A port loading and unloading device dynamic regulation method and system based on a job road

Abstract

The present application relates to the technical field of port scheduling, and particularly relates to a port loading and unloading equipment dynamic regulation method and system based on a work road, the method comprising the following steps: obtaining real-time state data of port loading and unloading equipment in a work road to obtain a coordination state index of the work road; combining a work environment factor to deduce a task work chain of the work road to obtain a dynamic time flow; obtaining an efficiency prediction index of the work road, generating a resource allocation regulation strategy by constructing a global resource reallocation model; converting the resource allocation regulation strategy into a device regulation instruction set for optimization to obtain a final regulation instruction sequence; executing the final regulation instruction sequence, obtaining a regulation effectiveness evaluation result, correcting the global resource reallocation model, and realizing dynamic regulation of the port loading and unloading equipment. The present application realizes active regulation of port loading and unloading equipment, effectively improves work efficiency, adaptability and intelligent level of the equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of port scheduling, and particularly relates to a port loading and unloading equipment dynamic regulation method and system based on a work road. BACKGROUND

[0002] Currently, port loading and unloading operation scheduling mainly adopts a static or segmented optimization method based on fixed rules. The prior art relies on predefined work time standards, fixed equipment efficiency parameters and relatively independent single machine scheduling logic. In specific implementation, a work instruction sequence is usually generated according to a ship stowage plan, and the planned time nodes of the entire work chain are calculated by calculating the standard time consumption of each link. The allocation and scheduling of resources are usually triggered before the task is issued or when the equipment is idle, and are performed according to simple rules such as priority and shortest distance, and the cooperation between devices mainly relies on the matching and alignment of the planned time window, and the entire scheduling process lacks state perception and real-time regulation capability.

[0003] The prior art method has some defects: first, it relies on static parameters and cannot effectively perceive and respond to real-time performance fluctuations of equipment in the work road, dynamic traffic congestion and chain delay effects between tasks, resulting in a large deviation between the planned time and the actual situation and poor fault tolerance; second, the resource scheduling and time calculation are decoupled, and the scheduling decision is not based on the real-time performance and future state prediction of the work road, which easily causes local optimization and low overall efficiency; finally, there is a lack of effective closed-loop learning and adaptive optimization mechanism, and the model parameters cannot be continuously corrected to update the strategy from the historical execution deviation, and the robustness and intelligence level are insufficient in the face of complex and variable work environment. The above defects result in that the resource utilization potential cannot be fully tapped, thereby limiting the overall work efficiency of the port loading and unloading equipment. SUMMARY

[0004] In view of the defects in the prior art, the present application provides a port loading and unloading equipment dynamic regulation method and system based on a work road.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a dynamic control method for port loading and unloading equipment based on a work route. The method includes the following steps: acquiring real-time status data of port loading and unloading equipment in the work route; obtaining a collaborative status index of the work route based on the real-time status data; deducing a dynamic time flow from the task chain of the work route based on the collaborative status index and in conjunction with operational environment factors; obtaining an efficiency prediction index of the work route based on the dynamic time flow; generating a resource allocation control strategy by constructing a global resource reallocation model; converting the resource allocation control strategy into a set of equipment control instructions; optimizing the set of equipment control instructions to obtain a final control instruction sequence; executing the final control instruction sequence; obtaining a control effectiveness evaluation result; and correcting the global resource reallocation model to achieve dynamic control of the port loading and unloading equipment. The present invention can dynamically adapt to the real-time status and environmental changes of the work route, reduce resource idleness and conflicts, significantly improve loading and unloading efficiency and control accuracy, and provide reliable technical support for the efficient operation of port operations.

[0006] Optionally, acquiring real-time status data of port loading and unloading equipment in the operation route, and obtaining a collaborative status index of the operation route based on the real-time status data, includes: acquiring the real-time status data, including the location information, speed information, and current task elapsed time of the port loading and unloading equipment; calculating collaborative status parameters based on the real-time status data, including rhythm synchronization rate, equipment health, and work load value; determining collaborative weighting coefficients; and weighting and fusing the collaborative status parameters using the collaborative weighting coefficients to obtain the collaborative status index. This invention makes the collaborative status index more closely reflect actual operational scenarios, avoids the one-sidedness of a single indicator, provides accurate data support for subsequent control decisions, and improves the scientific nature and pertinence of control schemes.

[0007] Optionally, the step of extrapolating the task chain of the work route based on the cooperative state index and in combination with the work environment factor to obtain the dynamic time flow includes: obtaining the real-time traffic density of the work route; obtaining the work environment factor based on the real-time traffic density; generating a dynamic time margin for each task stage of the work route by combining the cooperative state index and the work environment factor; and extrapolating the task chain based on the dynamic time margin to obtain the dynamic time flow. This invention generates a dynamic time margin to adapt to the characteristics of different task stages, improves the accuracy of dynamic time flow extrapolation, provides a reliable time series basis for subsequent performance prediction and resource allocation, and reduces the risk of work delays.

[0008] Optionally, the step of extrapolating the dynamic time flow based on the dynamic time margin to the task chain includes: using the start time of the first task stage in the task path as a time base; using the expected start and completion times of subsequent task stages as the task chain; and based on the time base, using the dynamic time margin to perform rolling extrapolation on the task chain to obtain a sequence of task time nodes, and using the sequence of task time nodes as the dynamic time flow. This invention clearly defines the temporal relationship of each task stage, facilitating the early prediction of task bottlenecks, reserving buffer space for equipment scheduling and task adjustment, and improving task continuity.

