Full-automatic multi-machine linkage system and method for port ship unloaders

Abstract

The invention relates to the technical field of ship unloader linkage, in particular to a port ship unloader full-automatic multi-machine linkage system and a method thereof.The port ship unloader full-automatic multi-machine linkage system comprises a distributed intelligent scheduling module, the distributed intelligent scheduling module is in coupling connection with a full-link data middle station, and the full-link data middle station is in coupling connection with a self-adaptive edge computer unit; the self-adaptive edge computer unit is in coupling connection with a three-dimensional perception safety monitoring network, the three-dimensional perception safety monitoring network is in coupling connection with a man-machine collaborative interaction platform, and the man-machine collaborative interaction platform is in coupling connection with a production management module. The port ship unloader full-automatic multi-machine linkage system and the method thereof solve the problems that in the prior art, a ship unloader full-automatic multi-machine linkage system is mostly controlled in a centralized mode in the actual using process, depends on a fixed algorithm and is difficult to dynamically deal with complex operation scenes (such as ship position deviation and uneven cargo distribution), and the working efficiency is high. And equipment conflict or idle waiting is caused.

Description

Technical Field

[0001] The present invention relates to the technical field of ship unloader linkage, and in particular to a port ship unloader fully automatic multi-machine linkage system and method thereof. Background Art

[0002] The fully automatic multi-machine linkage system of ship unloader is an advanced port loading and unloading equipment control system, which can realize the coordinated operation of multiple ship unloaders and improve the efficiency and safety of ship unloading operations. The fully automatic multi-machine linkage system of ship unloaders integrates advanced sensor technology, communication technology, artificial intelligence algorithm and automatic control technology to realize the fully automatic, efficient and coordinated operation of multiple ship unloaders. The system can monitor the ship's position, cargo distribution, ship unloader status and other information in real time, and dynamically adjust the operation plan and optimize the operation process based on this information, thereby improving the efficiency and safety of ship unloading operations. The fully automatic multi-machine linkage system of ship unloader is an advanced port loading and unloading equipment control system, which has the advantages of improving operation efficiency, reducing operating costs, enhancing system stability and enhancing port competitiveness. With the continuous advancement of technology and the continuous expansion of application scenarios, the system will play a more important role in future port loading and unloading operations.

[0003] However, in actual use, the fully automatic multi-machine linkage system of the ship unloader in the prior art is mostly centralized control and relies on fixed algorithms. It is difficult to dynamically respond to complex operating scenarios (such as ship position offset, uneven distribution of cargo), resulting in equipment conflicts or idle waiting and low unloading efficiency. To address the above problems, a fully automatic multi-machine linkage system and method for a port ship unloader are provided. Summary of the invention

[0004] The purpose of the present invention is to provide a fully automatic multi-machine linkage system for a port ship unloader and a method thereof, so as to solve the problems raised in the above-mentioned background technology. To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a fully automatic multi-machine linkage system for a port ship unloader, comprising a distributed intelligent scheduling module, the distributed intelligent scheduling module is coupled and connected to a full-link data middle platform, the full-link data middle platform is coupled and connected to an adaptive edge computer unit, the adaptive edge computer unit is coupled and connected to a three-dimensional perception safety monitoring network, the three-dimensional perception safety monitoring network is coupled and connected to a human-machine collaborative interaction platform, and the human-machine collaborative interaction platform is coupled and connected to a production management module.

[0005] Preferably, the distributed intelligent scheduling module uses a dynamic task allocation algorithm to perform operations, and constructs a state-action value function Q(s, a) based on a Q-learning algorithm of reinforcement learning, where s includes the location of the device, the task queue, and the environmental parameters, and action a is a task allocation decision;

[0006] The objective function is to minimize the total operation time T and the equipment idle time I:

[0007] MinimizeαT+βI;

[0008] Among them, α and β are weight coefficients, which are dynamically adjusted through training;

[0009] Update Q value in real time:

[0010]

[0011] Where η is the learning rate, γ is the discount factor, and r is the immediate reward.

[0012] Preferably, the distributed intelligent scheduling module adopts a priority queue-jumping mechanism. When the material distribution is uneven or the task cannot be completed due to delay, the system reallocates the task according to the task priority and the status of the adjacent equipment. The priority calculation is based on the task urgency U and the equipment complexity L:

[0013] P=ω 1 U+ω 2 L;

[0014] Among them, ω 1 and ω 2 is the weight coefficient.

