Railway container management system

The railway container management system addresses inefficiencies and safety risks through AI-driven automation and real-time data sharing, enhancing recognition accuracy and operational efficiency while reducing labor and environmental impact.

CN119067516APending Publication Date: 2024-12-03FANGCHENGGANG BEIBU GULF PORT CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411001597.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Railway container management relies heavily on manual operations, leading to low efficiency, high labor costs, safety risks, and environmental pollution, with existing systems lacking in equipment failure prevention, resource allocation, and safety management.

Method used

A railway container management system utilizing AI machine vision, Beidou positioning, and machine learning for automatic identification and scheduling, combined with real-time data sharing and monitoring, to optimize operations, reduce manual intervention, and enhance safety and resource utilization.

Benefits of technology

The system improves recognition accuracy, reduces manual labor, optimizes packing and resource use, predicts equipment failures, and ensures operational safety and efficiency, while minimizing environmental impact.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067516A_ABST
    Figure CN119067516A_ABST
Patent Text Reader

Abstract

The invention discloses a railway container management system, and relates to the technical field of container management systems. The AI machine vision and machine perception module is used for automatically identifying and processing image and video contents by using a deep learning algorithm and accurately identifying box number and vehicle number information; inner trailers and outer trailers are classified and recognized through a specific training model, and the recognition rate is increased; and the data sharing and interaction module is used for transmitting position relation data among the crane, the train card and the container lorry in real time by combining a Beidou positioning technology, feeding back the position relation data to an operator in time through message notification, and enhancing the visualization and cognition of an operation scene. According to the method, automatic processing and analysis of image and video contents are realized through AI machine vision and perception technologies, manual intervention is greatly reduced, and the recognition accuracy is improved; by means of the Beidou positioning technology and a high-speed network, real-time sharing of position data among a crane, a train card and a container truck is achieved, and visualization of an operation scene is enhanced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of container management systems, and in particular to a railway container management system. Background Art

[0002] In today's logistics and freight fields, railway container management still relies heavily on manual operations and faces many challenges, such as low operating efficiency, high labor costs, safety risks, and environmental pollution. Traditional management methods have obvious limitations in data processing, resource allocation, and safety management.

[0003] After searching, the application scheme of Chinese patent application No. 201910435879.5 discloses a field management system applied to container terminals, which involves the field of terminal management and solves the problem of inconvenient information management of container terminals; its technical scheme includes: terminal operation preset unit, which is preset with terminal operation preset information of unit time and work shift obtained through the information control center equipment; terminal personnel management unit, which stores the terminal personnel management information obtained through the information control center equipment; terminal site management unit, which stores the terminal site management information entered by the patrol personnel through the terminal equipment; the information control center equipment stores the terminal operation preset information, terminal personnel management information, terminal site management information and mechanical support management information locally and / or uploads them to the cloud server synchronously, so that the staff can query the required information through the terminal equipment. The field management system in the above patent has the following deficiencies: it is still insufficient in terms of preventing equipment failures, efficient resource allocation and safety risk management. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a railway container management system.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A railway container management system, comprising:

[0007] AI machine vision and machine perception module: Use deep learning algorithms to automatically identify and process image and video content, accurately identify container numbers and vehicle numbers; classify and identify inner and outer trailers through specific training models to improve recognition rates;

[0008] Data sharing and interaction module: Combined with Beidou positioning technology, it transmits the position relationship data between cranes, train cars and container trucks in real time, and provides timely feedback to operators through message notifications, enhancing the visualization and cognition of the operation scene;

[0009] Intelligent scheduling and operation optimization module: Based on the collected data and artificial intelligence algorithms, intelligent outbound planning is carried out to optimize the packing process and improve packing efficiency and accuracy;

[0010] Remote video automatic tallying module: It uses AI technology to upgrade the traditional tallying mode, and realizes remote automatic tallying through video recognition technology, reducing dependence on manpower;

[0011] Safety monitoring module: AI image recognition technology is used to automatically capture the status of the roof, ensure the integrity and cleanliness of the box, and improve operational safety.

