CN-120598296-B - Digital twin factory construction interaction method and system based on AI intelligent agent drive

CN120598296BCN 120598296 BCN120598296 BCN 120598296BCN-120598296-B

Abstract

The invention relates to the technical field of data processing analysis, in particular to a digital twin factory construction interaction method and system based on AI intelligent agent drive, wherein the disturbance degree of the insertion of an emergency order to each production line is analyzed through the dynamic data of the production line, the time sequence data used by equipment and the order characteristic data; the method comprises the steps of establishing an order adaptation degree analysis model through product line dynamic data and order characteristic data, analyzing adaptation degree of an emergency order and each product line, importing disturbance degree analysis results of the emergency order on each product line by insertion of the emergency order and adaptation degree analysis results of the emergency order and each product line into a product line dynamic bearing capacity assessment model, assessing dynamic bearing capacity of each product line when the emergency order is inserted, and inserting the emergency order according to assessment results of dynamic bearing capacity of each product line when the emergency order is inserted. The method can avoid capacity conflict or efficiency loss caused by blind order insertion, reduce resource mismatch cost and improve emergency order delivery response speed.

Inventors

QIN JUNQI
Zou Chunan
ZHANG QIDONG
SU YUEQI
Cui Tinghe

Assignees

北数博瑞数字科技(江苏)有限责任公司

Dates

Publication Date: 20260508
Application Date: 20250604

Claims (6)

