A digital twin method and system for edge computing

Abstract

The present invention discloses a digital twin method and system for edge computing, wherein the method includes: using an external network to network a MEC physical system with a digital twin of the MEC system; the MEC physical system is communicated through an external network based on information, and the digital twin of the MEC system is communicated through an external network to monitor the MEC physical system to obtain monitoring results, and the monitoring results are analyzed and evaluated to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and the constructed MEC physical system is digitally twinned, and the closed-loop optimization control of the MEC physical system is completed based on the evaluation results; the unconstructed MEC physical system is digitally twinned, and the optimization design of the MEC physical system is completed based on the evaluation results. The present invention can continuously optimize the MEC physical system through the digital twin of the MEC physical system, improve the management efficiency and resource utilization efficiency of the system, and reduce the operating cost and investment expenditure cost of the MEC service.

Description

Technical Field

[0001] The present invention relates to the field of computer simulation, and in particular to a digital twin method and system for edge computing. Background Art

[0002] In recent years, IT technology and mobile communication technology have made great progress. Various smart devices have emerged, enabling smart terminals to support some computing-intensive and data-intensive services based on cloud computing, bringing many conveniences to people's lives. However, most of these emerging applications are computing-intensive and data-intensive applications, which not only generate a large amount of data, but also, as users and applications continue to increase their requirements for service quality experience, the computing-intensive and data-intensive service provision paradigm based on cloud computing will not only bring a large service response delay, but also bring a huge load burden to the backhaul link, and also cause the battery life of mobile device terminals to be difficult to meet business needs.

[0003] To address the above challenges, ETSI introduced a new computing paradigm called Multi-Access Edge Computing (MEC). MEC deploys computing and storage resources to the edge of the network, providing real-time data analysis and intelligent processing at the edge of the network. Terminals can support computing-intensive and data-intensive services while meeting the strict latency requirements of the services, and can also reduce the load burden of mobile terminal devices on the backhaul link.

[0004] When deploying an actual system, a simulation method is usually used. First, a simulation / emulation system for the actual system is built, and then the simulation / emulation system is optimized. The actual physical system is deployed based on the optimization result information of the simulation and / or emulation. After the physical system is deployed, the closed-loop optimization management process for the physical system is completed by collecting data in the physical system and monitoring and evaluating the performance of the system. Due to the limitations of simulation / emulation software in simulation scenarios, modeling, computational complexity, and resource usage, simulation and simulation evaluation are usually performed separately for multiple aspects of the system. Therefore, the simulation / emulation of the system and the resulting closed-loop control and optimization constitute a digital twin for the system.

[0005] Similarly, in order to effectively solve the orchestration optimization and management problems of MEC physical systems, it is necessary to build a digital twin of the MEC physical system based on the concept of digital twins, collect the real-time operation data of the MEC physical system through real-time information interaction between the simulation system instance of the MEC physical system and the MEC physical system, perceive the actual operation status of the MEC physical system, and complete data synchronization, performance analysis and evaluation of the MEC physical system and its corresponding digital twin based on the actual operation status of the MEC physical system, thereby completing the closed-loop continuous optimization control process for the MEC physical system.

[0006] Through systematic research, it is found that there is currently no research result on building digital twins for MEC physical systems and realizing optimized control and orchestration of MEC physical systems. Summary of the invention

[0007] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

[0008] To this end, the purpose of the present invention is to build a digital twin for the MEC physical system, to achieve optimized control and orchestration of the MEC physical system, and to propose a digital twin method for edge computing.

[0009] Another object of the present invention is to propose a digital twin system for edge computing.

[0010] To achieve the above objectives, the present invention proposes a digital twin method for edge computing, comprising:

[0011] Step S1, using an external network to network the MEC physical system with the digital twin of the MEC system, so that information can be exchanged between the MEC physical system and the digital twin of the MEC system; step S2, the MEC physical system is passed through the external network and based on the information interaction, so that the MEC physical system is mirrored in the digital twin of the MEC system; and, step S3, the digital twin of the MEC system is passed through the external network, based on simulation and / or emulation, through data acquisition and storage, to monitor the MEC physical system, obtain monitoring results, and analyze and evaluate the monitoring results to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and, digital twinning of the constructed MEC physical system is performed, and closed-loop optimization control of the MEC physical system is completed based on the evaluation results; digital twinning of the unconstructed MEC physical system is performed, and optimization design of the MEC physical system is completed based on the evaluation results.

