CN-121981431-A - Port resource cooperative scheduling method, system, equipment and medium

Abstract

The invention discloses a port resource collaborative scheduling method, a system, equipment and a medium, wherein the method comprises the steps of obtaining historical resource data of a strategy library and a target port, wherein the strategy library comprises a plurality of different first solving strategies, carrying out calculation example generation on the historical resource data to obtain a plurality of port calculation examples, carrying out calculation example solving on all port calculation examples according to each first solving strategy in the strategy library to obtain a plurality of calculation example training groups, and carrying out parameter updating on an initialized strategy selection model according to all calculation example training groups to obtain the trained strategy selection model. According to the method, the target solving strategy is determined through the trained strategy selection model, and the resource computing example to be solved is solved, so that the efficiency and the effect of port resource collaborative scheduling can be improved. The invention relates to the technical field of port scheduling.

Inventors

• Lv Yaqiong
• WU ZIJUAN
• ZOU MINGKAI

Assignees

• 武汉理工大学

Dates

Publication Date

20260505

Application Date

20251217

Claims (10)

1. 1. The port resource cooperative scheduling method is characterized by comprising the following steps of: Acquiring a resource computing example to be solved of a target port; inputting the resource computing example to be solved into a trained strategy selection model to perform strategy selection, so as to obtain a target solving strategy; Carrying out an example solving on the resource example to be solved according to the target solving strategy to obtain a cooperative scheduling result of the target port; the trained strategy selection model is obtained through training by the following steps: Acquiring historical resource data of a strategy library and a target port, wherein the strategy library comprises a plurality of different first solving strategies; Performing calculation generation on the historical resource data to obtain a plurality of harbor calculation examples; Carrying out example solving on all the harbor examples according to each first solving strategy in the strategy library to obtain a plurality of example training groups, wherein each example training group comprises the harbor examples and a plurality of example labels, and each example label corresponds to one first solving strategy; And according to all the example training groups, updating parameters of the initialized strategy selection model to obtain the trained strategy selection model.
2. 2. The method of claim 1, wherein performing an instance generation on the historical resource data to obtain a plurality of port instances comprises: according to the historical resource data, basic parameter data and key parameter data are obtained; Carrying out statistical learning on the key parameter data to obtain a parameter distribution model; According to the parameter distribution model, performing parameter generation processing to obtain an example parameter; and obtaining the harbor calculation example according to the calculation example parameters and the basic parameter data.
3. 3. The method of claim 1, wherein said performing an example solution on all of said port examples according to each first solution strategy in said strategy library to obtain a plurality of example training sets comprises: Carrying out strategy solving on the harbor computing examples according to all the first solving strategies to obtain a plurality of computing example target values, wherein each computing example target value is used for representing an objective function value of a computing example solution obtained by solving the harbor computing examples under the corresponding first solving strategies; According to all the example target values, generating labels of the harbor examples to obtain a plurality of example labels; And obtaining the example training set according to the port example and all the example labels.
4. 4. A method according to claim 3, wherein said performing a policy solution on said port computing examples according to all of said first solution policies to obtain a plurality of computing example target values comprises: acquiring a second solving strategy, wherein the second solving strategy is used for representing all first solving strategies which are not subjected to strategy solving in the first solving strategies; and carrying out strategy solving on the harbor calculation example according to the second solving strategy to obtain the calculation example target value.
5. 5. The method of claim 4, wherein said performing a policy solution on said port computing case according to said second solution policy to obtain said computing case target value comprises: carrying out data coding on the port calculation example to obtain a coding data structure; Carrying out strategy initialization on the coded data structure according to the second solving strategy to obtain an initial individual; and performing individual optimizing treatment on the initial individual to obtain the example target value.
6. 6. The method according to claim 5, wherein the performing individual optimization processing on the initial individual to obtain the calculation target value includes: Obtaining an individual objective function and a current solution, wherein the current solution is an initial solution corresponding to the initial individual or a target optimization solution in the previous individual optimizing process; performing individual solution optimization on the current solution to obtain an intermediate optimized solution; Performing function evaluation on the current solution and the intermediate optimal solution according to the individual objective function to obtain an objective function value of the current solution and an objective function value of the intermediate optimal solution; comparing the objective function value of the current solution with the objective function value of the intermediate optimized solution to obtain a function value comparison result; If the function value comparison result is that the objective function value of the current solution is larger than the objective function value of the intermediate optimization solution, the current solution is determined to be the objective optimization solution in the current individual optimizing process, or if the function value comparison result is that the objective function value of the current solution is smaller than or equal to the objective function value of the intermediate optimization solution, the intermediate optimization solution is determined to be the objective optimization solution in the current individual optimizing process; and obtaining the example target value according to a target optimization solution in the current individual optimizing process.
7. 7. The method of claim 4, wherein said performing a policy solution on said port computing case according to said second solution policy to obtain said computing case target value comprises: Acquiring a population fitness function according to the second solving strategy; carrying out data coding on the port calculation example to obtain a coding data structure; Determining an initial population according to the encoded data structure; and carrying out population cross variation on the initial population according to the population fitness function to obtain the example target value.
8. 8. A port resource co-scheduling system, comprising: the first processing unit is used for acquiring a resource computing example to be solved of the target port; The second processing unit is used for inputting the resource computing example to be solved into a trained strategy selection model to perform strategy selection, so as to obtain a target solving strategy; The third processing unit is used for carrying out calculation case solving on the resource calculation case to be solved according to the target solving strategy to obtain a cooperative scheduling result of the target port; the trained strategy selection model is obtained through training by the following steps: Acquiring historical resource data of a strategy library and a target port, wherein the strategy library comprises a plurality of different first solving strategies; Performing calculation generation on the historical resource data to obtain a plurality of harbor calculation examples; Carrying out example solving on all the harbor examples according to each first solving strategy in the strategy library to obtain a plurality of example training groups, wherein each example training group comprises the harbor examples and a plurality of example labels, and each example label corresponds to one first solving strategy; And according to all the example training groups, updating parameters of the initialized strategy selection model to obtain the trained strategy selection model.
9. 9. An electronic device, comprising: At least one processor; At least one memory for storing at least one program; the at least one program, when executed by the at least one processor, causes the at least one processor to implement the method of any of claims 1-7.
10. 10. A computer readable storage medium, in which a processor executable program is stored, characterized in that the processor executable program is for implementing the method according to any of claims 1-7 when being executed by the processor.