1. The digital twin plant construction interaction method based on AI intelligent agent driving is characterized by comprising the following steps: S1, acquiring production line dynamic data and equipment use time sequence data of each production line through an AI intelligent agent, and simultaneously acquiring order feature data of an emergency order; S2, constructing an order insertion disturbance degree analysis model based on the production line dynamic data, the equipment use time sequence data and the order feature data, and analyzing the disturbance degree of the insertion of the emergency order to each production line; s3, constructing an order adaptation degree analysis model based on the dynamic data of the production lines and the order characteristic data, and analyzing the adaptation degree of the emergency order and each production line; S4, constructing a production line bearing capacity assessment model, importing disturbance degree analysis results of the emergency order insertion on each production line and adaptation degree analysis results of the emergency order and each production line into the production line dynamic bearing capacity assessment model, and assessing the dynamic bearing capacity of each production line when the emergency order is inserted; s5, inserting the emergency order according to the evaluation result of the dynamic bearing capacity of each production line when the emergency order is inserted; in the step S2, the disturbance degree of the insertion of the emergency order to each production line is analyzed, which includes the following specific steps: S21, extracting dynamic data of a production line and equipment use time sequence data and order feature data; S22, based on the dynamic data of the production line, the time sequence data of equipment and the order characteristic data, an order insertion disturbance degree analysis model is constructed, the disturbance degree of the insertion of the emergency order to each production line is analyzed, and a disturbance degree analysis result of the insertion of the emergency order to each production line is obtained; the disturbance degree calculation formula of the insertion of the emergency order to the current production line is as follows: ; Wherein RD is the disturbance degree of the insertion of the emergency order to the current production line, sc is the time sequence conflict degree of the insertion of the emergency order to the current production line, and Zz is the resource competition intensity degree of the insertion of the emergency order to the current production line; the construction process of the order insertion disturbance degree analysis model in the step S22 includes the following specific steps: S221, analyzing the time sequence conflict degree of the insertion of the emergency order to the current production line based on the production line dynamic data, the equipment use time sequence data and the order feature data, and obtaining a time sequence conflict degree analysis result of the insertion of the emergency order to the current production line; The calculation formula of the time sequence conflict degree is as follows: ; where Sc is the degree of timing conflict of the insertion of the emergency order to the current production line, For the historical emergency order arrival factor of the current production line in the dynamic data of the production line, D is the number of devices to be used for emergency order production on the current production line in the order feature data, txd is the required duration of the emergency order production in the D-th device to be used in the order feature data, tyd is the idle duration of the D-th device to be used for emergency order production in the device use time sequence data, max () is the maximum function in brackets, and D is any one of 1 to D; s222, analyzing the resource competition intensity of the insertion of the emergency order to the current production line based on the dynamic data of the production line and the order feature data, and obtaining an analysis result of the resource competition intensity of the insertion of the emergency order to the current production line; the calculation formula of the resource competition intensity is as follows: ; Where Zz is the severity of the resource competition of the current production line due to the insertion of the emergency order, pj is the index of the probability of collision of the j-th resource usage required by the emergency order on the current production line, Representing the operation of carrying out the continuous multiplication on the data in brackets, wherein n is the number of resource types required by the emergency order on the current production line in the order characteristic data, and j is any one of 1 to n; Wherein, the calculation formula using the collision probability index is: ; wherein Pj is the use conflict probability index of the j-th resource required by the emergency order on the current production line, rj is the required quantity of the j-th resource required by the emergency order on the current production line in the order characteristic data, aj is the current stock of the j-th resource required by the emergency order on the current production line in the production line dynamic data, and Uj is the historical average consumption rate of the j-th resource required by the emergency order on the current production line in the production line dynamic data.
2. The AI-agent-driven digital twin plant construction interaction method according to claim 1, wherein the step S3 of analyzing the degree of adaptation of the emergency order to each production line comprises the following specific steps: S31, extracting dynamic data of a production line and order feature data; S32, constructing an order adaptation degree analysis model based on the dynamic data of the production lines and the order characteristic data, and analyzing the adaptation degree of the emergency order and each production line to obtain an adaptation degree analysis result of the emergency order and each production line; the calculation formula of the adaptation degree of the emergency order and the current production line is as follows: ; wherein SP is the adaptation degree of the emergency order and the current production line, hx is the buffer absorption capacity of the current production line when the emergency order is inserted, rh is the flexible production exchange capacity of the current production line when the emergency order is inserted, rt is the total required quantity of all resources required by the emergency order on the current production line in order characteristic data, and At is the total current stock quantity of all resources required by the emergency order on the current production line in the dynamic data of the production line.
3. The AI-agent-driven digital twin plant construction interaction method according to claim 2, wherein the construction process of the order adaptation degree analysis model in step S32 comprises the following specific steps: s321, analyzing the buffer absorption capacity of each production line during emergency order insertion based on the dynamic data of the production line to obtain a buffer absorption capacity analysis result of each production line during emergency order insertion; S322, analyzing the flexible production capacity of each production line during emergency order insertion based on the dynamic data of the production lines and the order characteristic data, and obtaining a flexible production capacity analysis result of each production line during emergency order insertion.
4. The AI-agent-driven digital twin plant construction interaction method according to claim 3, wherein the construction of the line bearing capacity assessment model in step S4 comprises the following specific steps: S41, extracting disturbance degree analysis results of the insertion of the emergency order to each production line and adaptation degree analysis results of the emergency order and each production line; S42, according to disturbance degree analysis results of the emergency order insertion on each production line and adaptation degree analysis results of the emergency order and each production line, evaluating dynamic bearing capacity of each production line when the emergency order is inserted, and obtaining dynamic bearing capacity evaluation results of each production line when the emergency order is inserted; The evaluation formula of the dynamic bearing capacity is as follows: ; Where DC is the current line dynamic load-bearing capacity at the time of emergency order insertion.
5. The AI-agent-driven digital twin plant construction interaction method according to claim 4, wherein the step S5 is to insert the emergency order according to the evaluation result of the dynamic bearing capacity of each production line when the emergency order is inserted, and comprises the following specific steps: S51, acquiring dynamic bearing capacity evaluation results of all production lines obtained by evaluation during the insertion of the emergency order; S52, extracting the production line corresponding to the maximum value in the dynamic bearing capacity evaluation results of all the production lines when the evaluated emergency order is inserted as the production line for inserting the emergency order.
6. An AI-agent-driven digital twin plant construction interaction system implemented on the basis of the AI-agent-driven digital twin plant construction interaction method according to any one of claims 1-5, characterized in that the system comprises: The data acquisition module is used for acquiring the dynamic data of the production line and the equipment use time sequence data of each production line through the AI intelligent agent and acquiring the order characteristic data of the emergency order at the same time; the order insertion disturbance degree analysis module is used for constructing an order insertion disturbance degree analysis model based on the production line dynamic data, the equipment use time sequence data and the order characteristic data and analyzing the disturbance degree of the insertion of the emergency order to each production line; The order adaptation degree analysis module is used for constructing an order adaptation degree analysis model based on the dynamic data of the production lines and the order characteristic data and analyzing the adaptation degree of the emergency order and each production line; The production line dynamic bearing capacity assessment module is used for constructing a production line bearing capacity assessment model, importing disturbance degree analysis results of the insertion of the emergency order to each production line and adaptation degree analysis results of the emergency order and each production line into the production line dynamic bearing capacity assessment model, and assessing the dynamic bearing capacity of each production line when the emergency order is inserted; the emergency order insertion module is used for inserting the emergency order according to the evaluation result of the dynamic bearing capacity of each production line during the emergency order insertion; the control module is used for controlling the operation of the data acquisition module, the order insertion disturbance degree analysis module, the order adaptation degree analysis module, the production line dynamic bearing capacity assessment module and the emergency order insertion module.