[0009] Optionally, the step of obtaining the efficiency prediction index of the work route based on the dynamic time flow and generating a resource allocation control strategy by constructing a global resource reallocation model includes: obtaining the efficiency prediction index by combining the collaborative state index and the dynamic time flow within a short future period; obtaining the physical constraints of equipment operation and the optimization objectives of equipment operation to construct the global resource reallocation model; and generating the resource allocation control strategy by solving the global resource reallocation model when the efficiency prediction index is lower than a preset efficiency threshold. This invention achieves optimized resource allocation by generating a resource allocation control strategy, avoiding poor efficiency prediction indicators caused by local optima, and ensuring the efficient and stable operation of port loading and unloading equipment.

[0010] Optionally, the step of acquiring the physical constraints and optimization objectives of equipment operations to construct the global resource reallocation model includes: the physical constraints of equipment operations include the single allocation constraints, task timing constraints, and operation window constraints of the port loading and unloading equipment; maximizing the efficiency prediction index and minimizing the equipment utilization deviation are taken as the equipment operation optimization objectives; and a road-to-road equipment control model of the operation route is constructed by combining the physical constraints and optimization objectives of equipment operations as the global resource reallocation model. This invention balances efficiency maximization and utilization balance, avoids equipment overload or idleness, extends equipment lifespan, and improves the overall collaborative efficiency of the operation route, adapting to complex port operation needs.

[0011] Optionally, the step of converting the resource allocation and control strategy into a set of equipment control instructions, and optimizing the set of equipment control instructions to obtain a final control instruction sequence, includes: parsing the resource allocation and control strategy to generate the set of equipment control instructions; performing spatiotemporal conflict detection on the set of equipment control instructions to obtain a conflict detection result; and optimizing the set of equipment control instructions based on the conflict detection result to obtain the final control instruction sequence. This invention makes the set of equipment control instructions more aligned with the equipment's operational logic, improves the smoothness of instruction execution, avoids operational interruptions caused by invalid instructions, ensures the efficient implementation of control strategies, and improves operational stability.

[0012] Optionally, the step of executing the final control command sequence and obtaining the control effectiveness evaluation result to correct the global resource reallocation model, thereby realizing dynamic control of the port loading and unloading equipment, includes: executing the final control command sequence and collecting actual operating data of the port loading and unloading equipment; quantifying the control effectiveness evaluation result based on the actual operating data; adjusting the model parameters of the global resource reallocation model according to the control effectiveness evaluation result to obtain a decision correction model; obtaining a corrected control strategy through the decision correction model; and realizing dynamic control of the port loading and unloading equipment based on the corrected control strategy. This invention adapts to the dynamic changes in port operations, improves the accuracy of control decisions, and helps to achieve long-term stable and efficient operation of port loading and unloading equipment.

[0013] Optionally, adjusting the model parameters of the global resource reallocation model based on the evaluation results of the control effectiveness to obtain the decision correction model includes: performing sensitivity analysis on the model parameters based on the evaluation results of the control effectiveness to obtain a parameter sensitivity ranking; determining the parameter adjustment direction and parameter adjustment step size based on the parameter sensitivity ranking to obtain a set of corrected parameters for the model parameters; and correcting the global resource reallocation model using the set of corrected parameters to obtain the decision correction model. This invention, by constructing a decision correction model, adapts to changes in different operational scenarios, ensures the reliability and flexibility of the control strategy, and optimizes operational efficiency.

[0014] Secondly, this invention provides a dynamic control system for port loading and unloading equipment based on a work path. The system executes the dynamic control method for port loading and unloading equipment based on a work path provided by this invention. The system includes input devices, output devices, a processor, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions, and the processor is configured to call the program instructions. This invention, through high-performance hardware collaboration, adapts to the complex operating environment of ports, providing reliable hardware support for the dynamic control of port loading and unloading equipment. Attached Figure Description

[0015] Figure 1 This is a flowchart of a dynamic control method for port loading and unloading equipment based on a work path, according to an embodiment of the present invention.

[0016] Figure 2 This is a framework diagram of a port loading and unloading equipment dynamic control system based on a work path, according to an embodiment of the present invention. Detailed Implementation

[0017] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0018] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0019] Please see Figure 1 One embodiment of the present invention provides a dynamic control method for port loading and unloading equipment based on the operating path, the method comprising the following steps:

[0020] S1. Obtain real-time status data of port loading and unloading equipment in the operation route, and obtain the collaborative status index of the operation route based on the real-time status data.

[0021] In this embodiment, to achieve accurate perception of the collaborative status of the work path, real-time status data of the port loading and unloading equipment is collected through various sensors and data interfaces of the equipment control system, including but not limited to:

[0022] Location Information: For mobile devices, such as Automated Guided Vehicles (AGVs) and container trucks, their real-time latitude and longitude coordinates in the geodetic coordinate system or the local coordinate system of the dock are obtained through onboard GPS modules or ultra-wideband indoor positioning base stations. For rail-mounted equipment, such as quayside container cranes (QCs) and rubber-tired gantry cranes (RTGs), their precise displacement coordinates along the track direction are read by displacement sensors as their location information.