[0015] According to claim 1, a fully automatic multi-machine linkage system for port ship unloaders is characterized in that: the full-link data center uses InfluxDB to store device operation data, and predicts equipment failures and operation bottlenecks through a time series prediction model (such as LSTM);

[0016] The input of the LSTM model is the historical data sequence X = {x 1 ,x 2 ,...,x t}, the output is the future state yt+1:

[0017] yt+1=LSTM(X).

[0018] Preferably, the adaptive edge computer unit adjusts the grab bucket motion based on a PID controller, and the objective function is to minimize the swing angle θ:

[0019]

[0020] Where, e(t) is the error signal, K P , K i , K d are the proportional, integral, and differential coefficients.

[0021] Preferably, the three-dimensional perception security monitoring network uses Kalman filtering to fuse multi-sensor data to estimate the device position and motion state:

[0022]

[0023] in, is the state estimation, P k is the error covariance matrix, Q k is the process noise.

[0024] Preferably, the human-computer interaction platform provides an AR visualization interface, which intervenes in the automation process through gestures and voice commands, and automatically switches to manual assisted operation when the system detects an abnormality.

[0025] Preferably, the gesture recognition is based on a convolutional neural network to recognize the operation gesture and output a control instruction:

[0026] y = CNN(X);

[0027] Among them, X is the gesture image data, and y is the instruction category.

[0028] Preferably, the production management module optimizes energy allocation based on linear programming:

[0029]

[0030] Among them, c i is the energy cost, x i Assign a variable to energy.

[0031] A fully automatic multi-machine linkage method for port ship unloaders comprises the following steps:

[0032] S1. Task initialization: After the ship docks, a laser scanner is used to generate a three-dimensional point cloud of cargo distribution. The AI ​​algorithm automatically plans the unloading sequence. The data center synchronizes wind speed and tidal environment data to the scheduling module and adjusts the robot arm operation parameters.

[0033] S2, multi-machine collaborative operation. After the lifting mechanism grabs the goods, the edge computing unit calculates the optimal throwing trajectory based on the real-time position, and the conveyor belt synchronously adjusts the speed to match the rhythm of receiving goods. If a robotic arm fails, the learning algorithm immediately reallocates the task to the adjacent equipment and marks the fault point for the maintenance module to handle.

[0034] S3, dynamic safety monitoring, millimeter wave radar continuously monitors the movement trajectory of the robot arm. If the deviation from the preset path exceeds the threshold, it triggers an emergency stop and starts the AR auxiliary interface to guide manual intervention. The production management module optimizes the motor power in real time and stores the surplus energy in the supercapacitor for use during peak hours;

[0035] S4, data closed-loop optimization, the day's operation data is uploaded to the digital twin platform, AI analyzes equipment utilization and energy consumption curves, and generates a scheduling plan for the next day.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. After the ship berths, a laser scanner is used to generate a three-dimensional point cloud of cargo distribution. The AI ​​algorithm automatically plans the unloading sequence. The data center synchronizes wind speed and tidal environment data to the scheduling module, adjusts the operating parameters of the robotic arm, and after the lifting mechanism grabs the cargo, the edge computing unit calculates the optimal throwing trajectory according to the real-time position. The conveyor belt synchronously adjusts the speed to match the receiving rhythm. If a robotic arm fails, the learning algorithm immediately reallocates the task to the adjacent equipment and marks the fault point for the maintenance module to handle. The millimeter-wave radar continuously monitors the motion trajectory of the robotic arm. If the deviation from the preset path exceeds the threshold, an emergency stop is triggered and the AR auxiliary interface is started to guide manual intervention. The production management module optimizes the motor power in real time and stores the surplus energy in the supercapacitor for use during peak hours. The day's operation data is uploaded to the digital twin platform. AI analyzes the equipment utilization rate and energy consumption curve to generate a scheduling plan for the next day. This solves the problem that the existing technology of the fully automatic multi-machine linkage system of the ship unloader is mostly centralized in actual use and relies on fixed algorithms. It is difficult to dynamically respond to complex operating scenarios (such as ship position offset and uneven cargo distribution), resulting in equipment conflicts or idle waiting, and low unloading efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a system block diagram of a fully automatic multi-machine linkage system and method for a port ship unloader of the present invention;

