Port loading blending plan and yard operation line control method

Abstract

The application discloses a port loading mixing plan and yard operation line control method, and belongs to the technical field of port operation plan and equipment collaborative control. First, the port yard layout, loading operation task information in a planning period and yard equipment parameter information are collected to provide complete basic data for constructing a coupled space-time network model and realize high-precision collaborative scheduling of the yard equipment. Second, a coupled space-time network model considering loading operation task and mixing constraints is constructed, the spatial position, operation time sequence of each loading operation task in the yard, mixing constraints and equipment use are uniformly associated, and collaborative optimization of operation, mixing and conveying is realized. Finally, the operation time sequence of the loading operation task in the planning period and the running scheduling of the reclaimer are uniformly optimized and solved, the solution result is converted into an execution instruction of the yard equipment, and high-precision collaboration of the yard equipment is realized. The application can improve the operation efficiency of the port yard system under the premise of ensuring the mixing quality and the continuity of the loading operation.

Description

Technical Field

[0001] This invention belongs to the field of port operation planning and equipment collaborative control technology, and relates to a method for production organization and equipment control of port loading operations, particularly a method for port loading mixing planning and yard operation line control. Background Technology

[0002] Ports are crucial hubs for the cross-regional flow of bulk materials such as coal and ore, and their loading and unloading efficiency directly impacts ship turnaround, logistics costs, and the stability of downstream supply chains. In typical operational processes, the stockyard is responsible for receiving, storing, and organizing materials, while bucket wheel reclaimers and belt conveyor systems together constitute the main conveying channels within the terminal. With increasingly diversified user demands and more refined order granularities, ports face a growing number of tasks requiring the mixing of different batches and qualities of materials in specific proportions during the loading stage. Compared to traditional single-item loading, blending operations couple material sources, reclaiming sequences, and conveying paths, significantly increasing the complexity of operational organization.

[0003] Basic floor plan layout as follows Figure 1 As shown, the stockyard consists of several parallel stockpiles, each divided into multiple independent stacking positions, serving as the smallest storage unit for different material grades. Stacker cranes and reclaimers operate along their respective tracks, using robotic arms to perform stacking and reclaiming operations on adjacent stacking positions. A belt conveyor system arranged along the tracks handles material transport between the stockyard and loading berths or unloading facilities, achieving continuous connection from stacking and reclaiming to loading. This layout is typical of large bulk cargo terminals, helping to reduce equipment interference and improve operational efficiency.

[0004] Existing technologies have explored the issues of mixing and equipment organization in port loading operations from different perspectives. For example, Chinese invention patent CN121119877A constructs a collaborative configuration model for multi-line equipment clusters and combines it with intelligent decision-making methods to dynamically generate equipment configuration schemes, which has a certain effect on improving overall operational coordination. Chinese invention patent CN121119876A introduces large language models and reinforcement learning algorithms to optimize the equipment operation sequence during the cleaning and unloading stages, thereby improving equipment coordination efficiency. However, the above methods mostly focus on macro-level scheduling or the unloading process, and do not systematically model the mixing plan for multiple batches of bulk cargo during the loading stage, nor do they conduct in-depth research on the collaborative control and operation mechanism of equipment such as yard reclaimers and belt conveyors under mixing constraints, making it difficult to meet the actual needs of refined loading and mixing operations.

[0005] In summary, while existing technologies have proposed various improvement schemes in terms of equipment structure modification, operation scheduling, and flow control, they generally lack a unified coordination mechanism for the mixing ratio, material source, material picking rhythm, and conveying path during the loading stage. Furthermore, their coupling with the real-time loading process at the port and the operational status of the yard operation line is low, making it difficult to meet the continuous operation requirements of high-precision, dynamic mixing scenarios. Therefore, it is necessary to propose a scheduling optimization method that addresses the mixing requirements of loading, enables collaborative formulation of mixing plans within a unified framework, and controls the operation of the yard operation line. This will achieve efficient collaboration among multiple stages of operations from the yard to the terminal, improving mixing accuracy and overall operational efficiency. Summary of the Invention

[0006] This invention addresses the problems in existing port loading operations, such as insufficient mixing accuracy, frequent reclaimer wait times, conveyor interruptions, and overall reduced loading efficiency due to a lack of unified coordination between loading task timing, mixing scheme execution, and reclaimer scheduling. It proposes a port loading mixing plan and yard operation line control method. Since the yard stockpiling plan is usually pre-determined by the port and can be considered a known input, this invention does not optimize the stockpiling process but focuses on the overall coordination of reclaiming and mixing. This method centers on the coordination of loading task timing and reclaimer operation scheduling. By constructing a coupled spatiotemporal network model, it unifies the modeling of the spatial location, timing, mixing ratio, and available reclaimer resources for loading tasks, achieving temporal and spatial coordination and optimization of loading tasks. This improves mixing accuracy, ensures loading continuity, and optimizes yard operation efficiency.

[0007] To achieve the above objectives, the present invention provides the following solution:

[0008] A method for port loading and mixing planning and yard operation line control optimizes mixing scheduling by constructing a coupled spatiotemporal network model that coordinates the timing of loading tasks and the operation scheduling of material reclaimers. This method improves the operational efficiency and economy of the port yard system while ensuring the continuity of loading operations and the accuracy of mixing. Specifically, it includes the following steps:

[0009] Step S1: Collect port yard layout, ship loading operation task information within the planned period, and yard equipment parameter information to provide complete basic data for constructing a coupled spatiotemporal network model, and realize high-precision collaborative scheduling of yard equipment; specifically:

[0010] Step S1-1: Collect basic information on the layout of the port yard;

[0011] Collect information on the layout and spatial structure of the port yard, including the number of stockpiles, the number of stacks per stockpile, the material grade, material cost, and initial material quantity for each stack. Collect information on the number of reclaimer tracks, record the total planning time and time step, and provide a foundation for establishing a coupled spatiotemporal network model.