[0023] Speed ​​Information: The instantaneous speed information of the equipment is obtained through the encoder feedback signal of its drive motor or the on-board inertial measurement unit. For moving vehicles, the speed is the linear velocity of the vehicle's center of gravity; for the trolley and crane mechanisms of the lifting equipment, the speed is its horizontal movement component; the lifting / lowering speed of the hoisting mechanism is obtained through an encoder. All speed data is uploaded to the central dispatch server in real time at a fixed frequency via industrial Ethernet or wireless communication network.

[0024] Current task elapsed time: The system maintains a state machine and timestamp for each executing job instruction. When the device controller begins executing a specific instruction (such as "grab the box"), it sends an "instruction start" signal to the control system, and the system records this moment as... By continuously comparing the current system time and The difference between the two can be used to calculate the time elapsed for the current task in real time, and the device number and instruction number can be associated to ensure timing consistency.

[0025] It should be noted that a work route is a dynamically assembled collaborative work unit designed to complete a specific container movement task (such as loading / unloading). It consists of initial loading / unloading equipment, horizontal transport equipment, and terminal loading / unloading equipment. As the overall scheduling object, the work route's equipment composition and task sequence are dynamically adjusted according to the operation progress, making it the core carrier for achieving coordinated control of port equipment.

[0026] In this embodiment, cooperative state parameters are calculated based on real-time state data, including but not limited to:

[0027] Rhythm synchronization rate: Taking a certain unloading operation route (QC→AGV→RTG) as an example, the system records the time when the AGV completes the container loading at QC. and the time when unloading is completed at RTG. The difference between the two yields the actual cycle time for one loading and unloading operation. Simultaneously, the system obtains the theoretical cycle time based on the path distance and rated speed between the two devices. The rhythm synchronization rate is obtained by comparing the actual cycle time and the theoretical cycle time. This rate is used to quantify the degree of matching between the work cycles of adjacent equipment in the work path, satisfying the following relationship:

[0028]

[0029] in, For rhythm synchronization rate, To count the number of loops completed within a time window, The index is the number of iterations. It is an exponential function. For the first The actual cycle period of this cycle. For the first The theoretical cycle period of the next cycle.

[0030] Equipment health status: assessed by monitoring the deviation of key operating parameters (such as motor current, oil temperature, and vibration amplitude) from health benchmark values. It reflects the real-time performance status of the equipment and satisfies the following relationship:

[0031]

[0032] in, For equipment health, The types of key operating parameters, For the index of parameter types, For the first The weighting coefficients of the parameters, For the first The current monitoring value of each parameter, For the first The health baseline values ​​of these parameters under rated operating conditions.

[0033] The workload value consists of two parts: instantaneous queue length (i.e., the number of instructions to be executed by all equipment in the work route) and overall utilization rate (the proportion of actual equipment operation time to total time). It is used to characterize the current workload and pressure of the work route and satisfies the following relationship:

[0034]

[0035] in, This is the workload value. This is the balance coefficient (usually taken as 0.6). The instantaneous queue length, This is the maximum allowed queue length. For comprehensive utilization rate.

[0036] In this embodiment, the collaborative weight coefficients are dynamically adjusted and determined based on the business type and real-time stage of the work route. The system presets a weight configuration matrix. For example, during the core high-efficiency operation phase of loading and unloading, the rhythm synchronization rate has the greatest impact on overall smoothness and is given a higher weight; during periods of high equipment failure or preventative maintenance, the weight of equipment health adaptively increases; and when the work route is initialized or nearing completion, the weight of the workload value is relatively higher. The specific collaborative weight coefficients are output through a rule-based regression analysis of historical data, and always satisfy the condition that the sum of all weights is one.

[0037] Furthermore, the calculated cooperative state parameters are linearly weighted and fused with their corresponding cooperative weight coefficients, and the results are normalized to obtain a cooperative state index in the range [0,1], satisfying the following relationship:

[0038]

[0039] in, For the work route At any moment The collaborative state index, For normalized scaling factor, and For collaborative weighting coefficients, For rhythm synchronization rate, For equipment health, This represents the workload value.

[0040] S2. Based on the collaborative state index, and combined with the work environment factors, the task operation chain of the work route is deduced to obtain the dynamic time flow.

[0041] Specifically, S2 includes the following steps:

[0042] S21. Obtain the real-time traffic density of the work route, and obtain the work environment factor based on the real-time traffic density.

[0043] First, the port's horizontal transport road network and key equipment junctions (such as QC seaside platforms and RTG yard driveways) are divided into regular spatial grid cells on an electronic map. The size of each grid cell (e.g., 10m × 10m) is pre-set based on the size and safety clearance of typical AGVs or container trucks. The system receives the location information of all horizontal transport equipment in real time and maps it to the corresponding grid cell.

[0044] Secondly, for the current or soon-to-be-used path of the target operation route, at the statistical time, the total number of devices in all grid cells covered by the path is counted to obtain the real-time traffic density, satisfying the following relationship:

[0045]

[0046] in, For real-time traffic density, For the total number of devices, This represents the total area of ​​all grid cells covered by the path.

[0047] In an optional embodiment, the system simultaneously calculates the average moving speed of the devices, and low-speed devices or stationary devices will be assigned a higher congestion weight.

