CN-122018360-A - Port bank bridge-field bridge-AGV integrated scheduling system and method

CN122018360ACN 122018360 ACN122018360 ACN 122018360ACN-122018360-A

Abstract

The invention relates to a port shore bridge-field bridge-AGV integrated dispatching system and a method, the system comprises a full-element sensing and communication module, a unified task modeling and decomposing module, a collaborative scheduling optimization engine, an elastic scheduling layer, a digital twin simulation sandbox module, a conflict prediction and dynamic resolving module and an instruction distribution and execution monitoring module. The scheduling method comprises the following steps of system initialization and dynamic environment sensing, rolling time domain optimization triggering and task set determination, integrated collaborative scheduling scheme generation and verification, scheduling instruction issuing and online dynamic fine adjustment, and executing process monitoring and dynamic rescheduling. According to the invention, by constructing a unified space-time decision model and fusing digital twin verification, global collaborative optimization of production tasks, logistics resources and energy supply is realized, so that the ship berthing time is obviously shortened, and the comprehensive utilization rate of equipment and the capability of the system for coping with uncertainty are improved.

Inventors

ZHANG YU
YANG YANG
SONG ZHIGANG
ZHANG LICHUN
WU FAN

Assignees

中交机电工程局有限公司

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (10)

1. The port shore bridge-field bridge-AGV integrated dispatching system is deployed in a central control server of a wharf and is connected with wharf equipment through an industrial Internet of things, and is characterized by comprising, The full-element sensing and communication module is used for collecting equipment state information of a shore bridge, a field bridge and an AGV, container information and an operation plan from an upper system in real time and realizing bidirectional low-delay transmission of control instructions and state feedback; The unified task modeling and decomposing module is used for receiving the macroscopic operation plan and decomposing the macroscopic operation plan into atomic task units, wherein the atomic task units define the complete operation of transporting a specific container from one appointed resource point to another appointed resource point and are associated with time windows, priorities and task type attributes; The collaborative scheduling optimization engine is used for constructing a unified elastic space-time network model based on the real-time state provided by the full-element sensing and communication module and the atomic task set provided by the unified task modeling and decomposing module, establishing a multi-objective optimization model taking the minimum ship berthing time as a main objective and considering AGV total energy consumption and equipment load balance, solving by adopting an intelligent optimization algorithm, and outputting an integrated collaborative scheduling pre-scheme; The digital twin simulation sandbox module is used for constructing a high-fidelity virtual image synchronous with the physical wharf, receiving an integrated collaborative scheduling pre-scheme output by the collaborative scheduling optimization engine, carrying out accelerated simulation deduction, evaluating key performance indexes and identifying potential space-time conflicts and risks, feeding back an evaluation result to the collaborative scheduling optimization engine for scheme calibration and iterative optimization, and forming an optimization-simulation-verification decision closed loop; The conflict prediction and dynamic resolution module is used for predicting space-time conflicts among devices in a short-time domain in the future in an online rolling manner based on a scheduling scheme in current execution and real-time device track data, and carrying out dynamic resolution by adopting a speed adjustment, path node reservation or task sequence fine adjustment strategy; The instruction distribution and execution monitoring module is used for analyzing the final scheduling scheme verified and confirmed by the digital twin simulation sandbox module into a series of executable instruction sequences oriented to specific equipment and issuing the executable instruction sequences, simultaneously monitoring the instruction execution state and the progress deviation in real time, and triggering a dynamic rescheduling process when the deviation exceeds a threshold value or a specific event occurs.
2. The port bridge-field bridge-AGV integrated scheduling system of claim 1, wherein the co-scheduling optimization engine specifically comprises: The system comprises a space-time resource network modeling unit, a container transportation task and a control unit, wherein the space-time resource network modeling unit is used for discretizing a wharf physical area into space nodes and combining discretized time slices to construct a unified space-time network diagram, wherein the network nodes represent the space position and the operation state of equipment at specific moments, and the network sides represent the actions executed by the equipment in different space nodes, including driving, loading, unloading, waiting and charging; The multi-objective robust optimization model unit is used for establishing a mixed integer planning model, wherein an objective function at least comprises the steps of minimizing the maximum finishing time in all service ships, minimizing the total running energy consumption of all AGVs, maximizing the load balance between a shore bridge and a field bridge, limiting the quantity of AGVs waiting at the same time below by setting buffer zone capacity constraint for each shore bridge and the field bridge and limiting the capacity constraint of the task logic foundation constraint, modeling the battery electric quantity consumption and the charging requirement of the AGVs as special tasks capable of being inserted into a scheduling sequence, realizing the cooperation of production and energy supply, introducing robust constraint for key operation time length, and generating a scheduling scheme insensitive to uncertainty; The intelligent optimization solving unit is used for carrying out high-efficiency solving on the multi-objective robust optimization model by adopting a co-evolution algorithm based on space-time chromosome coding and reinforcement learning guidance, maintaining two co-evolved sub-populations of resource allocation and traffic scheduling, respectively optimizing a shore bridge/field bridge operation sequence, AGV path planning and fleet allocation, carrying out co-evolution among the sub-populations through exchanging excellent space-time scheduling fragments, and embedding a variable neighborhood search strategy guided by reinforcement learning agents to carry out local deep optimization.