[0012] Step S1-2: Collect information on loading operations during the planning period;

[0013] Collect information on loading operations during the planning period, including the number of loading operations, workload, earliest start time, and mixing scheme. The mixing scheme includes material mixing ratio requirements and mixing plan, which is used to achieve high-precision mixing in scheduling optimization.

[0014] Step S1-3: Collect yard equipment parameter information;

[0015] The storage yard equipment involved in this invention is a reclaimer. The equipment parameter information includes the number of reclaimers, the reclaiming efficiency of the reclaimers, the moving speed, and the range of stack positions where loading operations can be carried out.

[0016] Step S2: Based on the port yard layout, ship loading task information, and yard equipment parameter information obtained in Step S1, construct the spatiotemporal network of ship loading tasks and the spatiotemporal network of the reclaimer. On this basis, construct a coupled spatiotemporal network model considering ship loading tasks and mixing constraints, uniformly linking the spatial location, operation sequence, mixing constraints, and equipment availability of each ship loading task within the yard, achieving coordinated optimization of operations, mixing, and transportation, and providing support for optimizing ship loading efficiency and improving yard resource utilization. Specifically:

[0017] Step S2-1: Construct the nodes and arcs of the spatiotemporal network for the ship loading operation task;

[0018] First, establish a spatiotemporal network for the ship loading operation task, denoted as [missing information]. The spatiotemporal network of the loading operation task consists of a set of vertices. Sum of arcs Composition. Each vertex is represented as ,in For spatial nodes, For time indexing, Represents a set of spatial nodes. Represents a time set. The set of spatial nodes. It includes the following four types of nodes:

[0019] (1) Waiting for a subset of nodes : Indicates the idle state of the loading operation (i.e., no material handling is performed during this period). Each spatial node With material grade This is a linked system, allowing only loading tasks of the corresponding level to enter this node. The scheduling of each loading task originates from its corresponding waiting node. Therefore, the source vertex of the loading operation is ( , );

[0020] (2) Material picking node subset : Indicates the direct material handling operation for loading tasks, and is related to spatial nodes. The relevant stack locations and schemes are respectively denoted as and ;

[0021] (3) Mixed node subset This indicates the mixing and reclaiming operation for the loading task, according to the plan. For two stack positions and The materials are mixed and blended.

[0022] (4) Sink node : Indicates the termination of the scheduling of the loading operation in the spatiotemporal network, such as Figure 3 Node 9. All loading operations ultimately converge at the confluence. The process ends here.

[0023] The spatiotemporal network of the loading operation task not only includes the four types of nodes mentioned above, but also connects vertices through different types of arcs to represent the possible movements and operations of the task. Specifically, it includes the following seven types of arcs:

[0024] (1) Waiting arc: This indicates a unit of time to wait on the node;

[0025] (2) Transition arc: This indicates the transition of a loading operation task from a waiting state to a working state (such as material picking or mixing). These movements from waiting nodes to material picking or mixing nodes do not consume actual time in the spatiotemporal network of loading operation tasks; Indicates the index of the material picking node; Indicates a time index;

[0026] (3) Material taking arc: This represents one unit of time at the material handling node. The unit material handling capacity is defined as the amount of material that the direct material handling scheme can process within one unit of time.

[0027] (4) Mixed arc: Each mixed arc Indicates from time Begin, Continue At each time step, in the stack position and The mixing process is carried out on the stack. The mixing process reduces the unit material handling rate at each stack location; therefore, mixing requires multiple time steps to produce at least one effective mixed material unit.

[0028] (5) Reset arc: This indicates that the operation has been performed at the stacking position and the system has returned to a waiting state.

[0029] (6) Convergence arc: This indicates that the loading operation task has been transferred from the waiting node to the receiving node, completing the entire loading operation cycle;

[0030] (7) Cancel arc: This indicates a cancelled loading operation. The process is transferred directly from the source waiting node to the sink node, without any further action taken during the entire planning period.

[0031] Step S2-2: Construct the nodes and arcs of the spatiotemporal network of the material reclaimer;

[0032] Subsequently, a spatiotemporal network for the material reclaimer is constructed, denoted as... The spatiotemporal network of the material handling machine consists of a set of vertices. Sum of arcs Composition. Each vertex is represented as ,in For spatial nodes, Indexed by time. A collection of spatial nodes. It includes four types of nodes:

[0033] (1) Pile position node subset Because the material handling machine can only move on the track, it must occupy a specific stacking position at any given time. Indicates the index of the stack location set. Representing spatial nodes The stack position indicated;

[0034] (2) Subset of task nodes For each spatial node Define the corresponding work node to indicate that a material picking operation will be performed at this stack location;

[0035] (3) Material reclaimer source node : This represents the index of the set of reclaimers, each reclaimer having a known initial stack position. Therefore, in the spatiotemporal network of the material reclaimer, the material reclaimer... The source vertex is ,in ,and ; Indicates the initial stack position where the reclaimer source node is located;

[0036] (4) Sink node : Indicates the end of the spatiotemporal network trajectory of the material handling machine.

[0037] The spatiotemporal network of the material reclaimer not only includes the four types of nodes mentioned above, but also connects vertices through different types of arcs to represent the possible movements and operations of the material reclaimer. Specifically, it includes the following five types of arcs:

[0038] (1) Waiting arc: Vertex Connected to This indicates that the material reclaimer is at the stack node. The upper part remains idle;

[0039] (2) Moving arc: This indicates that the material reclaimer starts from the stack node. Move to another stack node ; Indicates a time index;

[0040] (3) Working arc: ,in ,and This indicates that the material handling machine has entered the working state from the idle state and is in the stack position. Above the time interval Perform material handling operations;

[0041] (4) Reset arc: ,in ,and This indicates that after completing the material picking operation, the material picker returns from the working state to the idle state at the same stack location;

[0042] (5) Convergence arc: This indicates that the material handling machine has completed scheduling and entered the sink node.

