations, fundamentally avoiding the risk of tipping over. Simultaneously, a digital twin system is used to construct a virtual-real synchronized three-dimensional scene. Based on queue theory, it simulates and predicts congestion nodes in real time and dynamically issues speed adjustment commands to adaptively balance the logistics rhythm of the entire station, effectively eliminating the risks of cargo accumulation and train delays. Furthermore, the use of universal roller conveyors adapted to various container sizes, such as baggage and pallets, along with robots replacing high-intensity manual labor, significantly reduces labor intensity and improves the system's flexibility and overall transfer efficiency.

[0042] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A digital twin-based unmanned transfer system for high-speed rail express stations, characterized in that: It includes a cargo transfer area (1), a cargo scanning and identification area (2), a circular conveyor line (3), an intensive operation area (4), a roller conveyor (6), and a railway platform loading and unloading operation area (5) arranged sequentially along the cargo transfer flow. It also includes a back-end data analysis platform for collecting multi-dimensional basic data of cargo and generating pre-loading simulation schemes for different destinations and issuing equipment control instructions, as well as a digital twin system for data interaction with the back-end data analysis platform and field equipment to construct a three-dimensional virtual scene of the entire station process, to map the dynamic operation data of field equipment and cargo in real time, to simulate and dynamically deduce the cargo flow operation conditions in the station, to predict congestion nodes and output conveying rhythm scheduling adjustment instructions to the back-end data analysis platform. The cargo receiving area (1) is used to receive cargo to be transferred; the cargo scanning and identification area (2) is equipped with a DWS device, which is used to synchronously collect the external dimensions, weight, and destination information of a single shipment and transmit it to the back-end data analysis platform in real time. The back-end data analysis platform is connected to the DWS detection equipment and configured with a simulated loading simulation module. The simulated loading simulation module performs pre-loading simulation calculations for different destinations based on the original cargo data, generates cargo diversion routing instructions and container loading instructions, and sends the two types of instructions to the control units corresponding to the circular conveyor line (3) and the intensive operation area (4). The railway platform loading and unloading area is equipped with AGV transfer trolleys, which are remotely dispatched and controlled by the back-end data analysis platform. The AGV transfer trolleys receive the loading instructions issued by the back-end data analysis platform, transfer and buffer the fully loaded containers output by the roller conveyor (6), and transfer them to the interior of the carriage after the train enters the station. The containers are arranged according to the pre-loading simulation scheme to balance the overall load distribution of the carriage. The digital twin system establishes data interaction channels with the backend data analysis platform and the automated equipment in various areas of the station.

2. The unmanned transfer system for digital twin loading simulation at a high-speed rail express station according to claim 1, characterized in that: The circular conveyor line (3) is a closed-loop conveyor structure used to receive goods output from the goods scanning and identification area (2) and, according to the routing instructions issued by the back-end data analysis platform, to divert and transport the goods to the corresponding destination intensive operation area (4) according to the pre-loading simulation scheme. When the backend data analysis platform determines that there is no available loading space in the destination intensive operation area (4) corresponding to the goods, it controls the circular conveyor line (3) to bypass the corresponding operation area and continue to transport the goods in a loop until the container at the destination makes room. Then, the DWS device will perform secondary identification and replan the route.

3. The unmanned transfer system for digital twin loading simulation at a high-speed rail express station according to claim 2, characterized in that: The intensive operation area (4) is set up in at least one group, and each group of intensive operation areas (4) corresponds to a single freight destination; Each intensive operation area (4) is equipped with a loading robot and an automatic lifting transmission device; the automatic lifting transmission device supports and places containers and integrates a weighing and verification device; the loading robot receives loading instructions and loads the corresponding goods into the container; the weighing and verification device collects the total load weight of the container in real time, and when the verification determines that the total weight of the container reaches the railway's preset load threshold, it drives the automatic lifting transmission device to complete the container replacement operation of pushing the fully loaded container and replacing the empty container.