[0039] Figure 2 This is a flow chart of a fully automatic multi-machine linkage system and method for a port ship unloader according to the present invention;. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technical personnel in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1 to Figure 2The present invention provides a technical solution: a fully automatic multi-machine linkage system for port ship unloaders, including a distributed intelligent scheduling module, which allocates tasks to each device (such as a robotic arm, a conveyor belt) in real time, dynamically adjusts the operation sequence and path, avoids equipment conflicts or idle waiting, supports multi-device collaborative operations, and optimizes the overall loading and unloading efficiency. The distributed intelligent scheduling module is coupled with a full-link data center, which integrates equipment operation data, environmental data, and cargo information, provides a unified data storage, analysis, and visualization platform, supports real-time data synchronization and historical data analysis, and optimizes the operation process. The full-link data center is coupled with an adaptive edge computer unit, which is deployed locally on each loading and unloading equipment to process real-time control instructions (such as take-off and landing speed adjustment), reduce cloud dependence, and support local decision-making (such as grab defense Swing control) to improve response speed, the adaptive edge computer unit is coupled with a three-dimensional perception safety monitoring network, which integrates millimeter-wave radar, infrared thermal imaging and high-frame rate cameras to build a three-dimensional perception network for the operating area, identify potential risks in real time (such as robot arm collision, cargo slippage), and trigger graded alarms. The three-dimensional perception safety monitoring network is coupled with a human-machine collaborative interaction platform to provide an AR visualization interface. Operators can intervene in the automation process through gestures or voice commands, and support "semi-automatic takeover mode". When the system detects an abnormality, it automatically switches to manual assisted operation. The human-machine collaborative interaction platform is coupled with a production management module to monitor the energy consumption of the multi-drive system in real time, optimize the energy feedback grid strategy, and use supercapacitors and lithium batteries for hybrid energy storage to smooth instantaneous load fluctuations.

[0042] In this embodiment, the distributed intelligent scheduling module uses a dynamic task allocation algorithm for calculation. The dynamic task allocation algorithm is one of the core modules of the fully automatic multi-machine linkage system of the port ship unloader. It aims to allocate tasks to various devices (such as robotic arms, conveyor belts) in real time, and dynamically adjust the operation sequence and path according to environmental changes (such as ship position offset, uneven distribution of cargo), avoid equipment conflicts or idle waiting, thereby maximizing the overall loading and unloading efficiency, ensuring that all tasks are completed in the shortest time, avoiding idleness or waiting of equipment due to uneven task allocation, and responding to environmental changes (such as wind speed, tide, equipment failure) in real time. The dynamic task allocation algorithm solves the limitations of traditional centralized control systems through reinforcement learning and priority queue-jumping mechanisms, and significantly improves the collaborative efficiency and adaptability of the port ship unloader multi-machine linkage system. Its dynamic adaptability and high efficiency make it an ideal solution for complex operating scenarios. The state-action value function Q(s, a) is constructed based on the Q-learning algorithm based on reinforcement learning, where s includes the location of the equipment, the task queue, and the environmental parameters, and action a is the task allocation decision;

[0043] The objective function is to minimize the total operation time T and the equipment idle time I:

[0044] MinimizeαT+βI;

[0045] Among them, α and β are weight coefficients, which are dynamically adjusted through training;

[0046] Update Q value in real time:

[0047]

[0048] Where η is the learning rate, γ is the discount factor, and r is the immediate reward.

[0049] In this embodiment, the distributed intelligent scheduling module adopts a priority queue-jumping mechanism, which is an important part of the dynamic task allocation algorithm and is intended to solve the problem of decreased operating efficiency caused by equipment failure, task delays or environmental changes. By calculating task priority and equipment load in real time, the system can dynamically adjust task allocation to ensure that high-priority tasks are completed first, while maximizing equipment utilization and overall operating efficiency, ensuring that high-priority tasks (such as urgent goods, time-sensitive tasks) are completed first, avoiding equipment overload or idleness caused by uneven task allocation, responding to equipment failure, task delays or environmental changes in real time, ensuring system robustness, ensuring that high-priority tasks are completed first through priority calculation and task redistribution, improving overall operating efficiency, and dynamically adjusting task allocation to ensure stable system operation when equipment fails or is delayed, avoiding equipment overload or idleness, and maximizing equipment utilization. The priority queue-jumping mechanism calculates task priority and equipment load in real time, dynamically adjusts task allocation, and solves the problem of decreased operating efficiency caused by equipment failure, task delays or environmental changes. Its high efficiency, robustness and load balancing capabilities make it one of the core components of the fully automatic multi-machine linkage system of port ship unloaders. When materials are unevenly distributed or delayed and unable to complete the task, the system reallocates tasks according to task priority and the status of adjacent equipment. The priority calculation is based on the task urgency U and the equipment complexity L:

[0050] P=ω 1 U+ω 2 L;

[0051] Among them, ω 1 and ω 2 is the weight coefficient.