[0043] The spatiotemporal networks of the loading tasks and the reclaimers respectively constructed candidate paths for all loading tasks and reclaimers. Since the reclaiming operation at a certain stack location must be performed by the reclaimer, there is a coupling relationship between the two. Specifically, if a loading task falls within a certain time interval... to Use stacking position Then a material handling machine must pass through the working arc. ,in In other words, the operation arc in the spatiotemporal network of the material reclaimer. Coupled with a ship loading operation task arc in the spatiotemporal network of the ship loading operation task, denoted as ,in Includes all stack positions (satisfy Related material handling, mixing, or transformation arcs, Indicates all within the time interval Inside, the stack is in use. The loading operation task arc (including material taking arc, mixing arc and transfer arc). This represents a time index. Conversely, if the loading operation uses a certain operation arc in the spatiotemporal network of the loading operation task, then... Then the set Each arc in the process must be executed by a reclaimer, among which, This represents the spatiotemporal network operation arc of the material handling machine.

[0044] Step S2-3: Establish the collaborative constraints of the spatiotemporal network of the loading operation task and the spatiotemporal network of the material reclaimer as follows:

[0045] (1)

[0046] (2)

[0047] (3)

[0048] (4)

[0049] (5)

[0050] (6)

[0051] (7)

[0052] (8)

[0053] (9)

[0054] (10)

[0055] (11)

[0056] (12)

[0057] In the formula, , , , , , , , , , , , , , , , , , , The definitions are as shown above; This indicates the arc index used to avoid symbol duplication; The spatiotemporal network spatial node index of the material handling machine is used to avoid symbol duplication; The weighting coefficient representing the total port dwell time for loading operations; The material cost weighting coefficient represents the material cost of the loading operation. This indicates the quantity weighting parameter for tasks that do not meet the requirements for loading operations. Represents the set of loading operations, using index; Arc representing the spatiotemporal network of the loading operation task Processing time, Indicates the loading operation task Arc of the spatiotemporal network for ship loading operations hour It is 1 if it is true, otherwise it is 0; Arc representing the spatiotemporal network of the loading operation task Material costs; Arc representing the spatiotemporal network of the loading operation task The quantity of materials already taken; Indicates loading operation task Required quantity of materials; To represent a set of storage locations in a port yard, use index; Indicates the material taking arc or mixing arc At the stack The quantity of materials processed above; Indicates stack position The initial quantity of materials; Indicates a collection of material handling machines, using index; Indicates when the material handling machine Arc of the spatiotemporal network using a material handling machine hour It is 1 if it is true, otherwise it is 0; Indicates when arc Feeder During execution, due to non-intersecting constraints, there are no reclaimer spatiotemporal network arcs that cannot be executed simultaneously with the arc in the reclaimer spatiotemporal network.

[0058] Formula (1) represents the objective function minimizing the weighted summation term, including: the total waiting time for the loading tasks, the total material cost incurred in meeting the loading requirements, and the number of unmet loading tasks. The constraints shown in Formulas (2) and (3) are flow balance constraints in the spatiotemporal network of the loading tasks, ensuring that each loading task is assigned to an effective path. The constraint shown in Formula (4) ensures that the total amount of material taken from each loading task on the selected arc is equal to the demand of that loading task. The constraint shown in Formula (5) limits the number of times the same loading task can be used at each stack to at most once. The constraint shown in Formula (6) ensures that the total amount of material taken from any stack does not exceed its initial amount of material. The constraint shown in Formula (7) enforces the coupling relationship between the spatiotemporal network flow of the loading tasks and the spatiotemporal network flow of the reclaimers, while ensuring that each stack can be occupied by at most one loading task at any given time. Formulas (8) and (9) are flow balance constraints in the spatiotemporal network of the reclaimers, ensuring that each reclaimer is assigned to a feasible trajectory. Constraint (10) is a non-intersecting constraint during the material handling process. Finally, the constraints shown in formulas (11) and (12) define the range of values ​​for the variables used in the coupled spatiotemporal network model.

[0059] Step S3: Based on the coupled spatiotemporal network model constructed in Step S2, the operation sequence of the loading operation tasks and the operation scheduling of the reclaimer within the planning period are uniformly optimized and solved. The operation sequence, mixing scheme and yard equipment scheduling scheme of each loading operation task within the planning period are generated, and the solution results are transformed into executable instructions for the yard equipment to achieve high-precision coordination of the yard equipment.

[0060] This invention directly solves the coupled spatiotemporal network model using the CPLEX solver within a C# environment. It employs the solver's built-in heuristic methods to comprehensively optimize the timing constraints, mixing constraints, and yard equipment occupancy conflicts of the loading operation. During the solution process, the coupled spatiotemporal network model outputs material handling paths, mixing schemes, and equipment scheduling results, providing complete and executable scheduling decision support for actual loading operations. Specifically:

[0061] Step S3-1: Solve for environment initialization and instantiation;

[0062] In the C# environment, the yard layout, loading operation tasks, and yard equipment parameter information collected in step S1 are loaded as input data, and the CPLEX solver instance is initialized based on the coupled spatiotemporal network model constructed in step S2. The variables, constraints, and objective function of the coupled spatiotemporal network model are fully imported into the solver at this stage, laying the foundation for subsequent solving.

[0063] Step S3-2: Solve the coupled spatiotemporal network model based on CPLEX;

[0064] The CPLEX solver is invoked to optimize and solve the coupled spatiotemporal network model, yielding a solution. The solver automatically handles job timing constraints, mixing constraints, and yard equipment conflict constraints within the model, using a built-in heuristic search mechanism to find the optimal scheduling scheme. This process requires no additional manual design of heuristics or manual parameter tuning, directly obtaining optimal or near-optimal scheduling results that satisfy spatiotemporal consistency.