[0012] Preferably: the AI ​​machine vision and machine perception module comprises:

[0013] Image preprocessing unit: performs enhancement and denoising preprocessing operations on the collected images to improve image quality;

[0014] Feature extraction unit: extracts key features of box number and vehicle number from images through deep learning algorithms;

[0015] Object detection and recognition unit: Use machine learning models to detect and accurately identify specific objects in processed images.

[0016] Preferably: the data sharing interaction module includes:

[0017] Positioning data processing unit: collects and processes data from Beidou and other positioning systems to ensure high-precision location information;

[0018] Real-time data transmission unit: Realize real-time data transmission through high-speed network to reduce delay;

[0019] User interface unit: provides an intuitive user interface, displays operation scenarios and related data, and supports user interactive operations.

[0020] Preferably: the intelligent scheduling and operation optimization module includes:

[0021] Scheduling strategy unit: formulates the optimal scheduling strategy based on current job requirements and historical data;

[0022] Operation path optimization unit: calculates the best operation path, shortens operation time and reduces energy consumption;

[0023] Job execution monitoring unit: monitors job execution in real time and adjusts scheduling strategies in time to deal with abnormal situations.

[0024] Preferably, the remote video automatic tallying module comprises:

[0025] Video analysis unit: Continuously analyzes video streams from the site to automatically identify and count goods;

[0026] Abnormal processing unit: when it is found that the goods are misplaced or damaged, it will automatically alarm and notify the operator to handle it;

[0027] Tally data management unit: store and manage tally data, and provide subsequent query and statistical analysis;

[0028] The remote video automatic tallying module is implemented through video analysis technology, using a recurrent neural network to process video frame sequences:

[0029] h t =f(W hh h t-1 +W xh x t +b h )

[0030] Among them, h t is the hidden state at time step t, x t is the input at time step t, W hh , W xh and b h They are the weights of the recurrent connections, the weights of the input to the hidden layer, and the bias term.

[0031] Preferably: the management system further comprises:

[0032] Prediction and maintenance module: Use machine learning algorithms to predict equipment failures and when maintenance is needed, thereby achieving preventive maintenance and reducing system downtime and maintenance costs;

[0033] Resource optimization module: By analyzing historical data, the optimization algorithm is used to schedule the use of forklifts and cranes to optimize resource allocation and improve resource utilization efficiency;

[0034] Environmental monitoring and response module: Integrates sensor data, monitors the working environment in real time, and responds quickly to possible risk factors through intelligent algorithms to ensure work safety.

[0035] Preferably: the prediction and maintenance module comprises:

[0036] Fault diagnosis unit: By analyzing equipment operation data, it uses machine learning algorithms to predict possible equipment failures;

[0037] Maintenance planning unit: automatically generates maintenance plans and allocates maintenance resources based on forecast results;

[0038] Effectiveness Evaluation Unit: Evaluate the effectiveness of maintenance activities and optimize maintenance strategies to improve equipment reliability;

[0039] The prediction and maintenance module uses a support vector machine to predict possible equipment failures, and its optimization objective function is:

[0040]

[0041] subject to you i ( <w,x i >+b)≥1-ξ i ,ξ i ≥0

[0042] in:

[0043] w is the normal vector in the high-dimensional space;

[0044] C is the penalty parameter for the error term;

[0045] ξt is a slack variable that allows some data points to violate the margin;

[0046] y i is the label of the data point, indicating whether it is faulty or not;

[0047] x i is the feature vector of the data point;

[0048] <w,x i > represents the dot product of w and x, which is equivalent to the linear combination result of data points in high-dimensional space.

[0049] Preferably: the resource optimization module comprises:

[0050] Resource scheduling unit: dynamically schedule forklifts and cranes based on current job requirements and resource status;

[0051] Energy consumption management unit: monitors and analyzes energy consumption and makes suggestions for energy-saving operations;

[0052] Resource Utilization Analysis Unit: Analyze resource utilization efficiency and provide decision support for subsequent procurement and resource allocation.