[0048] Subsequently, in addition to dynamic equipment, the system also accesses the yard planning data from the terminal operating system to identify static container storage areas on both sides of the path. If the storage area is too close to the lane (e.g., less than the safety distance), a visual or operational blind spot will be created. The additional safety distance introduced by the blind spot is obtained by combining the height of the containers in the storage area and the distance between the containers and the lane edge, along with the lane width from the geographic information system. This additional safety distance compresses the effective passage width of the lane, and its obstructive effect on traffic flow can be equivalent to static obstacle density, satisfying the following relationship:

[0049]

[0050] in, Static obstacle density, For correction factors, To add a safety distance, The height of the containers in the storage area. The distance between the container and the lane edge line. This refers to the lane width.

[0051] Finally, the operational environment factors are obtained through real-time traffic density and static obstacle density, satisfying the following relationship:

[0052]

[0053] in, As for work environment factors, For real-time traffic density, The influence coefficient of static obstacles. This refers to the static barrier density.

[0054] S22. For each task stage of the work route, a dynamic time margin is generated by combining the collaborative state index and the work environment factor.

[0055] In this embodiment, for the work path Each task phase The average operation time under ideal conditions of high traffic and low latency is retrieved from the system's historical database and used as the benchmark time.

[0056] Furthermore, the dynamic time margin is obtained by adjusting the baseline time consumption using the collaborative state index and the work environment factor, satisfying the following relationship:

[0057]

[0058] in, For the mission phase Dynamic time margin, Based on the baseline time, and The influence coefficient, For the work route At any moment The collaborative state index, As for work environment factors, This represents the critical density for traffic congestion.

[0059] It should be noted that the influence coefficient is initialized with an empirical value, aiming to minimize the time prediction error, and is determined by training and optimizing a machine learning model (such as gradient descent) on historical data; the critical density of traffic congestion is determined by analyzing the inflection point of traffic density and average vehicle speed in historical data.

[0060] In an alternative embodiment, this applies to equipment downtime phases (such as QC loading). The value can be set to 0 because the dynamic time margin of this task phase is not affected by road traffic and is determined solely by the cooperative state index.

[0061] S23. Based on the dynamic time margin, the task operation chain is deduced to obtain the dynamic time flow.

[0062] In this embodiment, the start time of the first task stage in the work route is used as the time reference; this time reference is determined by the actual completion time of the preceding work route, or by obtaining the planned time from the dock production system.

[0063] Furthermore, the expected start and end times of subsequent task stages are used as task operation chains. The system calculates the time of each task operation chain in the subsequent task stages according to the logical order of the operation process and obtains the sequence of operation time nodes as a dynamic time flow through dynamic time margin.

[0064] For the Task phases ( Its expected start and end times include:

[0065] The estimated start time is equal to the estimated completion time of the previous task phase. For the first task phase, its estimated start time is the time base.

[0066] The estimated completion time satisfies the following relationship:

[0067]

[0068] in, For the first The estimated completion time for each task phase For the first The estimated start time for each task phase. For a moment Task Phase Dynamic time margin, For the first The rolling calculation time for each task phase This is an index for the task phase.

[0069] In this embodiment, after calculating the expected start and completion times of all task stages in the task chain, the generated task time node sequence satisfies the following relationship:

[0070]

[0071] in, This is a sequence of task time nodes. For the first The estimated start time for each task phase. For the first The estimated completion time for each task phase This serves as an index for the task phase. This represents the total number of task phases.

[0072] In an optional embodiment, a dynamic time stream is obtained, and the system immediately compares key nodes in the dynamic time stream (such as the estimated arrival time of the AGV to the QC) with the required time window of the upstream device (QC). If the arrival time is later than the latest start time of the QC, a pre-delay flag is triggered, and the difference between the two is used as the delay amount.

[0073] It should be noted that during actual execution, if the work progress and environmental conditions change significantly, the system will use the latest time as the benchmark to re-trigger the dynamic time flow simulation to ensure that it always closely reflects the actual work situation.

[0074] S3. Obtain the performance prediction index of the operation path based on the dynamic time flow, and generate a resource allocation and control strategy by constructing a global resource reallocation model.

[0075] Specifically, S3 includes the following steps:

[0076] S31. In the near future, the efficiency prediction index is obtained by combining the collaborative state index and the dynamic time flow.

[0077] In this embodiment, an efficiency prediction index is obtained based on a defined prediction time window (e.g., the next 15 minutes) and by integrating multiple dynamic factors. This index is used to quantify the expected output efficiency of the work route in the near future. Specifically, the steps include the following:

[0078] First, in the dynamic time stream, all tasks whose expected start time falls within the forecast time window are selected to form a forecast task set. For each task, a standardized task weight is assigned according to its business type (such as loading or unloading) and container size (20 feet / 40 feet) (for example, a 40-foot standard container is assigned 1.0 and a 20-foot container is assigned 0.7).

[0079] Second, the task weights of all tasks in the prediction task set are summed to obtain a total weight; and the sum of the dynamic time margins required to complete all tasks is extracted from the dynamic time stream. This value already includes the influence of collaborative state and environmental factors, representing the actual expected resource consumption. The ratio between the total weights and the sum of the dynamic time margins constitutes the initial performance prediction value.