[0065] Step S3-3: Scheduling scheme parsing and executable instruction generation;

[0066] After solving the problem, the optimal operation sequence, mixing scheme execution, and reclaimer operation segment for each loading task are analytically obtained from the solution of the coupled spatiotemporal network model. The analytical results are further converted into executable instructions for the reclaimer's movement and operation plan. These instructions can be directly used to guide the on-site scheduling system, improving the overall efficiency of the loading operation.

[0067] The beneficial effects of this invention are as follows:

[0068] This invention achieves integrated optimization of loading and conveying processes by constructing a coupled spatiotemporal network model that aligns the timing of loading tasks with the scheduling of material reclaimers. Specifically, the model provides a unified description of the spatiotemporal relationships, mixing constraints, and availability of yard equipment for loading tasks. During the solution process, it automatically generates material reclaiming paths and equipment scheduling schemes that meet process requirements. This avoids mixing deviations and loading conflicts caused by manual rules, significantly improving the efficiency of material reclaimers, reducing waiting time during loading, and enhancing mixing accuracy and the continuity of loading operations. Therefore, this invention effectively improves the operational efficiency of port yard systems while ensuring mixing quality and loading continuity, providing an innovative technical solution for intelligent scheduling and loading operation optimization in ports. Attached Figure Description

[0069] Figure 1 This is a layout diagram of a port's storage yard.

[0070] Figure 2 This is an example of a port storage yard.

[0071] Figure 3 Spatiotemporal network path diagram for ship loading operations.

[0072] Figure 4 This is the spatiotemporal network path diagram of the material handling machine.

[0073] Figure 5 This is a flowchart of the present invention. Detailed Implementation

[0074] The technical solution of the present invention will be described in detail below.

[0075] This embodiment provides a method for port loading and mixing planning and yard operation line control, such as... Figure 1 As shown, it includes the following steps:

[0076] Step S1: Collect port yard layout, ship loading operation task information within the planned period, and yard equipment parameter information to provide complete basic data for constructing a coupled spatiotemporal network model, and realize high-precision collaborative scheduling of yard equipment; specifically:

[0077] Step S1-1: Collect basic information on the layout of the port yard;

[0078] Collect information on the layout and spatial structure of the storage yard, such as... Figure 2 As shown, the stockpile is arranged with 2 stockpiles, each divided into 3 stacking positions; the stacking position label "#1 (6, [5, 22])" indicates that the stacking position index is 1, the grade index of the stacked material is 6, and the initial material quantity follows a discrete uniform distribution on the interval [5, 22] (unit: thousand tons). For each randomly generated case, the grade of the stacked material is fixed, while the initial material quantity is sampled to reflect the dynamic changes in the stockpile operation; there is 1 reclaimer track and 2 reclaimers (e.g., ...). Figure 2 As shown in R1 and R2 (where R1 represents the first material reclaimer and R2 represents the second material reclaimer), the total planning time is 6 hours and the time step is 15 minutes, with a total of 24 time steps, which provides a foundation for the subsequent establishment of a coupled spatiotemporal network model.

[0079] Step S1-2: Collect information on loading operations during the planning period;

[0080] Information on loading operations during the planning period is collected, including the number of loading operations, workload, earliest start time, and mixing scheme. The mixing scheme information includes material mixing ratio requirements and the mixing scheme itself, used to achieve high-precision mixing during scheduling optimization. This invention collects a total of 6 sets of examples, each set including 6 loading operations.

[0081] Step S1-3: Collect yard equipment parameter information;

[0082] The stockpile equipment involved in this embodiment is a reclaimer. The equipment parameter information includes 2 reclaimers; the reclaiming efficiency of each reclaimer is 4000 tons per hour, or 1000 tons per unit time; the moving speed of each reclaimer is 1 stack position per unit time; the stack positions that reclaimer 1 can operate at are stack position 1, stack position 2, stack position 4, and stack position 5; the stack positions that reclaimer 2 can perform ship loading operations at are stack position 2, stack position 3, stack position 5, and stack position 6.

[0083] By collecting information on the yard layout, loading operations, and yard equipment parameters, complete basic data is provided for constructing a coupled spatiotemporal network model, enabling high-precision collaborative scheduling of yard equipment.

[0084] Step S2: Based on the port yard layout, ship loading task information, and yard equipment parameter information obtained in Step S1, construct the spatiotemporal network of ship loading tasks and the spatiotemporal network of the reclaimers. On this basis, construct a coupled spatiotemporal network model considering ship loading tasks and mixing constraints, uniformly linking the spatial location, operation sequence, mixing constraints, and equipment availability of each ship loading task within the yard. This achieves coordinated optimization of operations, mixing, and transportation, providing support for optimizing ship loading efficiency and improving yard resource utilization. To more clearly illustrate the spatiotemporal network path, Figure 3 , Figure 4 The example demonstrates the spatiotemporal network paths of the loading and unloading tasks and the unloading tasks in an instance consisting of one stockpile, three stacking positions, three loading tasks, and two reclaimers.

[0085] Step S2-1: Construct the nodes and arcs of the spatiotemporal network for the ship loading operation task;

[0086] First, establish a spatiotemporal network for the ship loading operation task, denoted as [missing information]. The spatiotemporal network of the loading operation task consists of a set of vertices. Sum of arcs Composition. Each vertex is represented as ,in For spatial nodes, For time indexing, Represents a set of spatial nodes. Represents a time set. The set of spatial nodes. It includes the following four types of nodes:

[0087] (1) Waiting for a subset of nodes : Indicates the idle state of the loading operation (i.e., no material handling is performed during this period). Each spatial node With material grade This is a linked system, allowing only loading tasks of the corresponding level to enter this node. The scheduling of each loading task originates from its corresponding waiting node. Therefore, the source vertex of the loading operation is ( , );

[0088] (2) Material picking node subset : Indicates the direct material handling operation for loading tasks, and is related to spatial nodes. The relevant stack locations and schemes are respectively denoted as and ;

[0089] (3) Mixed node subset This indicates the mixing and reclaiming operation for the loading task, according to the plan. For two stack positions and The materials are mixed and blended.