3. The port bridge-field bridge-AGV integrated scheduling system according to claim 2 wherein the unified elastic spatio-temporal network model is elastic, the granularity of the time slices can be dynamically adjusted according to the scheduling stage and the optimization accuracy requirement, and a coarser time granularity is adopted in the off-line macroscopic planning stage and a finer time granularity is adopted in the on-line rolling optimization stage.
4. The port shore bridge-field bridge-AGV integrated scheduling system according to claim 2, wherein in the intelligent optimization solving unit, the reinforcement learning Agent dynamically adjusts the neighborhood structure combination and the calling sequence adopted in the variable neighborhood searching strategy by continuously learning historical scheduling data and simulation feedback so as to adaptively improve the convergence speed and the solution quality of the algorithm under different working scenes.
5. The port shore bridge-field bridge-AGV integrated scheduling system according to claim 1 wherein the digital twin simulation sandbox module uses parallel computing and event driven simulation techniques to deduce the scheduling scheme faster than actual operation, and wherein a key conflict rule base is embedded in the module to automatically identify potential deadlock, congestion and equipment interference risks.
6. The integrated port bridge-field bridge-AGV scheduling system of claim 5 wherein the digital twin simulation sandbox module employs a federal twin architecture comprising a high fidelity core sandbox for final solution verification and multiple lightweight fast sandboxes for solution prescreening and fast evaluation inside the optimization engine to balance simulation accuracy and decision efficiency.
7. The port bridge-field bridge-AGV integrated scheduling system according to claim 1, further comprising an elastic scheduling layer, located between the cooperative scheduling optimization engine and the instruction distribution and execution monitoring module, configured to receive the scheduling scheme verified by simulation, and perform small elastic expansion and contraction on an execution time window of a part of non-critical tasks in the scheme according to a real-time perceived system load and a device health status, so as to further absorb micro-disturbance and enhance execution flexibility.
8. Scheduling method based on a port quay bridge-yard bridge-AGV integrated scheduling system according to any of claims 1-7, characterized by the following steps: s1, system initialization and dynamic environment sensing: loading a digital map of a wharf and an equipment parameter library, and continuously acquiring real-time states and to-be-processed job task lists of all controllable equipment through a full-element sensing and communication module; s2, triggering the rolling time domain optimization and determining a task set: starting a new round of optimization according to a preset fixed time period or triggered by equipment faults, serious task progress deviation and newly added emergency task events; S3, generating and verifying an integrated cooperative scheduling scheme: S31, calling a collaborative scheduling optimization engine, running an intelligent optimization algorithm based on the current wharf state and an atomic task set to be scheduled, and solving a multi-objective optimization model to obtain a preliminary integrated collaborative scheduling pre-scheme; S32, inputting a preliminary integrated collaborative scheduling pre-scheme into a digital twin simulation sandbox module, performing multi-speed acceleration simulation, evaluating theoretical performance indexes, and detecting potential hidden conflicts and execution risks in the scheme; s33, making a decision according to a simulation evaluation result, namely if the result meets a preset performance and robustness threshold, confirming the result as a final executable scheme, if the result does not meet the preset performance and robustness threshold, feeding bottleneck information and risk points identified by simulation back to a collaborative scheduling optimization engine, and after model parameters or constraint weights are adjusted, jumping to the step S31 for iterative optimization until a satisfactory scheme is obtained; S4, dispatching instruction issuing and online dynamic fine adjustment: The conflict prediction and dynamic resolution module continuously performs conflict prediction and online fine adjustment in the subsequent execution process based on the scheme and the real-time track; s5, monitoring an execution process and dynamically rescheduling: And (2) monitoring the actual execution progress of the scheme by the system, and triggering dynamic rescheduling when a new fixed optimization period is reached or when key equipment faults, accumulated task deviation exceeding a threshold value and high-priority new tasks being inserted are monitored, returning to the step (S2).
9. The method according to claim 8, wherein in step S2, the conditions and levels of event triggering are predefined, and different levels of event trigger dynamic rescheduling with different response speeds and optimization ranges.
10. The method for dispatching the port bank bridge-field bridge-AGV integrated dispatching system according to claim 9 is characterized in that in step S4, an online dynamic fine tuning strategy is based on the principle that speed adjustment is main, path nodes are about to be auxiliary, a smooth speed adjusting strategy is adopted when sufficient buffer time and space exist between equipment and potential conflict points, and when the conflict risk is high, a reservation mechanism for waiting resources at key intersections and loading and unloading positions is started, and the conflict is avoided by controlling the arrival time sequence of the equipment.