Description

Port resource cooperative scheduling method, system, equipment and medium Technical Field The invention relates to the technical field of port scheduling, in particular to a port resource collaborative scheduling method, a port resource collaborative scheduling system, port resource collaborative scheduling equipment and port resource collaborative scheduling medium. Background In order to ensure the sailing safety of the ship under the harbor sailing condition, the harbor entering and leaving operation processes of the ship at the harbor are uniformly and cooperatively scheduled by a harbor scheduling center. At present, a port scheduling center usually respectively establishes a port berth plan and a tug scheduling plan based on the waiting sequence of the ship, and performs scheduling decision based on the established port berth plan and tug scheduling plan to realize cooperative scheduling of port resources. Accordingly, there is a further need for solving and optimizing the problems associated with the related art. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the related art to a certain extent. Therefore, an object of the embodiments of the present invention is to provide a method, a system, an apparatus, and a medium for collaborative scheduling of port resources, where the method may improve efficiency and effect of collaborative scheduling of port resources. In order to achieve the technical purpose, the technical scheme adopted by the embodiment of the application comprises the following steps: in a first aspect, an embodiment of the present application provides a port resource cooperative scheduling method, including: Acquiring a resource computing example to be solved of a target port; inputting the resource computing example to be solved into a trained strategy selection model to perform strategy selection, so as to obtain a target solving strategy; Carrying out an example solving on the resource example to be solved according to the target solving strategy to obtain a cooperative scheduling result of the target port; the trained strategy selection model is obtained through training by the following steps: Acquiring historical resource data of a strategy library and a target port, wherein the strategy library comprises a plurality of different first solving strategies; Performing calculation generation on the historical resource data to obtain a plurality of harbor calculation examples; Carrying out example solving on all the harbor examples according to each first solving strategy in the strategy library to obtain a plurality of example training groups, wherein each example training group comprises the harbor examples and a plurality of example labels, and each example label corresponds to one first solving strategy; And according to all the example training groups, updating parameters of the initialized strategy selection model to obtain the trained strategy selection model. In addition, the method according to the above embodiment of the present application may further have the following additional technical features: further, in an embodiment of the present application, the performing an example generation on the historical resource data to obtain a plurality of harbor examples includes: according to the historical resource data, basic parameter data and key parameter data are obtained; Carrying out statistical learning on the key parameter data to obtain a parameter distribution model; According to the parameter distribution model, performing parameter generation processing to obtain an example parameter; and obtaining the harbor calculation example according to the calculation example parameters and the basic parameter data. Further, in an embodiment of the present application, the performing an example solution on all the harbor examples according to each first solution policy in the policy repository to obtain a plurality of example training sets includes: Carrying out strategy solving on the harbor computing examples according to all the first solving strategies to obtain a plurality of computing example target values, wherein each computing example target value is used for representing an objective function value of a computing example solution obtained by solving the harbor computing examples under the corresponding first solving strategies; According to all the example target values, generating labels of the harbor examples to obtain a plurality of example labels; And obtaining the example training set according to the port example and all the example labels. Further, in an embodiment of the present application, the performing, according to all the first solution policies, policy solution on the port computing case to obtain a plurality of computing case target values includes: acquiring a second solving strategy, wherein the second solving strategy is used for representing all first solving strategies which are not s