[0090] (4) Sink node : Indicates the termination of the scheduling of the loading operation in the spatiotemporal network, such as Figure 3 Node 9. All loading operations ultimately converge at the confluence. The process ends here.

[0091] The spatiotemporal network of the loading operation task not only includes the four types of nodes mentioned above, but also connects vertices through different types of arcs to represent the possible movements and operations of the task. Specifically, it includes the following seven types of arcs:

[0092] (1) Waiting arc: This indicates a unit of time to wait on the node;

[0093] (2) Transition arc: This indicates the transition of a loading operation task from a waiting state to a working state (such as material picking or mixing). These movements from waiting nodes to material picking or mixing nodes do not consume actual time in the spatiotemporal network of loading operation tasks; Indicates the index of the material picking node; Indicates a time index;

[0094] (3) Material taking arc: This represents one unit of time at the material handling node. The unit material handling capacity is defined as the amount of material that the direct material handling scheme can process within one unit of time.

[0095] (4) Mixed arc: Each mixed arc Indicates from time Begin, Continue At each time step, in the stack position and The mixing process is carried out on the stack. The mixing process reduces the unit material handling rate at each stack location; therefore, mixing requires multiple time steps to produce at least one effective mixed material unit.

[0096] (5) Reset arc: This indicates that the operation has been performed at the stacking position and the system has returned to a waiting state.

[0097] (6) Convergence arc: This indicates that the loading operation task has been transferred from the waiting node to the receiving node, completing the entire loading operation cycle;

[0098] (7) Cancel arc: This indicates a cancelled loading operation. The process is transferred directly from the source waiting node to the sink node, without any further action taken during the entire planning period.

[0099] S2-2: Constructing the nodes and arcs of the spatiotemporal network of the material reclaimer;

[0100] Subsequently, a spatiotemporal network for the material reclaimer is constructed, denoted as... The spatiotemporal network of the material handling machine consists of a set of vertices. Sum of arcs Composition. Each vertex is represented as ,in For spatial nodes, Indexed by time. A collection of spatial nodes. It includes four types of nodes:

[0101] (1) Pile position node subset Because the material handling machine can only move on the track, it must occupy a specific stacking position at any given time. Indicates the index of the stack location set. Representing spatial nodes The stack position indicated;

[0102] (2) Subset of task nodes For each spatial node Define the corresponding work node to indicate that a material picking operation will be performed at this stack location;

[0103] (3) Material reclaimer source node : This represents the index of the set of reclaimers, each reclaimer having a known initial stack position. Therefore, in the spatiotemporal network of the material reclaimer, the material reclaimer... The source vertex is ,in ,and ; Indicates the initial stack position where the reclaimer source node is located;

[0104] (4) Sink node : Indicates the end of the spatiotemporal network trajectory of the material handling machine.

[0105] The spatiotemporal network of the material reclaimer not only includes the four types of nodes mentioned above, but also connects vertices through different types of arcs to represent the possible movements and operations of the material reclaimer. Specifically, it includes the following five types of arcs:

[0106] (1) Waiting arc: Vertex Connected to This indicates that the material reclaimer is at the stack node. The upper part remains idle;

[0107] (2) Moving arc: This indicates that the material reclaimer starts from the stack node. Move to another stack node ; Indicates a time index;

[0108] (3) Working arc: ,in ,and This indicates that the material handling machine has entered the working state from the idle state and is in the stack position. Above the time interval Perform material handling operations;

[0109] (4) Reset arc: ,in ,and This indicates that after completing the material picking operation, the material picker returns from the working state to the idle state at the same stack location;

[0110] (5) Convergence arc: This indicates that the material handling machine has completed scheduling and entered the sink node.

[0111] The spatiotemporal networks of the loading tasks and the reclaimers respectively constructed candidate paths for all loading tasks and reclaimers. Since the reclaiming operation at a certain stack location must be performed by the reclaimer, there is a coupling relationship between the two. Specifically, if a loading task falls within a certain time interval... to Use stacking position Then a material handling machine must pass through the working arc. ,in In other words, the operation arc in the spatiotemporal network of the material reclaimer. Coupled with a ship loading operation task arc in the spatiotemporal network of the ship loading operation task, denoted as ,in Includes all stack positions (satisfy Related material handling, mixing, or transformation arcs, Indicates all within the time interval Inside, the stack is in use. The loading operation task arc (including material taking arc, mixing arc and transfer arc). This represents a time index. Conversely, if the loading operation uses a certain operation arc in the spatiotemporal network of the loading operation task, then... Then the set Each arc in the process must be executed by a reclaimer, among which, This represents the spatiotemporal network operation arc of the material handling machine.