[0053] Preferably: the environmental monitoring and response module comprises:

[0054] Environmental data collection unit: integrates multiple sensors to collect working environment data in real time;

[0055] Risk assessment unit: conducts comprehensive analysis of the collected data and assesses the safety hazards of the working environment;

[0056] Automatic adjustment unit: Automatically adjusts operating equipment settings according to environmental conditions.

[0057] Preferably: the workflow of the management system includes the following steps:

[0058] Acquire image and video data from the scene;

[0059] S1: Use AI machine vision and machine perception technology to process data and extract box number and vehicle number information;

[0060] S2: Combine identification information with Beidou positioning data and display the operation scene in real time through the data sharing interaction module;

[0061] S3: Based on the collected data, the intelligent scheduling and operation optimization module formulates a shipping plan;

[0062] S4: The remote video automatic tallying module verifies cargo to ensure the accuracy of loading and unloading operations;

[0063] S5: The safety monitoring module continuously monitors the operation process to ensure on-site safety;

[0064] S6: The prediction and maintenance module analyzes equipment data and predicts maintenance needs;

[0065] S7: Resource optimization module manages and optimizes resources;

[0066] S8: The environmental monitoring and response module monitors environmental factors and makes corresponding adjustments.

[0067] The beneficial effects of the present invention are:

[0068] 1. The present invention realizes automatic processing and analysis of image and video content through AI machine vision and perception technology, greatly reducing manual intervention and improving recognition accuracy; it utilizes Beidou positioning technology and high-speed network to realize real-time sharing of location data between cranes, train cars and container trucks, and enhances the visualization of the operation scene.

[0069] 2. The intelligent scheduling and operation optimization module of the present invention formulates the box delivery plan through advanced algorithms, shortens the operation path, reduces energy consumption, thereby speeding up loading and unloading and improving overall operation efficiency; it adopts remote video automatic tallying to reduce the demand for on-site personnel, and automatically monitors the roof status through AI image recognition technology to ensure operation safety.

[0070] 3. The prediction and maintenance module of the present invention predicts equipment failures and reduces downtime, and the resource optimization module dynamically schedules equipment to improve utilization efficiency. Both together reduce operating costs; the environmental monitoring module collects operating environment data in real time, evaluates and adjusts operating equipment settings to adapt to different environmental conditions, and ensures operation continuity and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 The present invention provides a workflow diagram of a railway container management system. DETAILED DESCRIPTION

[0072] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.

[0073] Embodiment 1:

[0074] A railway container management system, comprising:

[0075] AI machine vision and machine perception module: This module uses deep learning algorithms to automatically identify and process image and video content, accurately identify container numbers and vehicle numbers, and classify and identify inner and outer trailers through specific training models to improve recognition rates;

[0076] Data sharing and interaction module: Combined with Beidou positioning technology, it transmits the position relationship data between cranes, train cars and container trucks in real time, and provides timely feedback to operators through message notifications, enhancing the visualization and cognition of the operation scene;

[0077] Intelligent scheduling and operation optimization module: Based on the collected data and artificial intelligence algorithms, intelligent outbound planning is carried out to optimize the packing process and improve packing efficiency and accuracy;

[0078] Remote video automatic tallying module: It uses AI technology to upgrade the traditional tallying mode, and realizes remote automatic tallying through video recognition technology, reducing dependence on manpower;

[0079] Safety monitoring module: AI image recognition technology is used to automatically capture the status of the roof, ensure the integrity and cleanliness of the box, and improve operational safety.

[0080] The workflow of the management system includes:

[0081] S1: Acquire image and video data from the scene;

[0082] S2: Use AI machine vision and machine perception technology to process data and extract box number and vehicle number information;

[0083] S3: Combine identification information with Beidou positioning data and display the operation scene in real time through the data sharing interaction module;

[0084] S4: Based on the collected data, the intelligent scheduling and operation optimization module formulates a shipping plan;

[0085] S5: The remote video automatic tallying module verifies cargo to ensure the accuracy of loading and unloading operations;

[0086] S6: The safety monitoring module continuously monitors the operation process to ensure on-site safety.

