CN-120875363-B - Multi-model-adaptation-oriented intelligent material dynamic resource scheduling optimization method and system

CN120875363BCN 120875363 BCN120875363 BCN 120875363BCN-120875363-B

Abstract

The invention relates to an intelligent material dynamic resource scheduling optimization method and system for multi-model adaptation, and belongs to the technical field of resource scheduling. The method comprises the steps of obtaining equipment material multisource data and obtaining the equipment material multisource standard data through data preprocessing, processing the equipment material multisource standard data through knowledge graph modeling to obtain a process conflict rule base, optimizing the process conflict rule base and the equipment material multisource standard data through a dynamic conflict process to obtain a feasible resource scheduling solution space, processing the feasible resource scheduling solution space and the equipment material multisource standard data through multi-objective dynamic scheduling decision to obtain a material resource optimizing scheduling instruction, executing the material resource optimizing scheduling instruction through edge cloud to obtain material resource scheduling feedback information and obtaining a material resource optimizing rescheduling instruction through parameter self-evolution processing, and realizing intelligent material dynamic resource scheduling optimization of multi-number adaptation.

Inventors

LI TAOTAO
ZHANG XUEGANG

Assignees

上海安托信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20250715

Claims (6)

1. The intelligent material dynamic resource scheduling optimization method for multi-model adaptation is characterized by comprising the following steps of: S1, acquiring equipment material multi-source data, and preprocessing the equipment material multi-source data through data to obtain equipment material multi-source standard data; s2, processing the multi-source standard data of the equipment material through knowledge graph modeling to obtain a process conflict rule base; s3, optimizing and processing the process conflict rule base and the equipment material multi-source standard data through a dynamic conflict process to obtain a feasible resource scheduling solution space; the generation process of the feasible resource scheduling solution space in the S3 is as follows: s301, processing the process conflict rule base and the equipment material multisource standard data through structural integration to obtain a process material resource mapping matrix; s302, processing the process material resource mapping matrix through demand matching to obtain task allocation material resource information; s303, optimizing the task allocation material resource information through dynamic conflict detection to obtain a feasible task resource allocation scheme set; s304, evaluating the feasible task resource allocation scheme set through space-time resource mapping to obtain a feasible resource scheduling solution space; The process of generating the process material resource mapping matrix in S301 includes: s301-1, processing the process conflict rule base through conflict relation semantic analysis to obtain a structured conflict relation diagram; S301-2, processing the multi-source standard data of the equipment material through multi-dimensional resource feature fusion to obtain a material resource capacity feature matrix; s301-3, processing the structural conflict relation graph and the material resource capacity feature matrix through conflict association modeling to obtain a process material resource mapping matrix; S4, processing the feasible resource scheduling solution space and the equipment material multi-source standard data through a multi-target dynamic scheduling decision to obtain a material resource optimization scheduling instruction; S5, cooperatively executing the material resource optimization scheduling instruction through the edge cloud to obtain material resource scheduling feedback information, and processing the material resource scheduling feedback information through parameter self-evolution to obtain a material resource optimization rescheduling instruction; The generation process of the material resource optimization rescheduling instruction in the S5 is as follows: s501, optimizing and processing the material resource scheduling feedback information through scheduling feedback parameters to obtain a resource scheduling correction factor; S502, acquiring resource scheduling dynamic data, and processing the resource scheduling correction factors and the resource scheduling dynamic data through cloud rescheduling to obtain a material resource optimization rescheduling instruction; the generating process of the resource scheduling correction factor in S501 is as follows: s501-1, extracting and processing the material resource scheduling feedback information through multidimensional feedback characteristics to obtain a structural scheduling deviation characteristic vector; s501-2, obtaining resource scheduling parameter updating information by optimizing and processing the structural scheduling deviation feature vector through incremental parameters; and S501-3, verifying the resource scheduling parameter updating information through knowledge distillation to generate a resource scheduling correction factor.
2. The method for optimizing dynamic resource scheduling of intelligent materials according to claim 1, wherein the generating process of the multi-source standard data of the equipment materials in the step S1 is as follows: s101, acquiring equipment material multi-source data through an edge data acquisition engine; S102, processing the equipment material multi-source data through stream cleaning to obtain equipment material multi-source cleaning data; and S103, processing the multi-source cleaning data of the equipment materials through time stamp normalization to obtain multi-source standard data of the equipment materials.