[0112] Step S2-3: Establish the following collaborative constraints for the spatiotemporal network of the loading operation task and the reclaimer:

[0113] (1)

[0114] (2)

[0115] (3)

[0116] (4)

[0117] (5)

[0118] (6)

[0119] (7)

[0120] (8)

[0121] (9)

[0122] (10)

[0123] (11)

[0124] (12)

[0125] In the formula, , , , , , , , , , , , , , , , , The definitions are as shown above; This indicates the arc index used to avoid symbol duplication; The spatiotemporal network spatial node index of the material handling machine is used to avoid symbol duplication; The weighting coefficient for the total port dwell time of the loading operation is 1. The material cost weighting coefficient for the loading operation is 0.00001. This parameter represents the quantity weighting parameter for tasks that do not meet the loading operation requirements; it is set to 160. Represents the set of loading operations, using index; Arc representing the spatiotemporal network of the loading operation task Processing time, Indicates the loading operation task Arc of the spatiotemporal network for ship loading operations hour It is 1 if it is true, otherwise it is 0; Arc representing the spatiotemporal network of the loading operation task Material costs; Arc representing the spatiotemporal network of the loading operation task The quantity of materials already taken; Indicates loading operation task Required quantity of materials; To represent a set of storage locations in a port yard, use index; Indicates the material taking arc or mixing arc At the stack The quantity of materials processed above; Indicates stack position The initial quantity of materials; Indicates a collection of material handling machines, using index; Indicates when the material handling machine Arc of the spatiotemporal network using a material handling machine hour It is 1 if it is true, otherwise it is 0; Indicates when arc Feeder During execution, due to non-intersecting constraints, there are no reclaimer spatiotemporal network arcs that cannot be executed simultaneously with the arc in the reclaimer spatiotemporal network.

[0126] Formula (1) represents the objective function minimizing the weighted summation term, which includes: the total waiting time for the loading tasks, the total material cost incurred in meeting the loading requirements, and the number of unmet loading tasks. Constraints (2)–(3) are flow balance constraints in the spatiotemporal network of the loading tasks, ensuring that each loading task is assigned to an effective path. Constraint (4) ensures that the total amount of material taken from each loading task on the selected arc is equal to the demand of that loading task. Constraint (5) limits the number of times the same loading task can be used at each stack to at most once. Constraint (6) ensures that the total amount of material taken from any stack does not exceed its initial amount of material. Constraint (7) enforces the coupling relationship between the spatiotemporal network flow of the loading tasks and the spatiotemporal network flow of the reclaimers, while ensuring that each stack can be occupied by at most one loading task at any given time. Constraints (8)–(9) are flow balance constraints in the spatiotemporal network of the reclaimers, ensuring that each reclaimer is assigned to a feasible trajectory. Constraint (10) is a non-crossing constraint during the material handling process. Finally, constraints (11)–(12) define the range of values ​​for the variables used in the coupled spatiotemporal network model.

[0127] Step S3: Based on the coupled spatiotemporal network model constructed in Step S2, the operation sequence of the loading operation tasks and the operation scheduling of the reclaimer within the planning period are uniformly optimized and solved. The operation sequence, mixing scheme and yard equipment scheduling scheme of each loading operation task within the planning period are generated, and the solution results are transformed into executable instructions for the yard equipment to achieve high-precision coordination of the yard equipment.

[0128] This invention directly solves the coupled spatiotemporal network model using the CPLEX solver within a C# environment. It employs the solver's built-in heuristic methods to comprehensively optimize the timing constraints, mixing constraints, and yard equipment occupancy conflicts of the loading operation. During the solution process, the coupled spatiotemporal network model outputs material handling paths, mixing schemes, and equipment scheduling results, providing complete and executable scheduling decision support for actual loading operations. Specifically:

[0129] Step S3-1: Solve for environment initialization and instantiation;

[0130] In the C# environment, the yard layout, loading operation tasks, and yard equipment parameter information collected in step S1 are loaded as input data, and the CPLEX solver instance is initialized based on the coupled spatiotemporal network model constructed in step S2. The variables, constraints, and objective function of the coupled spatiotemporal network model are fully imported into the solver at this stage, laying the foundation for subsequent solving.

[0131] Step S3-2: Solve the coupled spatiotemporal network model based on CPLEX;

[0132] The CPLEX solver is invoked to optimize and solve the coupled spatiotemporal network model, yielding a solution. The solver automatically handles job timing constraints, mixing constraints, and yard equipment conflict constraints within the model, using a built-in heuristic search mechanism to find the optimal scheduling scheme. This process requires no additional manual design of heuristics or manual parameter tuning, directly obtaining optimal or near-optimal scheduling results that satisfy spatiotemporal consistency. Two examples are shown in Table 1, and the solution efficiency is shown in Table 2. Gap represents the optimality gap reported by the CPLEX solver, used to measure the relative deviation between the current feasible solution and the global optimal solution, thus reflecting the convergence quality and solution stability of the model within a given solution time.

[0133] Table 1: Validation Examples of Coupled Spatiotemporal Network Models

[0134]

[0135] Table 2: Validation results of the coupled spatiotemporal network model

[0136]

[0137] Step S3-3: Scheduling scheme parsing and executable instruction generation;

[0138] After solving the problem, the optimal timing, mixing scheme execution, and reclaimer operation segment for each loading task are analyzed from the solution of the coupled spatiotemporal network model. The analytical results are further converted into executable instructions for the reclaimer's movement and operation plan. These instructions can be directly used to guide the on-site scheduling system, improving the overall efficiency of the loading operation.

[0139] This embodiment constructs a collaborative control optimization model encompassing stacker-reclaimer operation sequence, stacker-reclaimer scheduling, and belt conveyor scheduling based on the yard layout structure, operational task characteristics, and stacker-reclaimer operation process. A squeak wheel algorithm is introduced to solve the mixed-integer programming model, and the model is applied to specific embodiments, yielding high-quality scheduling results. Compared to the conventional method that relies on manual experience to arrange the operation sequence and passively adjusts belt conveyor start / stop according to operation switching, the method of this invention can obtain a higher-quality scheduling scheme in a shorter time, significantly reducing belt conveyor idling time and ineffective energy consumption while ensuring yard operation efficiency. Through collaborative optimization of stacker-reclaimer and belt conveyor operations, energy utilization efficiency is effectively improved, providing an implementable technical path and innovative solution for the green, refined, and intelligent operation of port yards.