[0087] The AI ​​machine vision and machine perception module includes:

[0088] Image preprocessing unit: performs preprocessing operations such as enhancement and denoising on the collected images to improve image quality;

[0089] Feature extraction unit: extracts key features such as box number and vehicle number from images through deep learning algorithms;

[0090] Object detection and recognition unit: Use machine learning models to detect and accurately identify specific objects in processed images.

[0091] Wherein, the data sharing interaction module includes:

[0092] Positioning data processing unit: collects and processes data from Beidou and other positioning systems to ensure high-precision location information;

[0093] Real-time data transmission unit: Realize real-time data transmission through high-speed network to reduce delay;

[0094] User interface unit: provides an intuitive user interface, displays operation scenarios and related data, and supports user interactive operations.

[0095] The intelligent scheduling and operation optimization module includes:

[0096] Scheduling strategy unit: formulates the optimal scheduling strategy based on current job requirements and historical data;

[0097] Operation path optimization unit: calculates the best operation path, shortens operation time and reduces energy consumption;

[0098] Job execution monitoring unit: monitors job execution in real time and adjusts scheduling strategies in time to deal with abnormal situations.

[0099] Wherein, the remote video automatic tallying module includes:

[0100] Video analysis unit: Continuously analyzes video streams from the site to automatically identify and count goods;

[0101] Abnormal processing unit: when it is found that the goods are misplaced or damaged, it will automatically alarm and notify the operator to handle it;

[0102] Tally data management unit: stores and manages tally data, and provides subsequent query and statistical analysis.

[0103] The remote video automatic tallying module is implemented through video analysis technology, using a recurrent neural network to process video frame sequences:

[0104] h t =f(W hh h t-1 +W xh x t +bh )

[0105] Among them, h t is the hidden state at time step t, x t is the input at time step t (i.e., video frame), W hh , W xh and b h They are the weights of the recurrent connections, the weights of the input to the hidden layer, and the bias term.

[0106] Embodiment 2:

[0107] A railway container management system. This embodiment is based on the first embodiment, with the following changes:

[0108] The system includes:

[0109] AI machine vision and machine perception module: This module uses deep learning algorithms to automatically identify and process image and video content, accurately identify container numbers and vehicle numbers, and classify and identify inner and outer trailers through specific training models to improve recognition rates;

[0110] Data sharing and interaction module: Combined with Beidou positioning technology, it transmits the position relationship data between cranes, train cars and container trucks in real time, and provides timely feedback to operators through message notifications, enhancing the visualization and cognition of the operation scene;

[0111] Intelligent scheduling and operation optimization module: Based on the collected data and artificial intelligence algorithms, intelligent outbound planning is carried out to optimize the packing process and improve packing efficiency and accuracy;

[0112] Remote video automatic tallying module: It uses AI technology to upgrade the traditional tallying mode, and realizes remote automatic tallying through video recognition technology, reducing dependence on manpower;

[0113] Safety monitoring module: AI image recognition technology is used to automatically capture the status of the roof to ensure the integrity and cleanliness of the box and improve operational safety;

[0114] Prediction and maintenance module: Use machine learning algorithms to predict equipment failures and when maintenance is needed, thereby achieving preventive maintenance and reducing system downtime and maintenance costs;

[0115] Resource optimization module: By analyzing historical data, the optimization algorithm is used to schedule the use of forklifts, cranes and other equipment to optimize resource allocation and improve resource utilization efficiency;

[0116] Environmental monitoring and response module: Integrates sensor data, monitors the working environment (such as temperature, humidity, wind speed, etc.) in real time, and responds quickly to possible risk factors through intelligent algorithms to ensure work safety.

[0117] The workflow of the management system includes the following steps:

[0118] Acquire image and video data from the scene;

[0119] S1: Use AI machine vision and machine perception technology to process data and extract box number and vehicle number information;

[0120] S2: Combine identification information with Beidou positioning data and display the operation scene in real time through the data sharing interaction module;

[0121] S3: Based on the collected data, the intelligent scheduling and operation optimization module formulates a shipping plan;

[0122] S4: The remote video automatic tallying module verifies cargo to ensure the accuracy of loading and unloading operations;

[0123] S5: The safety monitoring module continuously monitors the operation process to ensure on-site safety;

[0124] S6: The prediction and maintenance module analyzes equipment data and predicts maintenance needs;

[0125] S7: Resource optimization module manages and optimizes resources;

[0126] S8: The environmental monitoring and response module monitors environmental factors and makes corresponding adjustments.