[0080] Third, the coordination state index is used as a multiplier to discount the initial performance prediction value by the same proportion. The lower the coordination state index, the more serious the coordination problem within the operation route. The number of tasks marked as pre-delayed in the prediction task set is counted, and a delay penalty factor less than 1 (e.g., 0.95) is introduced to exponentially penalize the initial performance prediction value. The impact of delayed tasks on overall performance grows non-linearly.

[0081] Fourth, based on the above prediction time window, task weight, dynamic time margin, collaborative state index, and delay penalty factor, the performance prediction index is obtained, satisfying the following relationship:

[0082]

[0083] in, As an indicator for performance prediction, To predict the time window, As task weight, For the mission phase Dynamic time margin, For the work route At any moment The collaborative state index, To delay the punishment factor, The number of tasks expected to be delayed.

[0084] S32. Obtain the physical constraints of equipment operation and the optimization objectives of equipment operation to construct the global resource reallocation model.

[0085] In this embodiment, the core decision variable of the global resource reallocation model is the binary device-task allocation variable. This indicates that the plan is being adjusted. In China, will adjustable equipment be included? (e.g., an idle AGV) is assigned to the work route Specific tasks to be performed Resource allocation and regulation strategies are the collection of all such adjustment schemes.

[0086] To ensure the feasibility of the solution, the global resource reallocation model must meet the physical constraints of equipment operation, including but not limited to:

[0087] Single assignment constraint: Each port loading and unloading equipment can be assigned to at most one task at any given time.

[0088] Task timing constraints: For the current task and its predecessor within the same work path, the estimated completion time of the predecessor task must be earlier than the estimated start time of the current task. This is achieved by linking the equipment operation time with the dynamic time margin in the model.

[0089] Job window constraint: Obtain the estimated start time of the current task and the latest start time required by the upstream device. Use the interval between the estimated start time and the latest start time as the job window. The model must ensure that the start time of the task after the device is allocated falls within the job window.

[0090] In this embodiment, maximizing the performance prediction index and minimizing the equipment utilization deviation are taken as the equipment operation optimization objectives.

[0091] Maximize performance forecast metrics: Re-estimate dynamic time flow and performance forecast metrics based on the adjustment plan, and obtain the sum of the adjusted performance forecast metrics.

[0092] Minimize equipment utilization deviation: Under the adjustment scheme, the expected utilization rate of all adjustable equipment (i.e., the ratio of the total time of the assigned tasks to the predicted duration) is minimized, while the sum of the absolute deviations of the utilization rate of each equipment from the global average utilization target is minimized to achieve utilization balance.

[0093] In this embodiment, the physical constraints of equipment operation and the optimization objectives of equipment operation are combined to construct an inter-road equipment control model for the operation route as a global resource reallocation model, satisfying the following relationship:

[0094]

[0095] in, To maximize, For the number of work routes, For the index of the work route, For the work route After adopting the adjustment plan Post-performance prediction indicators For adjustment coefficients, To adjust the plan Adjustable equipment Expected utilization rate This represents the global average target utilization rate.

[0096] S33. When the performance prediction index is lower than the preset performance threshold, the resource allocation control strategy is generated by solving the global resource reallocation model.

[0097] In this embodiment, the system continuously monitors the performance prediction indicators of all work routes. When the performance prediction indicator of any work route falls below its preset performance threshold (which can be dynamically set according to the work route type and vessel priority), or when its cumulative predicted delay exceeds the delay tolerance, the global resource reallocation model is immediately triggered to carry out a global reallocation decision process.

[0098] Specifically, the system calls the optimization solver to solve the constructed global resource reallocation model, using a decomposition-coordination algorithm: First, it determines the job path corresponding to each performance prediction index that does not meet the threshold; second, it quickly generates candidate adjustment schemes through the global resource reallocation model, including resource supplementation schemes or task rearrangement schemes; finally, it coordinates the candidate adjustment schemes at the global level to solve them in an approximately optimal manner, and uses the resource adjustment scheme output by the solver as the resource allocation control strategy.

[0099] S4. The resource allocation and control strategy is transformed into a set of equipment control instructions, and the set of equipment control instructions is optimized to obtain the final control instruction sequence.

[0100] In this embodiment, the resource allocation and control strategy is decoded into specific operation instructions to construct a set of equipment control instructions. The operation instructions include, but are not limited to: equipment transfer instructions (e.g., dispatching an AGV from its current idle state to a certain work route to take over the transportation task of a certain equipment), task sequence adjustment instructions (e.g., swapping the execution order of two tasks in a certain work route), and work parameter adjustment instructions (e.g., increasing the working speed of a certain equipment).

[0101] Before the equipment control command set is issued, spatiotemporal conflict detection is performed to obtain the command conflict detection results.

[0102] Constructing predictive spatiotemporal trajectories: For each instruction in the instruction set involving movement or equipment occupancy, the system simulates its complete process from the planned start time to the completion time. For movement instructions (such as AGV movement), a continuous spatial trajectory pipeline (with width to represent the safety zone) and time occupancy interval are generated based on its path planning and speed curve. For equipment operation instructions (such as RTG loading and unloading), a time occupancy window for the equipment at a specific spatial location is generated. Executing a conflict detection algorithm: All the above spatiotemporal information is compared with the real-time occupancy status generated by the currently executing task in the same spatiotemporal coordinate system.