[0012] The edge computing digital twin method of the embodiment of the present invention can optimize the MEC physical system, improve the network management efficiency and resource utilization efficiency of the system, and reduce the operating cost and investment expenditure cost of MEC services. The present invention is based on the multi-perspective, multi-instance function and performance simulation / emulation of the MEC physical system, constructs the digital twin of the MEC system, and completes the state perception, monitoring and closed-loop continuous optimization control process for the MEC physical system based on the MEC digital twin.

[0013] To achieve the above object, the present invention proposes, on the other hand, an edge computing digital twin system, comprising:

[0014] An external network entity, used to use an external network to network the MEC physical system with the digital twin of the MEC system, so that information can be exchanged between the MEC physical system and the digital twin of the MEC system;

[0015] The MEC physical system is used to transmit the MEC physical system through the external network and based on the interaction of the information, so that the MEC physical system is mirrored in the digital twin of the MEC system; and

[0016] The digital twin of the MEC system is used to monitor the MEC physical system through the external network, based on simulation and / or emulation, through data acquisition and storage, to obtain monitoring results, and analyze and evaluate the monitoring results to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and to perform digital twinning on the constructed MEC physical system, and complete the closed-loop optimization control of the MEC physical system based on the evaluation results; to perform digital twinning on the unconstructed MEC physical system, and complete the optimization design of the MEC physical system based on the evaluation results.

[0017] The edge computing digital twin system of the embodiment of the present invention can optimize the MEC physical system, improve the network management efficiency and resource utilization efficiency of the system, and reduce the operating cost and investment expenditure cost of MEC services. The present invention constructs a digital twin of the MEC system based on multi-perspective, multi-instance function and performance simulation / emulation of the MEC physical system, and completes the state perception, monitoring and closed-loop continuous optimization control process of the MEC physical system based on the MEC digital twin.

[0018] Beneficial effects of the present invention:

[0019] (1) The present invention can be applied to optimize the MEC physical system that has been built. A digital twin instance is constructed for the MEC physical system that has been built to complete real-time monitoring and closed-loop continuous optimization control of the MEC physical system;

[0020] (2) The present invention can be applied to the optimized deployment of MEC physical systems. A digital twin is generated for the MEC physical system to be constructed, and its digital twin instance is constructed to support the optimized design of the MEC physical system. Based on the optimized design results of the MEC digital twin instance, the actual MEC physical system is constructed.

[0021] (3) Based on the MEC digital twin method of the present invention, it is possible to support monitoring, control and optimization of the MEC system, and improve the network management efficiency of the MEC system in terms of fault management, performance management, security management, log management and configuration management; by improving the resource utilization efficiency and management efficiency of the MEC system, the operating cost (OpEx) and investment expenditure cost (CapEx) when providing MEC services can be effectively reduced.

[0022] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0024] Figure 1 A flowchart of a digital twin method for edge computing according to an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of the functional architecture of a digital twin for a MEC system according to an embodiment of the present invention;

[0026] Figure 3 A schematic diagram of the framework of the main functional modules included in the MEC physical system according to an embodiment of the present invention;

[0027] Figure 4 It is a schematic diagram of the framework of the functional modules included in the digital twin of the MEC system according to an embodiment of the present invention;

[0028] Figure 5 A schematic diagram of the framework of the main functional submodules included in the MEC model management control module according to an embodiment of the present invention;

[0029] Figure 6 A schematic diagram of the framework of the main functional submodules included in the MEC digital twin operation management module according to an embodiment of the present invention;

[0030] Figure 7 Schematic diagram of the structure of a digital twin system for edge computing according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] The following first describes a digital twin method and system for edge computing. The functional architecture of the method for digital twinning of MEC physical systems is as follows: Figure 2 As shown, the functional architecture includes the MEC physical system, the digital twin of the MEC system, and external network entities.

[0034] Figure 1 It is a flow chart of a digital twin method for edge computing according to an embodiment of the present invention.