[0127] Wherein, the prediction and maintenance module includes:

[0128] Fault diagnosis unit: By analyzing equipment operation data, it uses machine learning algorithms to predict possible equipment failures;

[0129] Maintenance planning unit: automatically generates maintenance plans and allocates maintenance resources based on forecast results;

[0130] Effectiveness Evaluation Unit: Evaluate the effectiveness of maintenance activities and optimize maintenance strategies to improve equipment reliability.

[0131] The prediction and maintenance module uses a support vector machine to predict possible equipment failures, and its optimization objective function is:

[0132]

[0133] subject to you t ( <w,x i >+b)≥1-ξ i ,ξ i ≥0

[0134] in:

[0135] W is the normal vector in the high-dimensional space;

[0136] C is the penalty parameter for the error term;

[0137] ξ i is a slack variable that allows some data points to violate the margin;

[0138] y i is the label of the data point, indicating whether it is faulty or not;

[0139] x i is the feature vector of the data point;

[0140] <w,x i > represents the dot product of w and x, which is equivalent to the linear combination result of data points in high-dimensional space.

[0141] Wherein, the resource optimization module includes:

[0142] Resource scheduling unit: dynamically schedules forklifts, cranes and other equipment based on current job requirements and resource status;

[0143] Energy consumption management unit: monitors and analyzes energy consumption and makes suggestions for energy-saving operations;

[0144] Resource Utilization Analysis Unit: Analyze resource utilization efficiency and provide decision support for subsequent procurement and resource allocation.

[0145] Wherein, the environmental monitoring and response module includes:

[0146] Environmental data collection unit: integrates multiple sensors to collect working environment data in real time;

[0147] Risk assessment unit: conducts comprehensive analysis of the collected data and assesses the safety hazards of the working environment;

[0148] Automatic adjustment unit: Automatically adjusts operating equipment settings according to environmental conditions, such as adjusting crane speed, lighting, etc.

[0149] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A railway container management system, characterized in that: include: AI machine vision and machine perception module: Use deep learning algorithms to automatically identify and process image and video content, accurately identify container numbers and vehicle numbers; classify and identify inner and outer trailers through specific training models to improve recognition rates; Data sharing and interaction module: Combined with Beidou positioning technology, it transmits the position relationship data between cranes, train cars and container trucks in real time, and provides timely feedback to operators through message notifications, enhancing the visualization and cognition of the operation scene; Intelligent scheduling and operation optimization module: Based on the collected data and artificial intelligence algorithms, intelligent outbound planning is carried out to optimize the packing process and improve packing efficiency and accuracy; Remote video automatic tallying module: It uses AI technology to upgrade the traditional tallying mode, and realizes remote automatic tallying through video recognition technology, reducing dependence on manpower; Safety monitoring module: AI image recognition technology is used to automatically capture the status of the roof, ensure the integrity and cleanliness of the box, and improve operational safety.

2. A railway container management system according to claim 1, characterized in that: The AI ​​machine vision and machine perception module includes: Image preprocessing unit: performs enhancement and denoising preprocessing operations on the collected images to improve image quality; Feature extraction unit: extracts key features of box number and vehicle number from images through deep learning algorithms; Object detection and recognition unit: Use machine learning models to detect and accurately identify specific objects in processed images.

3. A railway container management system according to claim 1, characterized in that: The data sharing interaction module includes: Positioning data processing unit: collects and processes data from Beidou and other positioning systems to ensure high-precision location information; Real-time data transmission unit: Realize real-time data transmission through high-speed network to reduce delay; User interface unit: provides an intuitive user interface, displays operation scenarios and related data, and supports user interactive operations.