[0103] Specifically, the conflict detection algorithm mainly identifies two types of conflicts: 1) Time conflict: The usage time windows allocated to the same device by different instructions overlap, or one device needs to use a shared resource (such as a lane or a hoist) before another device has released it; 2) Spatial conflict: The pipelines of two moving trajectories overlap spatially at a future time (such as two AGVs having their predetermined paths intersect and are expected to pass through the intersection at the same time), or the moving trajectory overlaps with the area occupied by static equipment.

[0104] All detected conflicts will be recorded in detail to form an instruction conflict detection result, which will clearly list the conflicting instruction pair, conflict type, estimated time and location of the conflict.

[0105] In this embodiment, after obtaining the conflict detection results, the system activates the optimization engine to reconstruct the device control instruction set to resolve the conflict. The optimization follows the principle of minimum disturbance, that is, while ensuring logical correctness and security, it maintains the original strategy intent and efficiency as much as possible.

[0106] Specifically, optimization methods include, but are not limited to: First, instruction sequence rearrangement: For conflicting instructions involving the same equipment or shared resources, their planned execution order is adjusted while satisfying the logical order of tasks (e.g., unloading must precede loading); this is achieved through a sortable set based on task priority and dependencies. Second, inserting dynamic buffer time: For conflicts caused by overly tight time windows, dynamic buffer time is inserted between conflicting instructions without affecting subsequent critical nodes (e.g., ship windows). The dynamic buffer time is dynamically determined based on equipment acceleration / deceleration performance and system safety redundancy time. Third, local path / speed fine-tuning: For spatial path conflicts, an alternative path is replanned for one of the movement instructions without significantly increasing travel time, or its planned speed through the conflict point is adjusted to achieve spatiotemporal peak avoidance.

[0107] In this embodiment, the optimization process is modeled as a constrained scheduling optimization problem. It is solved quickly by combining heuristic rules (such as prioritizing the resolution of the most pressing conflicts) with search algorithms (such as genetic algorithms). The final output is a sequence of control instructions that eliminates all identified spatiotemporal conflicts and has accurate timestamps and device action descriptions, which can be directly distributed for execution.

[0108] S5. Execute the final control command sequence, obtain the control effectiveness evaluation results, correct the global resource reallocation model, and realize the dynamic control of the port loading and unloading equipment.

[0109] Specifically, S5 includes the following steps:

[0110] S51. Execute the final control command sequence and collect the actual operation data of the port loading and unloading equipment.

[0111] In this embodiment, the final control command sequence is distributed to the programmable logic controller or vehicle control unit of the target device through an industrial Internet of Things (IoT) communication network deployed in the port. After receiving and parsing the commands, the device controller drives the device to perform precise actions as required by the commands (such as moving to a specified coordinate or performing a grasping and releasing operation).

[0112] During execution, the system collects actual operation data in parallel through a sensor network, including at least: 1) High-precision spatiotemporal trajectory data: the actual movement path, real-time position, speed of the equipment, and the precise timestamp of each key action; 2) Operation efficiency data: the actual time consumed by each instruction and the actual cycle of loading and unloading actions; 3) Equipment status data: motor current, energy consumption, and fault alarm codes during execution.

[0113] It should be noted that all actual operation data is transmitted to the central server in real time and is strictly bound and time-aligned with the issued equipment to form a traceable instruction-response data chain.

[0114] S52. Quantify the evaluation results of the control effectiveness based on the actual operation data.

[0115] In this embodiment, the effectiveness of the regulation is quantitatively evaluated from multiple dimensions based on the collected actual operational data. The evaluation revolves around the core objectives of the regulation: the accuracy of time prediction and the degree of improvement in overall efficiency.

[0116] First, calculate the time prediction bias: compare the actual time taken for each task stage with the dynamic time margin, and calculate the average absolute percentage error as the core indicator of prediction accuracy.

[0117] Secondly, calculate the efficiency achievement rate: within the time window of the regulation, count the actual number of standard containers or the total weight of tasks completed on the operation route, and compare it with the efficiency prediction indicators before regulation.

[0118] Finally, assess the cost of disruption: calculate the additional waiting time or efficiency loss caused to the original work route due to this adjustment (such as equipment transfer).

[0119] In summary, the evaluation result of the regulation effectiveness is quantified into a comprehensive score. For example, the evaluation result of the regulation effectiveness can be obtained by weighting and integrating the above indicators using adjustable weight parameters.

[0120] S53. Based on the evaluation results of the regulation effectiveness, the model parameters of the global resource reallocation model are adjusted to obtain the decision correction model.

[0121] In this embodiment, parameter sensitivity analysis of the model parameters is performed to identify the most critical model parameters for improving the evaluation results of regulation effectiveness.

[0122] Locality sensitivity analysis is employed. The system performs analysis based on historical datasets. For each parameter to be analyzed... With other parameters fixed, at its baseline value Based on this, a small perturbation quantity is determined. The parameters are calculated separately and then become and When the model is simulated on a set of validation cases to obtain the change in the evaluation results of the regulatory effectiveness, the sensitivity of the parameters satisfies the following relationship:

[0123]

[0124] in, For the first Sensitivity of each parameter To control the changes in the effectiveness evaluation results, For the first A tiny perturbation of each parameter.