3. The method for optimizing dynamic resource scheduling of intelligent materials according to claim 1, wherein the process of generating the process conflict rule base in step S2 is as follows: s201, extracting and processing the equipment material multi-source standard data through entity relation to obtain an equipment material knowledge triplet set; S202, processing the equipment material knowledge triplet set through a map fusion engine to obtain an equipment material knowledge map; S203, processing the equipment material knowledge graph through rule reasoning to obtain an original process conflict rule base; s204, optimizing and processing the original process conflict rule base through association rule mining to obtain a process conflict rule base.
4. The method for optimizing dynamic resource scheduling of intelligent materials according to claim 1, wherein the generating process of the material resource optimizing scheduling instruction in step S4 is as follows: s401, quantitatively processing the feasible resource scheduling solution space and the equipment material multi-source standard data through multi-objective evaluation to obtain a resource scheduling comprehensive evaluation matrix; s402, obtaining an optimal resource scheduling scheme by processing the comprehensive evaluation matrix of the resource scheduling through multi-objective optimization; S403, obtaining a material resource optimization scheduling instruction by analyzing the optimal resource scheduling scheme and mapping the optimal resource scheduling scheme to a device material control system.
5. The method for optimizing dynamic resource scheduling of intelligent material according to claim 4, wherein the generating process of the comprehensive evaluation matrix of resource scheduling in step S401 is as follows: S401-1, processing the equipment material multisource standard data through multi-target dimension definition to obtain a material resource quantitative evaluation reference rule; s401-2, traversing the feasible resource scheduling solution space through objective function calculation to obtain a material resource initial evaluation matrix; S401-3, processing the initial evaluation matrix of the material resources and the quantitative evaluation reference rule of the material resources through target normalization and weighted aggregation to obtain a comprehensive evaluation matrix of resource scheduling.
6. An intelligent material dynamic resource scheduling optimization system oriented to multi-model adaptation, which is applied to the intelligent material dynamic resource scheduling optimization method as claimed in any one of claims 1-5, and is characterized by comprising an equipment material processing module, a process conflict rule base construction module, a scheduling conflict processing module, a dynamic scheduling decision module and a feedback rescheduling module; The equipment material processing module is used for acquiring equipment material multi-source data and preprocessing the equipment material multi-source data through data to obtain equipment material multi-source standard data; The process conflict rule base construction module is used for processing the equipment material multi-source standard data through knowledge graph modeling to obtain a process conflict rule base; The scheduling conflict processing module is used for optimizing and processing the process conflict rule base and the equipment material multi-source standard data through a dynamic conflict process to obtain a feasible resource scheduling solution space; the generation process of the feasible resource scheduling solution space is as follows: Processing the process conflict rule base and the equipment material multi-source standard data through structural integration to obtain a process material resource mapping matrix; processing the process material resource mapping matrix through demand matching to obtain task allocation material resource information; Optimizing the task allocation material resource information through dynamic conflict detection to obtain a feasible task resource allocation scheme set; Evaluating the feasible task resource allocation scheme set through space-time resource mapping to obtain a feasible resource scheduling solution space; the process material resource mapping matrix is generated by the following steps: processing the process conflict rule base through conflict relation semantic analysis to obtain a structured conflict relation graph; Processing the multi-source standard data of the equipment material through multi-dimensional resource feature fusion to obtain a material resource capacity feature matrix; Processing the structural conflict relation graph and the material resource capacity feature matrix through conflict association modeling to obtain a process material resource mapping matrix; The dynamic scheduling decision module is used for processing the feasible resource scheduling solution space and the equipment material multi-source standard data through multi-objective dynamic scheduling decision to obtain a material resource optimization scheduling instruction; the feedback rescheduling module is used for cooperatively executing the material resource optimization scheduling instruction through the edge cloud to obtain material resource scheduling feedback information, and processing the material resource scheduling feedback information through parameter self-evolution to obtain the material resource optimization rescheduling instruction; the generation process of the material resource optimization rescheduling instruction comprises the following steps: Optimizing the material resource scheduling feedback information through scheduling feedback parameters to obtain a resource scheduling correction factor; Acquiring resource scheduling dynamic data, and processing the resource scheduling correction factors and the resource scheduling dynamic data through cloud rescheduling to obtain a material resource optimization rescheduling instruction; The generation process of the resource scheduling correction factor comprises the following steps: extracting and processing the material resource scheduling feedback information through multidimensional feedback characteristics to obtain a structural scheduling deviation characteristic vector; the structural scheduling deviation feature vector is optimized through incremental parameters to obtain resource scheduling parameter updating information; and verifying the resource scheduling parameter updating information through knowledge distillation to generate a resource scheduling correction factor.