4. The unmanned transfer system for digital twin loading simulation at a high-speed rail express station according to claim 3, characterized in that: The simulated loading module uses a load-volume dual-constraint judgment algorithm to perform real-time feasibility verification on each piece of cargo to be loaded. Formula for calculating the load constraint of a container: in, Number the target container, such as A1, A2, B1, B2; The serial number of the goods to be loaded; The container has been filled. The collection consisting of all the goods inside; For container Total weight of cargo currently loaded; For the actual test of DWS device Weight of goods, in kg; The maximum rated load of a single container is limited by railway express transport regulations, in kg. Formula for calculating container volume constraints: in, For container The total volume occupied by the currently loaded cargo; For the actual test of DWS device Volume of a piece of cargo, in dm³; This represents the upper limit of the effective loading volume inside the container, in dm³. Comprehensive formula for determining whether cargo can be loaded: in, Assign identification tags to the cargo; Representative goods Can be loaded into container ; This indicates that the current container is underloaded or its volume is insufficient, and the goods need to be circulated back along the circular conveyor line to wait for an empty container. In addition to determining the feasibility of a single cargo, the simulated loading module is configured to: acquire the weight and volume data of each cargo to be loaded at the same destination; optimize the total weight variance of each container by minimizing the variance; and distribute the cargo to multiple containers at the destination by combining the rated load and volume constraints of each container, so that the loading weight of each container tends to be balanced to meet the train's load-bearing requirements. Furthermore, the module dynamically adjusts the cargo arrangement based on the multi-point weighing data at the bottom of the container to optimize the internal center of gravity distribution of the container and prevent overturning during on-site transfer or train operation. The objective optimization function is: The constraints are: in, This represents the total number of containers required for the current freight destination. The average total load of all containers at that destination; For container load variance; , For any two different numbered containers; Indicates any two different numbered containers and The intersection is empty; Gather all goods to the same destination The iterative solution model outputs the cargo allocation set for each container. The system generates a cargo loading sequence and robot grasping coordinate control parameters, which are then issued as global loading commands.

5. The unmanned transfer system for digital twin loading simulation at a high-speed rail express station according to claim 4, characterized in that: The automatic lifting transmission device is configured to descend gradually as the cargo accumulates in the container during loading to maintain a constant relative height between the loading robot and the top of the cargo; when the container is fully loaded, the automatic lifting transmission device rises to a height level with the roller conveyor, pushes the fully loaded container onto the roller conveyor (6), and then descends and positions a new empty container for the next round of loading operations.

6. A high-speed rail express station digital twin loading simulation unmanned transfer system according to any one of claims 1-5, characterized in that: The digital twin system includes a 3D modeling module, a real-time data mapping module, and a simulation module. The real-time data mapping module is used to map the operating status of field equipment, cargo location, and transportation speed to the 3D model. The simulation module is used to perform bottleneck prediction and operation cycle optimization based on the mapped data.

7. A high-speed rail express station digital twin loading simulation unmanned transfer system according to any one of claims 1-5, characterized in that: It also includes a monitoring and control room, which is equipped with a large visualization screen. The visualization screen is connected to the digital twin system and is used to display the daily cargo volume, the volume of cargo waiting to be transported, the equipment utilization rate and the three-dimensional dynamic simulation screen. It also supports operators to issue manual intervention instructions based on the simulation results to adjust the operation parameters.

8. A high-speed rail express station digital twin loading simulation unmanned transfer system according to any one of claims 1-5, characterized in that: The DWS device integrates a volume measurement sensor, a weighing sensor, and a barcode / QR code recognition camera. The DWS device is set along the conveying path of the circular conveyor line and is used to collect and verify data multiple times on the goods being transported in a loop.

9. A digital twin-based unmanned transfer method for high-speed rail express stations, characterized in that... The system is implemented using the digital twin loading simulation unmanned transfer system for high-speed rail express stations as described in any one of claims 1-8, comprising: S1. Receive the goods, use DWS equipment to collect the weight, dimensions and destination information of the goods and upload them to the backend data analysis platform; S2. The back-end data analysis platform performs automatic loading simulation by destination based on the collected information, calculates the optimal loading scheme for each container, and generates corresponding loading instructions and routing instructions. S3. Control the circular conveyor line to transport goods according to the routing instructions. If the target intensive operation area is full, control the goods to circulate back on the circular conveyor line and wait for the next identification and diversion. S4. Control the loading robot to load the goods into the container according to the optimal loading scheme, and at the same time check the loading weight in real time through the weighing device. In response to the signal that the container is full, execute the operation of automatic container changing and conveying to the roller conveyor. S5. Control the AGV transfer trolley to transfer the fully loaded container on the roller conveyor to the train car and perform balanced loading within the car; S6. The entire process from steps S1 to S5 is monitored and simulated in real time using a digital twin system. The work rhythm is dynamically adjusted based on the simulation results to prevent congestion.

10. A digital twin-based unmanned transfer method for high-speed rail express stations according to claim 9, characterized in that, The automatic destination-specific loading simulation described in step S2 includes: Collect statistics on the weight and volume of all goods awaiting loading to the same destination; Based on greedy algorithms or genetic algorithms, cargo is combined and allocated to multiple containers to minimize the total weight variance of each container and ensure that the weight of a single container does not exceed the limit. Generate the cargo loading sequence for each container and the corresponding robot grasping coordinates; Step S6 involves dynamically adjusting the work cycle time, including: Monitor the dwell time of goods at each workstation using a digital twin system; When the time the goods are held up in any link exceeds the preset threshold, the system automatically generates a speed adjustment command to reduce the delivery speed of upstream goods or increase the processing priority of downstream equipment, and displays the adjusted status on the large visualization screen as an early warning.