[0125] Furthermore, parameters are sorted from highest to lowest sensitivity to obtain a parameter sensitivity ranking. Parameters ranked higher have a greater impact on the overall control effectiveness from minor adjustments and are therefore prioritized for subsequent corrections.

[0126] In this embodiment, based on parameter sensitivity ranking, the system aims to improve the evaluation results of regulation effectiveness and determines the adjustment direction and step size of parameters through iterative optimization search.

[0127] First, determine the direction of adjustment: observe the statistical correlation between the parameter value and the effectiveness evaluation results of the control measures based on historical data. For example, if historical data shows that an increase in the value of the delay penalty factor (more severe delay penalties) is often accompanied by an increase in the effectiveness evaluation results of the control measures, then the direction of this adjustment is to increase the delay penalty factor.

[0128] Next, an adaptive step size search method is used to determine the adjustment step size: starting from an initial step size (such as 5% of the baseline value), the parameters are adjusted in the determined adjustment direction with this step size, and the model is simulated and run on the historical validation set to predict the new control effect evaluation result. If it is higher than the current control effect evaluation result, the step size is increased in the same direction (such as increasing to 10%); if it is lower than the current control effect evaluation result, the step size is reduced or the opposite direction is tried.

[0129] Finally, the top-ranked parameters are optimized in the aforementioned direction and step size. The adjustment process is carried out rapidly within a predefined maximum number of iterations, and a set of parameter values ​​that enable the estimated control effectiveness assessment results to achieve local optima are output as a set of correction parameters.

[0130] In this embodiment, after obtaining the set of correction parameters, the system updates the global resource reallocation model and generates a decision correction model. This process is not simply replacing parameters, but a process that includes verification and version management.

[0131] First, model reloading and validation: replace the corresponding parameters in the original model with the new values ​​in the model to form a new variant of the model.

[0132] Secondly, the new model is run on an independent set of historical test cases that are not involved in parameter optimization. The predicted performance of the simulated decision is compared with the actual historical results of the cases to ensure that its comprehensive evaluation index is better than or at least not worse than the original model, so as to prevent overfitting.

[0133] Finally, after successful verification, the model is switched and archived: the system marks the currently used online model as the old version and archives it, and loads the corrected new model into the online decision engine, ready to respond to the next regulatory needs.

[0134] At the same time, the system records a complete log of this parameter calibration, including the triggering reason, analysis process, new and old parameter values ​​and verification results, forming a knowledge base for continuous improvement.

[0135] S54. Obtain a correction and control strategy through the decision correction model, and realize dynamic control of the port loading and unloading equipment based on the correction and control strategy.

[0136] In this embodiment, obtaining the decision correction model signifies that the system has completed a full closed-loop control cycle. When the system detects a decline in performance prediction indicators again and triggers a new round of resource reallocation decisions, it will invoke the decision correction model that has been corrected and optimized based on historical experience.

[0137] When generating corrective control strategies, the modified model will use more accurate parameters (e.g., penalty factors that are more sensitive to delays and impact coefficients that are more closely aligned with current traffic conditions) for prediction and optimization calculations, thereby generating higher-quality and more effective corrective control strategies.

[0138] After the modified control strategy is converted into a sequence of instructions for execution, the actual results are collected, evaluated, and fed back to complete the model correction. This cycle forms a closed-loop adjustment, enabling the global resource reallocation model to dynamically adapt to the long-term drift and changes in the port operation environment, equipment performance, and operation mode, thereby achieving dynamic control of port loading and unloading equipment with continuous self-optimization capabilities.

[0139] Please see Figure 2In an optional embodiment, the present invention provides a dynamic control system for port loading and unloading equipment based on a work path. The system includes input devices, output devices, a processor, and a memory, all interconnected. The memory stores a computer program comprising program instructions, and the processor is configured to invoke the program instructions to execute specific steps as described in the relevant embodiments of the dynamic control method for port loading and unloading equipment based on a work path provided by the present invention. The dynamic control system for port loading and unloading equipment based on a work path provided by the present invention has a complete and stable structure, enhancing the overall applicability and practical application capability of the present invention.