Description

Multi-model-adaptation-oriented intelligent material dynamic resource scheduling optimization method and system Technical Field The invention belongs to the technical field of resource scheduling, and particularly relates to an intelligent material dynamic resource scheduling optimization method and system for multi-model adaptation. Background The existing intelligent material manufacturing system faces a fundamental bottleneck in multi-type parallel scheduling, wherein a static scheduling mechanism based on fixed rules or offline optimization is difficult to respond to real-time working condition fluctuation (such as equipment failure and emergency bill insertion), so that a scheduling plan is frequently invalid, meanwhile, serious adaptation conflict is caused in multi-type mixed line production due to ignoring strong constraint of process parameters between materials and equipment, and the equipment empty rate is directly pushed up. Meanwhile, the material performance data and the production scheduling system are cut to form a data island, so that the resource allocation lacks real-time data driving, and the technological parameters and the production requirements are continuously misplaced. The series of problems form vicious circle, and severely restrict the efficiency and reliability of manufacturing of multi-model intelligent materials. Therefore, research on a method and a system for optimizing dynamic resource scheduling of intelligent materials for multi-model adaptation are needed. Disclosure of Invention In order to solve the problems in the prior art, the invention provides an intelligent material dynamic resource scheduling optimization method and system for multi-model adaptation. The aim of the invention can be achieved by the following technical scheme: a multi-model adaptation-oriented intelligent material dynamic resource scheduling optimization method comprises the following steps: S1, acquiring equipment material multi-source data, and preprocessing the equipment material multi-source data through data to obtain equipment material multi-source standard data; s2, processing the multi-source standard data of the equipment material through knowledge graph modeling to obtain a process conflict rule base; s3, optimizing and processing the process conflict rule base and the equipment material multi-source standard data through a dynamic conflict process to obtain a feasible resource scheduling solution space; S4, processing the feasible resource scheduling solution space and the equipment material multi-source standard data through a multi-target dynamic scheduling decision to obtain a material resource optimization scheduling instruction; And S5, cooperatively executing the material resource optimization scheduling instruction through the edge cloud to obtain material resource scheduling feedback information, and processing the material resource scheduling feedback information through parameter self-evolution to obtain a material resource optimization rescheduling instruction. Preferably, the generating process of the multi-source standard data of the device material in the step S1 is as follows: s101, acquiring equipment material multi-source data through an edge data acquisition engine; S102, processing the equipment material multi-source data through stream cleaning to obtain equipment material multi-source cleaning data; and S103, processing the multi-source cleaning data of the equipment materials through time stamp normalization to obtain multi-source standard data of the equipment materials. Preferably, the generating process of the process conflict rule base in step S2 is as follows: s201, extracting and processing the equipment material multi-source standard data through entity relation to obtain an equipment material knowledge triplet set; S202, processing the equipment material knowledge triplet set through a map fusion engine to obtain an equipment material knowledge map; S203, processing the equipment material knowledge graph through rule reasoning to obtain an original process conflict rule base; s204, optimizing and processing the original process conflict rule base through association rule mining to obtain a process conflict rule base. Preferably, the generating process of the feasible resource scheduling solution space in the step S3 is: s301, processing the process conflict rule base and the equipment material multisource standard data through structural integration to obtain a process material resource mapping matrix; s302, processing the process material resource mapping matrix through demand matching to obtain task allocation material resource information; s303, optimizing the task allocation material resource information through dynamic conflict detection to obtain a feasible task resource allocation scheme set; S304, the feasible task resource allocation scheme set is evaluated through space-time resource mapping to obtain a feasible resource scheduling solution space