[0035] like Figure 1 As shown, the method includes but is not limited to:

[0036] Step S1, using an external network to network the MEC physical system with the digital twin of the MEC system, so that information can be exchanged between the MEC physical system and the digital twin of the MEC system.

[0037] Specifically, in the present invention, the network entity of the external network mainly provides networking connectivity between the MEC physical system and the digital twin of the MEC system, and completes information interaction between the MEC physical system and the digital twin of the MEC system. The network entity includes but is not limited to mobile wireless networks, wired networks and other external networks, such as the Internet.

[0038] Step S2: The MEC physical system is mirrored in the digital twin of the MEC system through an external network and based on information interaction.

[0039] Specifically, the MEC physical system is an actual MEC system that supports MEC functions. The MEC physical system exchanges information with the digital twin of the MEC system through external network entities to complete the mirroring of its functions and performance on the digital twin of the MEC system.

[0040] Furthermore, the MEC physical system is deployed in the NFV environment, including two components: MEC system level and MEC host level. Figure 3 Schematic diagram of the framework of the main functional modules of the MEC physical system according to an embodiment of the present invention, including:

[0041] (1) MEC system level: mainly includes MEC application orchestrator, NFV orchestrator, operation support system, CFS portal, device APP, user APP and user APP lifecycle management agent. The main functions of each functional module are as follows:

[0042] A. MEC Application Orchestrator (MEAO): From a global perspective, completes the orchestration and maintenance of the MEC system in terms of deployed MEC hosts, available resources, available MEC services, and topology.

[0043] B. NFV Orchestrator (NFVO): Responsible for orchestrating virtualized resources; responsible for managing the life cycle of network services including MEC APP instances, and can orchestrate MEC-based user APPs as VNFs.

[0044] C. Operation Support System (OSS): Responsible for receiving requests for instantiating or terminating APPs through the CFS portal and device APPs, and deciding whether to approve these requests. Authorized requests are forwarded to the NFV orchestrator so that the NFV orchestrator can further process the request based on the current status and resources of the MEC physical system.

[0045] D.CFS portal: It is the portal for operators to subscribe to and monitor edge apps for third-party customers.

[0046] E. Device APP: refers to the APP in the device (such as user equipment, laptop with Internet connection).

[0047] F. User APP: refers to the MEC APP instantiated in the MEC system, which responds to user requests through the device APP.

[0048] G. User APP lifecycle management agent: allows device APP to request to start, instantiate, terminate user APP, and redirect user APP if supported by the MEC system; allows device APP to be notified of the status of user APP.

[0049] (2) MEC host level: It includes two parts: MEC host and MEC host level management. Specifically:

[0050] A. MEC host: includes the MEC platform and virtualized infrastructure, providing computing, storage, and network resources for MEC APP, including:

[0051] MEC platform: provides an environment where MEC APP can discover, notify, consume and provide MEC services; is responsible for receiving traffic rules from the MEC platform manager, applications or services, and instructs the data plane accordingly; is responsible for receiving DNS records from the MEC platform manager and configuring the corresponding DNS proxy / server; is responsible for hosting MEC services; is responsible for providing access to persistent storage and time information. The MEC platform can be deployed as a VNF in an NFV environment.

[0052] Virtualized infrastructure: Responsible for providing virtualized computing, storage, and network resources to MEC APP. Virtualized infrastructure can be deployed as NFVI in the NFV environment and managed by VIM.

[0053] MEC APP: As a virtualized application (such as a virtual machine or containerized application) running on the virtualized infrastructure provided by the MEC host, it can interact with the MEC platform to use and provide MEC services.

[0054] B. MEC host-level management: MEC host-level management includes MEC platform manager, MEC platform lifecycle management, MEC APP lifecycle management, and virtualized infrastructure manager, which are used to accept the scheduling of MEC application orchestrator and NFV orchestrator.

[0055] MEC Platform Manager (MEPM-V): responsible for providing element management functions to the MEC platform; responsible for managing application rules and requirements, including traffic rules, service authorization, DNS configuration and conflict resolution.

[0056] MEC platform lifecycle management: Complete the lifecycle management function of the MEC platform.

[0057] MEC APP lifecycle management: Each MEC APP is regarded as a VNF instance and is responsible for the lifecycle management of the MEC APP.