Description

Port bank bridge-field bridge-AGV integrated scheduling system and method Technical Field The invention relates to the technical field of port automation and intelligent scheduling, in particular to a port shore bridge-field bridge-AGV integrated scheduling system and method. Background The automatic container terminal is a core development direction for improving port operation efficiency and reducing labor cost. The core operation flow of the method relates to three key devices, namely a shore bridge is responsible for loading and unloading containers between ships and the shore, a yard bridge is responsible for storing and taking containers in a yard, and an AGV is responsible for horizontal transportation of a wharf. The traditional dispatching mode generally adopts a layered or segmented architecture, namely, a quay bridge operation plan is firstly established, an AGV is allocated to each operation line, and finally, a yard bridge operation is allocated to the container reaching the stacking area. The decoupling scheduling has inherent defects of global optimization deficiency, inconsistent scheduling targets and information lag of each level, easiness in forming local optimum and low overall performance, serious mutual waiting among devices, delayed dynamic response, slow response of a segmentation adjustment strategy in the face of equipment faults, operation delay and other disturbance, easiness in causing operation chain breakage, passive collision avoidance, non-prospective consideration of space-time coupling relation between AGV path planning and operation of a shore bridge and a field bridge, frequent intersection deadlock and operation area congestion, insufficient resource cooperation, separation of AGV charging and buffer area management and production scheduling, and influence on continuous operation capability. In the prior art, although research attempts are made to perform joint scheduling on every two devices, the shore bridge, the field bridge and the AGV cannot be placed under a unified space-time frame to perform millisecond collaborative decision, so that global optimization of a full link from a ship-horizontal transportation-storage yard is difficult to realize, and an efficient online adjustment and verification mechanism for coping with a complex dynamic environment is lacking. Disclosure of Invention The invention aims to solve the defects of the prior art, and provides a port bank bridge-field bridge-AGV integrated scheduling system and a port bank bridge-field bridge-AGV integrated scheduling method. The invention adopts the following technical scheme to realize the aim: The port bank bridge-field bridge-AGV integrated dispatching system is deployed in a central control server of a wharf and is connected with wharf equipment through an industrial Internet of things, The full-element sensing and communication module is used for collecting equipment state information of a shore bridge, a field bridge and an AGV, container information and an operation plan from an upper system in real time and realizing bidirectional low-delay transmission of control instructions and state feedback; The unified task modeling and decomposing module is used for receiving the macroscopic operation plan and decomposing the macroscopic operation plan into atomic task units, wherein the atomic task units define the complete operation of transporting a specific container from one appointed resource point to another appointed resource point and are associated with time windows, priorities and task type attributes; The collaborative scheduling optimization engine is used for constructing a unified elastic space-time network model based on the real-time state provided by the full-element sensing and communication module and the atomic task set provided by the unified task modeling and decomposing module, establishing a multi-objective optimization model taking the minimum ship berthing time as a main objective and considering AGV total energy consumption and equipment load balance, solving by adopting an intelligent optimization algorithm, and outputting an integrated collaborative scheduling pre-scheme; The digital twin simulation sandbox module is used for constructing a high-fidelity virtual image synchronous with the physical wharf, receiving an integrated collaborative scheduling pre-scheme output by the collaborative scheduling optimization engine, carrying out accelerated simulation deduction, evaluating key performance indexes and identifying potential space-time conflicts and risks, feeding back an evaluation result to the collaborative scheduling optimization engine for scheme calibration and iterative optimization, and forming an optimization-simulation-verification decision closed loop; The conflict prediction and dynamic resolution module is used for predicting space-time conflicts among devices in a short-time domain in the future in an online rolling manner based on a scheduling scheme in c