[0140] In summary, the present invention provides a dynamic control method and system for port loading and unloading equipment based on work routes. First, real-time status data is collected and a collaborative status index is calculated. Second, a dynamic time flow is generated by rolling out the task operation chain in conjunction with operational environment factors. Then, based on the dynamic time flow, an efficiency prediction index is calculated to drive a global resource reallocation model to generate a resource allocation control strategy. Subsequently, the resource allocation control strategy is transformed into a set of equipment control instructions, which is optimized to obtain the final control instruction sequence. Finally, the final control instruction sequence is executed, and the control effectiveness evaluation results are quantified and fed back to the global resource reallocation model for parameter self-correction, thus forming a closed-loop dynamic control system. The method of this invention is easy to understand, computationally simple, requires minimal workload, and is convenient for engineering applications, providing a theoretical foundation and technical support for the further development of port scheduling technology.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A dynamic control method for port loading and unloading equipment based on operating paths, characterized in that, Includes the following steps: S1. Obtain real-time status data of port loading and unloading equipment in the operation route, and obtain the cooperative status index of the operation route based on the real-time status data; S2. Based on the collaborative state index, and combined with the work environment factors, the task operation chain of the work route is deduced to obtain the dynamic time flow. S3. Obtain the efficiency prediction index of the operation route based on the dynamic time flow, and generate a resource allocation and control strategy by constructing a global resource reallocation model; S4. The resource allocation and control strategy is converted into a set of equipment control instructions, and the set of equipment control instructions is optimized to obtain the final control instruction sequence. S5. Execute the final control command sequence, obtain the control effectiveness evaluation results, correct the global resource reallocation model, and realize the dynamic control of the port loading and unloading equipment. S1 includes: The real-time status data is obtained, including the location information, speed information, and current task elapsed time of the port loading and unloading equipment; Based on the real-time status data, collaborative status parameters are calculated, including rhythm synchronization rate, equipment health, and workload value; The rhythm synchronization rate satisfies the following relationship: in, For rhythm synchronization rate, To count the number of loops completed within a time window, The index is the number of iterations. It is an exponential function. For the first The actual cycle period of this cycle. For the first The theoretical cycle period of this cycle; Determine the collaborative weighting coefficient, and then use the collaborative weighting coefficient to weight and fuse the collaborative state parameters to obtain the collaborative state index; S2 includes: Obtain the real-time traffic density and static obstacle density of the work route, and obtain the work environment factor based on the real-time traffic density and the static obstacle density; The real-time traffic density satisfies the following relationship: in, For real-time traffic density, For the total number of devices, The total area of ​​all grid cells covered by the path; The static barrier density satisfies the following relationship: in, Static obstacle density, For correction factors, To add a safe distance, The height of the containers in the storage area. The distance between the container and the lane edge line. Lane width; The work environment factors satisfy the following relationship: in, As for work environment factors, For real-time traffic density, The influence coefficient of static obstacles. Static barrier density; For each task stage of the work route, a dynamic time margin is generated by combining the collaborative state index and the work environment factors; The dynamic time flow is obtained by extrapolating the task chain based on the dynamic time margin. The process of deriving the dynamic time flow from the task chain based on the dynamic time margin includes: The start time of the first task phase in the aforementioned work path is used as the time reference; The expected start and completion times of subsequent task phases are used as the task chain; Based on the time base, the task operation chain is rolled out using the dynamic time margin to obtain the operation time node sequence, and the operation time node sequence is used as the dynamic time flow.

2. The dynamic control method for port loading and unloading equipment based on the operating path according to claim 1, characterized in that, The step of obtaining the performance prediction index of the work route based on the dynamic time flow and generating a resource allocation control strategy by constructing a global resource reallocation model includes: In the near future, the performance prediction index will be obtained by combining the collaborative state index and the dynamic time stream. Obtain the physical constraints and optimization objectives of equipment operations to construct the global resource reallocation model; When the performance prediction index is lower than the preset performance threshold, the resource allocation control strategy is generated by solving the global resource reallocation model.

3. The dynamic control method for port loading and unloading equipment based on the operating path according to claim 2, characterized in that, The step of obtaining the physical constraints and optimization objectives of equipment operations to construct the global resource reallocation model includes: The physical constraints of equipment operation include the single allocation constraint, task timing constraint, and operation window constraint of the port loading and unloading equipment. The optimization objectives for the equipment operation are to maximize the performance prediction index and minimize the equipment utilization deviation. The inter-road equipment control model of the operation route is constructed by combining the physical constraints of the equipment operation and the optimization objectives of the equipment operation as the global resource redistribution model.

4. The dynamic control method for port loading and unloading equipment based on the operating path according to claim 1, characterized in that, The step of converting the resource allocation and control strategy into a set of equipment control instructions, and optimizing the set of equipment control instructions to obtain the final control instruction sequence includes: The resource allocation and control strategy is parsed to generate the equipment control instruction set; Spatiotemporal conflict detection is performed on the device control instruction set to obtain instruction conflict detection results; Based on the command conflict detection results, the device control command set is sequence optimized to obtain the final control command sequence.

5. The dynamic control method for port loading and unloading equipment based on the operating path according to claim 1, characterized in that, The process of executing the final control command sequence, obtaining the control effectiveness evaluation results, and correcting the global resource reallocation model to achieve dynamic control of the port loading and unloading equipment includes: Execute the final control command sequence and collect the actual operation data of the port loading and unloading equipment; The evaluation results of the control effectiveness are quantified based on the actual operational data. Based on the evaluation results of the regulation effectiveness, the model parameters of the global resource reallocation model are adjusted to obtain a decision correction model; The decision correction model is used to obtain a correction control strategy, and the port loading and unloading equipment is dynamically controlled based on the correction control strategy.

6. The dynamic control method for port loading and unloading equipment based on the operating path according to claim 5, characterized in that, The step of adjusting the model parameters of the global resource reallocation model based on the evaluation results of the regulation effectiveness to obtain the decision correction model includes: Based on the evaluation results of the regulation effectiveness, a sensitivity analysis was performed on the model parameters to obtain a parameter sensitivity ranking; The parameter adjustment direction and step size are determined based on the parameter sensitivity ranking to obtain the set of correction parameters for the model parameters; The decision correction model is obtained by correcting the global resource reallocation model using the set of correction parameters.

7. A dynamic control system for port loading and unloading equipment based on a work path, characterized in that, The system includes an input device, an output device, a processor, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the dynamic control method for port loading and unloading equipment based on the work path as described in any one of claims 1-6.

Images