Search

CN-121979156-A - Real-time data decision support system of production line based on cloud edge cooperation

CN121979156ACN 121979156 ACN121979156 ACN 121979156ACN-121979156-A

Abstract

The invention discloses a cloud edge cooperation-based real-time data decision support system of a production line, which relates to the technical field of industrial Internet of things and intelligent manufacturing, and comprises a plurality of edge intelligent nodes, a plurality of edge intelligent nodes and a plurality of edge intelligent nodes, wherein the edge intelligent nodes are distributed at manufacturing resources of the production line and are used for acquiring multi-source heterogeneous real-time data and executing local decisions; and the cloud decision platform is used for executing global data analysis and strategy generation. The cloud edge cooperation-based real-time data decision support system of the production line converts a traditional passive exception handling mode into an active prediction and cooperation decision mode. The system can continuously sense the production line state in real time, forecast and locally intervene before the potential problem influence is enlarged, and rapidly adjust multiparty resources through an intelligent cooperative mechanism to form a solution when needed, so that the overall response time from abnormal occurrence to implementation of management and control is effectively shortened, and the stability and continuity of production operation are improved.

Inventors

  • LI QIANSHU
  • YIN FENG

Assignees

  • 成都云中行科技有限公司

Dates

Publication Date
20260505
Application Date
20260206

Claims (10)

  1. 1. Real-time data decision support system of production line based on cloud limit is cooperated, characterized by comprising: the intelligent edge nodes are distributed at manufacturing resources on the production line and are used for collecting multi-source heterogeneous real-time data and executing local decisions; the cloud decision platform is used for executing global data analysis and strategy generation; The self-adaptive decision coordinator is in communication connection with the plurality of edge intelligent nodes and the cloud decision platform; The cloud decision platform is also used for generating a decision optimization strategy or an updating model based on the information acquired from the edge intelligent nodes and the self-adaptive decision coordinator, and sending the decision optimization strategy or the updating model to the relevant edge intelligent nodes through the self-adaptive decision coordinator.
  2. 2. The cloud edge collaboration-based production line real-time data decision support system as claimed in claim 1, wherein the edge intelligent nodes comprise a data acquisition module, a local digital twin micro model and a micro decision module; The data acquisition module is used for continuously acquiring vibration, acoustics, images and process parameter data of the deployed manufacturing resources; The local digital twin micro model is used for dynamically simulating and predicting key performance index flows of corresponding manufacturing resources when the current production task is completed according to the real-time data flows acquired by the data acquisition module; The micro decision module is used for comparing the predicted stream of the local digital twin micro model with the actual data stream of the data acquisition module, generating and executing a local control instruction when the deviation exceeds a dynamically adjusted threshold value, and generating a decision experience data packet containing an abnormal situation segment, an execution action and instant feedback.
  3. 3. The cloud edge collaboration-based production line real-time data decision support system as recited in claim 2, wherein the edge intelligent node further comprises a local model updating module; The local model updating module is used for receiving model updating data or parameters sent by the self-adaptive decision coordinator, carrying out on-line adjustment on the local digital twin micro-model, and/or carrying out self-adaptive optimization on internal parameters of the local digital twin micro-model based on feedback information in a decision experience data packet generated by the micro-decision module.
  4. 4. The cloud edge collaboration-based production line real-time data decision support system as claimed in claim 3, wherein the cloud decision platform comprises a hybrid game strategy synthesis engine; the hybrid game policy synthesis engine is used for abstracting a production line into a multi-agent collaborative network, wherein each edge intelligent node or manufacturing resources represented by the edge intelligent node are regarded as an agent; The hybrid game strategy synthesis engine is configured to receive state information and decision experience data packets from each agent, dynamically construct a strategy gain matrix for coping with resource conflict or complex abnormal events, simulate gain changes of each agent and a global system under different collaborative strategies, and synthesize collaborative decision strategies for improving the overall system by solving the collaborative game model.
  5. 5. The cloud edge collaboration-based production line real-time data decision support system as claimed in claim 4, wherein the cloud decision platform further comprises a global knowledge base and model evolution module; The global knowledge base is used for storing decision experience data packets gathered from all edge intelligent nodes and collaborative decision strategies generated by the hybrid game strategy synthesis engine, and constructing an industrial decision knowledge graph; The model evolution module is used for carrying out iterative optimization on game model parameters of the hybrid game strategy synthesis engine based on the data accumulated in the global knowledge base and/or training a new decision-making auxiliary model, and the new model obtained through training or the optimized parameters are packaged into model updating data.
  6. 6. The cloud edge collaboration-based production line real-time data decision support system as claimed in claim 5, wherein the self-adaptive decision coordinator comprises a system situation awareness module and a dynamic routing scheduling module; The system situation awareness module is used for fusing microcosmic abnormal signals, production plan information and resource load states from all edge intelligent nodes in real time to generate and maintain a real-time situation map of the whole system; The dynamic routing scheduling module is embedded with a decision routing rule set and is used for dynamically judging the type and the emergency degree of a current decision task according to the real-time situation map provided by the system situation awareness module, and distributing the decision task to at least one path for execution according to the type and the emergency degree, wherein the path is used for triggering local micro-decisions of a single edge intelligent node, initiating direct transverse collaboration among a plurality of edge intelligent nodes or submitting the direct transverse collaboration to the cloud decision platform for global arbitration and strategy synthesis.
  7. 7. The cloud edge collaboration-based production line real-time data decision support system of claim 6, wherein the dynamic routing scheduling module is further configured to parse decision optimization policies or model update data from the cloud decision platform, determine a target set of edge intelligent nodes according to the real-time situation map, and send the decision optimization policies or model update data to each of the target set of edge intelligent nodes in a streaming manner.
  8. 8. The cloud edge collaboration-based production line real-time data decision support system as described in claim 7, further comprising an inter-vehicle layer data bus and a unified interface gateway; the workshop layer data bus provides a data exchange channel for the plurality of edge intelligent nodes, the self-adaptive decision coordinator and the cloud decision platform; The unified interface gateway is connected between the workshop layer data bus and an external network and used for protocol conversion, data encapsulation and safety isolation, and reliable communication between the cloud decision platform and a system in a workshop is achieved.
  9. 9. The cloud edge collaboration-based production line real-time data decision support system as claimed in claim 2, wherein the local digital twin micro model is a lightweight machine learning model trained based on the historical operation data of the manufacturing resources and the knowledge fusion of physical mechanisms, and the simulated key performance index flow comprises at least one of a dimensional tolerance change sequence, a surface quality evolution sequence or an energy consumption fluctuation sequence.
  10. 10. The cloud edge collaboration-based real-time data decision support system of the production line is characterized in that in the hybrid game strategy synthesis engine, a plurality of factors of production efficiency index, product quality index, equipment health index and energy consumption index of manufacturing resources represented by each agent are comprehensively considered for a benefit function defined by each agent, and the strategy benefit matrix construction and solving process is completed within a preset time window constraint.