[0140] It should be noted that the above embodiments are merely illustrative of the implementation methods of the present invention, but should not be construed as limiting the scope of the present invention. It should be pointed out that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

Claims

1. A method for port loading blending plan and yard operation line control, characterized in that, The port loading blending plan and yard operation line control method optimizes the blending scheduling by constructing a coordinated and consistent coupled spatiotemporal network model between the loading operation task sequence and the reclaimer operation scheduling. The method includes the following steps: Step S1: Collect port yard layout, ship loading operation task information during the planning period, and yard equipment parameter information to provide complete basic data for building a coupled spatiotemporal network model and realize high-precision collaborative scheduling of yard equipment; Step S2: Based on the port yard layout, ship loading task information, and yard equipment parameter information obtained in Step S1, construct the spatiotemporal network of ship loading tasks and the spatiotemporal network of the reclaimers. On this basis, construct a coupled spatiotemporal network model considering ship loading tasks and mixing constraints, uniformly linking the spatial location, operation sequence, mixing constraints, and equipment usability of each ship loading task within the yard to achieve coordinated optimization of operations, mixing, and transportation. Specifically: Step S2-1: Construct the nodes and arcs of the spatiotemporal network for the ship loading operation task; A space-time network of the loading operation task is established, denoted as The space-time network of the loading operation task is composed of a vertex set and an arc set Each vertex is expressed as , wherein is a space node, is a time index, denotes a space node set, denotes a time set; the space node set contains four types of nodes; the space-time network of the loading operation task further includes arcs connecting the vertices of different types to represent the movement and operation of the operation task; Step S2-2: Construct the nodes and arcs of the spatiotemporal network of the material reclaimer; Construct a spatiotemporal network for the material reclaimer, denoted as . The spatiotemporal network of the material handling machine consists of a set of vertices. Sum of arcs Composition; each vertex is represented as ,in For spatial nodes, For time indexing; spatial node set It contains four types of nodes; the spatiotemporal network of the reclaimer connects vertices through different types of arcs to represent the movement and operation of the reclaimer; The spatiotemporal network of the loading operation task and the spatiotemporal network of the reclaimer respectively construct all candidate paths for the loading operation task and the reclaimer. The material reclaiming operation performed at a certain stack position is executed by the reclaimer. There is a coupling relationship between the spatiotemporal network of the loading operation task and the spatiotemporal network of the reclaimer. Step S2-3: Establish the collaborative constraints of the spatiotemporal network of the loading operation task and the spatiotemporal network of the material reclaimer; Step S3: Based on the coupled spatiotemporal network model constructed in Step S2, the operation sequence of the loading tasks and the operation scheduling of the reclaimer within the planning period are uniformly optimized and solved. The operation sequence, mixing scheme and yard equipment scheduling scheme of each loading task within the planning period are generated, and the solution results are transformed into the execution instructions of the yard equipment to achieve high-precision coordination of the yard equipment.

2. The method for port loading and mixing planning and yard operation line control according to claim 1, characterized in that, The specific steps of S1 are as follows: Step S1-1: Collect basic information on the layout of the port yard; Collect information on the layout and spatial structure of the port yard, including the number of stockpiles, the number of stacks for each stockpile, the material grade, material cost, and initial material quantity for each stack; collect information on the number of reclaimer tracks, and record the total planning time and time step. Step S1-2: Collect information on loading operations during the planning period; Collect information on loading operations during the planning period, including the number of loading operations, workload, earliest start time, and blending scheme; Step S1-3: Collect yard equipment parameter information; The yard equipment consists of reclaimers; equipment parameter information includes the number of reclaimers, the reclaimer's reclaiming efficiency, the moving speed, and the stacking position range for loading operations.

3. The method for port loading and mixing planning and yard operation line control according to claim 2, characterized in that, In step S2-1, the spatial node set The four types of nodes included are as follows: (1) Waiting for a subset of nodes : Indicates the idle status of the loading operation task; each spatial node With material grade Associative, only loading tasks of the corresponding level are allowed to enter this node; the scheduling of each loading task starts from its corresponding waiting node. Initially, the source vertex of the loading operation is ( , ); (2) Material picking node subset : Indicates the direct material handling operation for loading tasks, and is related to spatial nodes. The relevant stack locations and schemes are respectively denoted as and ; (3) Mixed node subset This indicates the mixing and reclaiming operation for the loading task, according to the plan. For two stack positions and The materials are mixed and blended. (4) Sink node : Indicates the termination of the scheduling of loading operations in the spatiotemporal network; all loading operations ultimately converge at the sink vertex. The process ends here.

4. The method for port loading and mixing planning and yard operation line control according to claim 3, characterized in that, Step S2-1 includes the following seven types of arcs: (1) Waiting arc: This indicates a unit of time to wait on the node; (2) Transition arc: This indicates that the loading operation has changed from a waiting state to a working state; Indicates the index of the material picking node; Indicates a time index; (3) Material taking arc: , representing one unit of time at the material intake node; (4) Mixed arc: Each mixed arc Indicates from time Begin, Continue At each time step, in the stack position and Mixing and blending operations are carried out on top; (5) Reset arc: This indicates that the operation has been performed at the stacking position and the system has returned to a waiting state. (6) Convergence arc: This indicates that the loading operation task has been transferred from the waiting node to the receiving node, completing the entire loading operation cycle; (7) Cancel arc: This indicates a cancelled loading operation; the process is directly transferred from the source waiting node to the sink node, without any further action taken during the entire planning period.

5. The method for port loading and mixing planning and yard operation line control according to claim 4, characterized in that, In step S2-2, the spatial node set The four types of nodes included are as follows: (1) Pile position node subset Because the material handling machine can only move on the track, it must occupy a specific stacking position at any given time. Indicates the index of the stack location set. Representing spatial nodes The stack position indicated; (2) Subset of task nodes For each spatial node Define the corresponding work node to indicate that a material picking operation will be performed at this stack location; (3) Material reclaimer source node : This represents the index of the set of reclaimers, each reclaimer having a known initial stack position. Therefore, in the spatiotemporal network of the material reclaimer, the material reclaimer... The source vertex is ,in ,and ; Indicates the initial stack position where the reclaimer source node is located; (4) Sink node : Indicates the end of the spatiotemporal network trajectory of the material handling machine.