4. A railway container management system according to claim 1, characterized in that: The intelligent scheduling and operation optimization module includes: Scheduling strategy unit: formulates the optimal scheduling strategy based on current job requirements and historical data; Operation path optimization unit: calculates the best operation path, shortens operation time and reduces energy consumption; Job execution monitoring unit: monitors job execution in real time and adjusts scheduling strategies in time to deal with abnormal situations.

5. A railway container management system according to claim 1, characterized in that: The remote video automatic tallying module comprises: Video analysis unit: Continuously analyzes video streams from the site to automatically identify and count goods; Abnormal processing unit: when it is found that the goods are misplaced or damaged, it will automatically alarm and notify the operator to handle it; Tally data management unit: store and manage tally data, and provide subsequent query and statistical analysis; The remote video automatic tallying module is implemented through video analysis technology, using a recurrent neural network to process video frame sequences: h t =f(W hh h t-1 +W xh x t +b h ) Among them, h t is the hidden state at time step t, x t is the input at time step t, W hh , W xh and b h They are the weights of the recurrent connections, the weights of the input to the hidden layer, and the bias term.

6. A railway container management system according to claim 1, characterized in that: The management system also includes: Prediction and maintenance module: Use machine learning algorithms to predict equipment failures and when maintenance is needed, thereby achieving preventive maintenance and reducing system downtime and maintenance costs; Resource optimization module: By analyzing historical data, the optimization algorithm is used to schedule the use of forklifts and cranes to optimize resource allocation and improve resource utilization efficiency; Environmental monitoring and response module: Integrates sensor data, monitors the working environment in real time, and responds quickly to possible risk factors through intelligent algorithms to ensure work safety.

7. A railway container management system according to claim 6, characterized in that: The prediction and maintenance module includes: Fault diagnosis unit: By analyzing equipment operation data, it uses machine learning algorithms to predict possible equipment failures; Maintenance planning unit: automatically generates maintenance plans and allocates maintenance resources based on forecast results; Effectiveness Evaluation Unit: Evaluate the effectiveness of maintenance activities and optimize maintenance strategies to improve equipment reliability; The prediction and maintenance module uses a support vector machine to predict possible equipment failures, and its optimization objective function is: subject to y i (<w,x i >+b)≥1-ξ i ,ξ i ≥0 in: w is the normal vector in the high-dimensional space; C is the penalty parameter for the error term; ξ i is a slack variable that allows some data points to violate the margin; y i is the label of the data point, indicating whether it is faulty or not; x i is the feature vector of the data point; <w,x i > represents the dot product of w and x, which is equivalent to the linear combination result of data points in high-dimensional space.

8. A railway container management system according to claim 6, characterized in that: The resource optimization module includes: Resource scheduling unit: dynamically schedule forklifts and crane equipment based on current job requirements and resource status; Energy consumption management unit: monitors and analyzes energy consumption and makes suggestions for energy-saving operations; Resource Utilization Analysis Unit: Analyze resource utilization efficiency and provide decision support for subsequent procurement and resource allocation.

9. A railway container management system according to claim 6, characterized in that: The environmental monitoring and response module includes: Environmental data collection unit: integrates multiple sensors to collect working environment data in real time; Risk assessment unit: conducts comprehensive analysis of the collected data and assesses the safety hazards of the working environment; Automatic adjustment unit: Automatically adjusts operating equipment settings according to environmental conditions.

10. A railway container management system according to claim 6, characterized in that: The workflow of the management system includes the following steps: Acquire image and video data from the scene; S1: Use AI machine vision and machine perception technology to process data and extract box number and vehicle number information; S2: Combine identification information with Beidou positioning data and display the operation scene in real time through the data sharing interaction module; S3: Based on the collected data, the intelligent scheduling and operation optimization module formulates a shipping plan; S4: The remote video automatic tallying module verifies cargo to ensure the accuracy of loading and unloading operations; S5: The safety monitoring module continuously monitors the operation process to ensure on-site safety; S6: The prediction and maintenance module analyzes equipment data and predicts maintenance needs; S7: Resource optimization module manages and optimizes resources; S8: The environmental monitoring and response module monitors environmental factors and makes corresponding adjustments.

Citation Information

Patent Citations

  • A field management system for container terminals

    CN110262411B