Description

Real-time data decision support system of production line based on cloud edge cooperation Technical Field The invention relates to the technical field of industrial Internet of things and intelligent manufacturing, in particular to a cloud edge cooperation-based real-time data decision support system of a production line. Background Along with the rapid development of new generation information technologies such as industrial Internet of things, cloud computing, edge computing and the like, manufacturing enterprises are gradually transformed into intelligent and digital directions. In the production process, the acquisition, analysis and decision support of real-time data of a production line become key links for improving the production efficiency and guaranteeing the production stability. Especially in a large-scale and multi-process complex production environment, how to realize real-time sensing, abnormal early warning and intelligent decision on the state of a production line becomes a core problem to be solved in the current intelligent manufacturing field. Currently, some research has been dedicated to the monitoring and optimization of the production process in a data-driven manner. For example, the patent of the invention with the publication number of CN112149866A discloses an intelligent manufacturing workshop anomaly prediction and control method based on edge cloud cooperation, which provides a cooperative decision framework integrating edge calculation and cloud calculation, wherein the anomaly prediction is carried out by deploying a convolutional neural network at a manufacturing resource end, and an edge cloud cooperation mechanism is adopted for carrying out anomaly processing. The method realizes the transition from 'post processing' to 'pre prediction' to a certain extent, and improves the timeliness of abnormal response. However, the prior art still has the defects of limited data acquisition and integration dimension, weak real-time decision support capability, solidification of a collaborative decision mechanism and insufficient system expansibility and adaptability. Therefore, a production line data decision support system which can deeply fuse cloud edge collaborative architecture, support multi-source real-time data high-efficiency integration and has strong real-time analysis and intelligent decision capability is still lacking in the prior art. In particular, how to achieve seamless collaboration of data acquisition, processing and decision making in the face of highly concurrent, highly dynamic production environments remains a significant bottleneck in the current state of the art. Based on the method, the invention provides a cloud-edge cooperation-based real-time data decision support system of a production line, which aims to construct an integrated platform which supports multi-source data real-time access, has the capabilities of intelligent edge preprocessing and cloud deep analysis and can realize dynamic optimization and cooperation decision so as to make up the defects of the existing system in terms of instantaneity, self-adaption and decision support capability. Disclosure of Invention The invention aims to provide a Yun Bian-collaboration-based real-time data decision support system for a production line, so as to solve the problems in the background technology. In order to solve the technical problems, the invention provides the following technical scheme that the cloud edge cooperation-based real-time data decision support system of the production line comprises: the intelligent edge nodes are distributed at manufacturing resources on the production line and are used for collecting multi-source heterogeneous real-time data and executing local decisions; the cloud decision platform is used for executing global data analysis and strategy generation; The self-adaptive decision coordinator is in communication connection with the plurality of edge intelligent nodes and the cloud decision platform; The cloud decision platform is also used for generating a decision optimization strategy or an updating model based on the information acquired from the edge intelligent nodes and the self-adaptive decision coordinator, and sending the decision optimization strategy or the updating model to the relevant edge intelligent nodes through the self-adaptive decision coordinator. Preferably, the edge intelligent node comprises a data acquisition module, a local digital twin micro model and a micro decision module; The data acquisition module is used for continuously acquiring vibration, acoustics, images and process parameter data of the deployed manufacturing resources; The local digital twin micro model is used for dynamically simulating and predicting key performance index flows of corresponding manufacturing resources when the current production task is completed according to the real-time data flows acquired by the data acquisition module; The micro decision module is used for