[0052] In this embodiment, the full-link data middle station uses InfluxDB to store device operation data. InfluxDB is a high-performance time series database specifically used for storing, querying and analyzing timestamp data. In the fully automatic multi-machine linkage system of the port ship unloader, InfluxDB is used as the core data storage platform to record and analyze equipment operation data (such as motor power, grab speed), environmental data (such as wind speed, tide) and cargo information (such as weight, size). It is designed for timestamp data and supports efficient writing and querying. Data is stored by time partitions, which is suitable for high-frequency data collection scenarios. It supports writing and querying millions of data points per second, has a built-in compression algorithm, reduces storage space usage, adopts the "measurement-tag-field" data model, supports multi-dimensional data storage and query, supports complex time series data analysis, such as aggregation, filtering, connection, etc., supports cluster deployment, meets large-scale data storage requirements, and provides high availability and data redundancy mechanisms. As a high-performance time series database, InfluxDB plays an important role in the fully automatic multi-machine linkage system of the port ship unloader. Its efficient storage, flexible query and powerful expansion capabilities make it an ideal storage platform for equipment operation data, environmental data and cargo information. By properly designing data models and query strategies, InfluxDB can significantly improve the system's data management capabilities and analysis efficiency, and predict equipment failures and operational bottlenecks through time series prediction models (such as LSTM);

[0053] The input of the LSTM model is the historical data sequence X = {x 1 ,x 2 ,...,x t}, the output is the future state yt+1:

[0054] yt+1=LSTM(X).

[0055] In this embodiment, the adaptive edge computer unit adjusts the grab bucket movement based on the PID controller, which is a feedback control algorithm widely used in industrial control systems. It adjusts the control quantity (such as motor speed, valve opening) to make the system output (such as temperature, speed) reach the expected value (set point). In the fully automatic multi-machine linkage system of the port unloader, the PID controller is often used in scenarios such as grab bucket anti-sway control and motor speed adjustment. The objective function is to minimize the swing angle θ:

[0056]

[0057] Where, e(t) is the error signal, K P , K i , K d are the proportional, integral, and differential coefficients.

[0058] In this embodiment, the three-dimensional perception security monitoring network uses Kalman filtering to fuse multi-sensor data to estimate the device position and motion state:

[0059]

[0060] in, is the state estimation, P k is the error covariance matrix, Q k is the process noise.

[0061] In this embodiment, the human-computer interaction platform provides an AR visualization interface, which intervenes in the automation process through gestures and voice commands. When the system detects an abnormality, it automatically switches to manual assisted operation.

[0062] In this embodiment, gesture recognition is based on convolutional neural network to recognize operation gestures and output control instructions:

[0063] y = CNN(X);

[0064] Among them, X is the gesture image data, and y is the instruction category.

[0065] In this embodiment, the production management module optimizes energy allocation based on linear programming:

[0066]

[0067] Among them, c i is the energy cost, x i Assign a variable to energy.

[0068] A fully automatic multi-machine linkage method for port ship unloaders comprises the following steps:

[0069] S1. Task initialization: After the ship docks, a laser scanner is used to generate a three-dimensional point cloud of cargo distribution. The AI ​​algorithm automatically plans the unloading sequence. The data center synchronizes wind speed and tidal environment data to the scheduling module and adjusts the robot arm operation parameters.

[0070] S2, multi-machine collaborative operation. After the lifting mechanism grabs the goods, the edge computing unit calculates the optimal throwing trajectory based on the real-time position, and the conveyor belt synchronously adjusts the speed to match the rhythm of receiving goods. If a robotic arm fails, the learning algorithm immediately reallocates the task to the adjacent equipment and marks the fault point for the maintenance module to handle.