6. The method for port loading and mixing planning and yard operation line control according to claim 5, characterized in that, In step S2-2, the spatiotemporal network of the material handling machine includes the following five types of arcs: (1) Waiting arc: Vertex Connected to This indicates that the material reclaimer is at the stack node. The upper part remains idle; (2) Moving arc: This indicates that the material reclaimer starts from the stack node. Move to another stack node ; Indicates a time index; (3) Working arc: ,in ,and This indicates that the material handling machine has entered the working state from the idle state and is in the stack position. Above the time interval Perform material handling operations; (4) Reset arc: ,in ,and This indicates that after completing the material picking operation, the material picker returns from the working state to the idle state at the same stack location; (5) Convergence arc: This indicates that the material handling machine has completed scheduling and entered the sink node.

7. The method for port loading and mixing planning and yard operation line control according to claim 6, characterized in that, In step 2, the coupling relationship between the spatiotemporal network of the loading operation task and the spatiotemporal network of the material reclaimer is specifically as follows: If a certain loading operation is within the time frame to Use stacking position Then the material handling machine passes through the working arc. ,in In other words, the operation arc in the spatiotemporal network of the material reclaimer. Coupled with a ship loading operation task arc in the spatiotemporal network of the ship loading operation task, denoted as ,in Includes all stack positions Related material handling, mixing, or transformation arcs, Indicates all within the time interval Inside, the stack is in use. The loading operation task arc, Indicates a time index; Conversely, if the loading operation uses a certain operation arc in the spatiotemporal network of the loading operation task... Then the set Each arc in the process is executed by a reclaimer, in which, This represents the spatiotemporal network operation arc of the material handling machine.

8. The method for port loading and mixing planning and yard operation line control according to claim 7, characterized in that, The collaborative constraints established in steps S2-3 are as follows: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) In the formula, This indicates the arc index used to avoid symbol duplication; The spatiotemporal network spatial node index of the material handling machine is used to avoid symbol duplication; The weighting coefficient representing the total port dwell time for loading operations; The material cost weighting coefficient represents the material cost of the loading operation. This indicates the quantity weighting parameter for tasks that do not meet the requirements for loading operations. Represents the set of loading operations, using index; Arc representing the spatiotemporal network of the loading operation task Processing time, Indicates the loading operation task Arc of the spatiotemporal network for ship loading operations hour It is 1 if it is true, otherwise it is 0; Arc representing the spatiotemporal network of the loading operation task Material costs; Arc representing the spatiotemporal network of the loading operation task The quantity of materials already taken; Indicates loading operation task Required quantity of materials; To represent a set of storage locations in a port yard, use index; Indicates the material taking arc or mixing arc At the stack The quantity of materials processed above; Indicates stack position The initial quantity of materials; Indicates a collection of material handling machines, using index; Indicates when the material handling machine Arc of the spatiotemporal network using a material handling machine hour It is 1 if it is true, otherwise it is 0; Indicates when arc Feeder During execution, due to non-intersecting constraints, there are no reclaimer spatiotemporal network arcs that can be executed simultaneously with the arc in the reclaimer spatiotemporal network. Formula (1) represents the objective function minimization of the weighted summation term, which includes: total waiting time for loading operations, total material cost incurred in meeting loading requirements, and number of unmet loading operations. The constraints shown in formulas (2) and (3) are flow balance constraints in the spatiotemporal network of loading tasks, ensuring that each loading task is assigned to an effective path; the constraint shown in formula (4) ensures that the total amount of material taken from each loading task on the selected arc is equal to the demand of the loading task; the constraint shown in formula (5) restricts the use of each stack position by the same loading task to at most once; the constraint shown in formula (6) ensures that the total amount of material taken from any stack position does not exceed its initial amount of material; the constraint shown in formula (7) enforces the coupling relationship between the spatiotemporal network flow of the loading task and the spatiotemporal network flow of the material reclaimer, while ensuring that each stack position can be occupied by at most one loading task at any given time; formulas (8) and (9) are flow balance constraints in the spatiotemporal network of the material reclaimer, ensuring that each material reclaimer is assigned to a feasible trajectory; constraint (10) is the non-crossing constraint during the material reclaiming process; finally, the constraints shown in formulas (11) and (12) define the range of values ​​of the variables used in the coupled spatiotemporal network model.

9. A method for port loading and mixing planning and yard operation line control according to claim 8, characterized in that, The specific steps of step S3 are as follows: The CPLEX solver is used to solve the coupled spatiotemporal network model, performing overall optimization of the timing constraints, blending constraints, and yard equipment occupancy conflicts of the loading operation. During the solution process, the coupled spatiotemporal network model can output material handling paths, blending schemes, and equipment scheduling results, providing complete and executable scheduling decision support for actual loading operations. Specifically: Step S3-1: Solve for environment initialization and instantiation; The information on yard layout, loading operation tasks, and yard equipment parameters collected in step S1 is used as input data, and the CPLEX solver instance is initialized based on the coupled spatiotemporal network model constructed in step S2; the variables, constraints, and objective function of the coupled spatiotemporal network model are fully imported into the CPLEX solver at this stage. Step S3-2: Solve the coupled spatiotemporal network model based on CPLEX; The CPLEX solver is used to optimize and solve the coupled spatiotemporal network model, and the solution of the coupled spatiotemporal network model is obtained. Step S3-3: Parse the scheduling scheme and generate execution instructions; After the solution is obtained, the optimal operation sequence, mixing scheme execution and reclaimer operation section of each loading operation task are obtained analytically from the solution of the coupled spatiotemporal network model. The analysis results are further converted into instructions for the movement of the material handling machine and the execution of the work plan.

Images