Description

Digital twin factory construction interaction method and system based on AI intelligent agent drive Technical Field The invention relates to the technical field of data processing analysis, in particular to a digital twin factory construction interaction method and system based on AI intelligent agent driving. Background With the advancement of industry 4.0 wave and the deepening of intelligent manufacturing concepts, the traditional manufacturing industry is experiencing unprecedented reform. In order for traditional production plants to cope with increasingly aggressive market competition, it is highly desirable to construct a highly flexible, intelligent response production system. The digital twin technology is used as a key technology for realizing the deep fusion of the physical world and the information world, and provides an important approach for the intelligent upgrading of a factory by constructing a virtual mirror image of a physical entity. Traditional digital twin plant construction is often limited to static modeling and offline analysis, and it is difficult to capture dynamic changes in the production process in real time. Especially when dealing with emergency order insertion such emergencies, the decision mechanism still depends on manual intervention or preset rules, and the autonomous learning capability is lacking. Meanwhile, the traditional digital twin system has obvious short plates on the data fusion layer, and multiple data in the factory production process are often subjected to splitting treatment and cannot form an organic whole, so that the response to complex production scenes such as emergency order insertion is not attractive. Meanwhile, the existing construction interaction method of the digital twin factory is used for sudden insertion of the orders. The cascade effect of time sequence conflict and resource competition of the urgent order insertion on the production line cannot be quantified in real time, which causes frequent trapping of local optimization in dynamic conflict of an order scheduling system, meanwhile, the prior art fuses equipment capacity and order characteristics by manually setting weights, and ignores the nonlinear saturation characteristics of a buffer area on the production line and the entropy increasing effect of the yield change fluctuation, so that the misjudgment rate of the high-flexibility production line is continuously higher in the analysis process. In order to solve the problems, the application designs a digital twin plant construction interaction method and system based on AI intelligent agent driving. Disclosure of Invention In order to overcome the defects and shortcomings in the prior art, the invention provides a digital twin factory construction interaction method and system based on AI intelligent agent driving, which are used for analyzing disturbance degree of emergency order insertion on production lines and adaptation degree of the emergency order to the production lines, further realizing evaluation of dynamic bearing capacity of the production lines when the emergency order is inserted, and inserting the emergency order according to evaluation results of the dynamic bearing capacity of the production lines when the emergency order is inserted. The method can avoid capacity conflict or efficiency loss caused by blind order insertion, reduce resource mismatch cost and improve emergency order delivery response speed. In order to achieve the above purpose, the present invention adopts the following technical scheme: in a first aspect, an embodiment of the present invention provides an AI-agent-driven digital twin plant construction interaction method, including the steps of: S1, acquiring production line dynamic data and equipment use time sequence data of each production line through an AI intelligent agent, and simultaneously acquiring order feature data of an emergency order; S2, constructing an order insertion disturbance degree analysis model based on the production line dynamic data, the equipment use time sequence data and the order feature data, and analyzing the disturbance degree of the insertion of the emergency order to each production line; s3, constructing an order adaptation degree analysis model based on the dynamic data of the production lines and the order characteristic data, and analyzing the adaptation degree of the emergency order and each production line; S4, constructing a production line bearing capacity assessment model, importing disturbance degree analysis results of the emergency order insertion on each production line and adaptation degree analysis results of the emergency order and each production line into the production line dynamic bearing capacity assessment model, and assessing the dynamic bearing capacity of each production line when the emergency order is inserted; S5, inserting the emergency order according to the evaluation result of the dynamic bearing capacity of each production line when