[0071] S3, dynamic safety monitoring, millimeter wave radar continuously monitors the movement trajectory of the robot arm. If the deviation from the preset path exceeds the threshold, it triggers an emergency stop and starts the AR auxiliary interface to guide manual intervention. The production management module optimizes the motor power in real time and stores the surplus energy in the supercapacitor for use during peak hours;

[0072] S4, data closed-loop optimization, the day's operation data is uploaded to the digital twin platform, AI analyzes equipment utilization and energy consumption curves, and generates a scheduling plan for the next day.

[0073] The above shows and describes the basic principles, main features and advantages of the present invention. Technical personnel in this industry should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A fully automatic multi-machine linkage system for port ship unloaders, characterized in that: It includes a distributed intelligent scheduling module, which is coupled to a full-link data middle platform, which is coupled to an adaptive edge computer unit, which is coupled to a three-dimensional perception safety monitoring network, which is coupled to a human-machine collaborative interaction platform, and which is coupled to a production management module.

2. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1 is characterized in that: The distributed intelligent scheduling module uses a dynamic task allocation algorithm to perform operations and constructs a state-action value function Q(s, a) based on the Q-learning algorithm of reinforcement learning, where s includes the location of the device, the task queue, and the environmental parameters, and action a is the task allocation decision; The objective function is to minimize the total operation time T and the equipment idle time I: MinimizeαT+βI; Among them, α and β are weight coefficients, which are dynamically adjusted through training; Update Q value in real time: Where η is the learning rate, γ is the discount factor, and r is the immediate reward.

3. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1 is characterized in that: The distributed intelligent scheduling module adopts a priority queue-jumping mechanism. When materials are unevenly distributed or delayed and the task cannot be completed, the system reallocates the task according to the task priority and the status of the adjacent equipment. The priority calculation is based on the task urgency U and the equipment complexity L: P = ω1U + ω2L; Among them, ω1 and ω2 are weight coefficients.

4. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1 is characterized in that: The full-link data center uses InfluxDB to store device operation data and predicts device failures and operation bottlenecks through time series prediction models (such as LSTM); The input of the LSTM model is the historical data sequence X = {x1, x2, ..., x t }, the output is the future state yt+1: yt+1=LSTM(X).

5. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1 is characterized in that: The adaptive edge computer unit regulates the grab bucket motion based on a PID controller, and the objective function is to minimize the swing angle θ: Where, e(t) is the error signal, K P , K i , K d are the proportional, integral, and differential coefficients.

6. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1 is characterized in that: The three-dimensional perception security monitoring network uses Kalman filtering to fuse multi-sensor data to estimate the device position and motion state: in, is the state estimation, P k is the error covariance matrix, Q k is the process noise.

7. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1 is characterized in that: The human-computer interaction platform provides an AR visualization interface, intervenes in the automation process through gestures and voice commands, and automatically switches to manual assisted operation when the system detects an abnormality.

8. The fully automatic multi-machine linkage system of a port ship unloader according to claim 7 is characterized in that: The gesture recognition is based on convolutional neural network to recognize the operation gesture and output the control instruction: y = CNN(X); Among them, X is the gesture image data, and y is the instruction category.

9. The fully automatic multi-machine linkage system of a port ship unloader according to claim 1, characterized in that: The production management module optimizes energy distribution based on linear programming: Among them, c i is the energy cost, x i Assign a variable to energy.

10. A fully automatic multi-machine linkage method for port ship unloaders, characterized in that: The steps include: S1. Task initialization: After the ship docks, a laser scanner is used to generate a three-dimensional point cloud of cargo distribution. The AI ​​algorithm automatically plans the unloading sequence. The data center synchronizes wind speed and tidal environment data to the scheduling module to adjust the robot arm operation parameters. S2, multi-machine collaborative operation. After the lifting mechanism grabs the goods, the edge computing unit calculates the optimal throwing trajectory based on the real-time position, and the conveyor belt synchronously adjusts the speed to match the rhythm of receiving goods. If a robotic arm fails, the learning algorithm immediately reallocates the task to the adjacent equipment and marks the fault point for the maintenance module to handle; S3, dynamic safety monitoring, millimeter wave radar continuously monitors the movement trajectory of the robot arm. If the deviation from the preset path exceeds the threshold, it triggers an emergency stop and starts the AR auxiliary interface to guide manual intervention. The production management module optimizes the motor power in real time and stores the surplus energy in the supercapacitor for use during peak hours; S4, data closed-loop optimization, the day's operation data is uploaded to the digital twin platform, AI analyzes equipment utilization and energy consumption curves, and generates a scheduling plan for the next day.

Images