[0058] Virtualization Infrastructure Manager (VIM): Responsible for allocating, managing, and releasing virtualized computing, storage, and network resources of the virtualized infrastructure; responsible for preparing the virtualized infrastructure to run software images, preparing to configure the infrastructure, which may also include receiving and storing software images; responsible for collecting and reporting performance and fault information of virtualized resources.

[0059] It should be understood that the above-mentioned MEC physical system can be deployed in a centralized manner or in a distributed manner.

[0060] Step S3, the digital twin of the MEC system is used through an external network, based on simulation and / or emulation, through data collection and storage, to monitor the MEC physical system, obtain monitoring results, and analyze and evaluate the monitoring results to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and, the constructed MEC physical system is digitally twinned, and the closed-loop optimization control of the MEC physical system is completed based on the evaluation results; the unconstructed MEC physical system is digitally twinned, and the optimization design of the MEC physical system is completed based on the evaluation results.

[0061] Specifically, the digital twin of the MEC system completes the mirroring of the MEC physical system through an external network entity. Based on simulation and / or emulation methods, the system performs state perception, function and performance monitoring of the MEC physical system through data collection and storage, and completes the functions and performance of the digital twin corresponding to the MEC physical system based on different perspectives; analyzes and evaluates the monitoring results of the MEC physical system; for MEC physical systems that have not been built, completes the optimization design of the MEC physical system based on the analysis and evaluation results; for established MEC physical systems, performs closed-loop continuous optimization control based on the analysis and evaluation results.

[0062] Further, Figure 4 This is a schematic diagram of the framework of the functional modules included in the digital twin of the MEC system of an embodiment of the present invention, including: MEC digital twin service module, MEC digital twin providing module, MEC model providing module, MEC data management module and communication module. The main functions completed by each functional module are as follows:

[0063] (1) MEC digital twin service module: including MEC digital twin service description module, MEC digital twin service development module, MEC digital twin service deployment and operation module, MEC digital twin service management module, MEC digital twin service evaluation module and MEC digital twin service transaction module. The functions of each module are introduced below:

[0064] A. MEC digital twin service description module: It is a digital twin service configuration technical document that describes the MEC digital twin service, including but not limited to providing the modeling environment, modeling rules and modeling logic flow corresponding to the digital twin service, as well as key service quality indicators, system performance indicators and system key parameter requirements during simulation / emulation;

[0065] B. MEC digital twin service development module: used to support the development of MEC digital twin services;

[0066] C. MEC digital twin service deployment and operation module: used to support the deployment and operation of MEC digital twin services;

[0067] D. MEC digital twin service management module: used for the full life cycle management of MEC digital twin services;

[0068] E. MEC digital twin service evaluation module: used to evaluate MEC digital twin services based on multiple perspectives;

[0069] F.MEC digital twin service transaction module: used to support functions related to MEC digital twin service transactions.

[0070] (2) MEC digital twin provision module: including several MEC digital twin instances, MEC digital twin orchestration modules and MEC digital twin data storage modules. Each MEC digital twin instance consists of a MEC digital twin module, a MEC digital twin operation management module and a MEC digital twin visualization module. The main functions of each module are as follows:

[0071] A. MEC digital twin instance: refers to a mirror instance of the MEC physical system formed by associating, combining, integrating and fusing one or more MEC models based on different perspectives;

[0072] B. MEC digital twin orchestration module: used to receive and parse service requests from different digital twin instances, and perform related operations such as construction, deletion, orchestration, and management on one or more MEC digital twin instances;

[0073] C. MEC digital twin data storage module: used to store information about the MEC digital twin, including but not limited to one or more MEC models contained in the digital twin and their semantic associations; modeling scenario information, platform, code of the digital twin, and system key parameter configuration information of the contained MEC model;

[0074] D. MEC digital twin module: It is a characterization of the MEC physical system formed by associating, combining, integrating and fusing multiple MEC models based on different perspectives. It is the core component of a MEC digital twin instance.

[0075] E. MEC digital twin operation management module: used to support the management of the MEC digital twin in the digital twin instance during operation, including but not limited to the initialization, operation monitoring, status update and optimization of different MEC models in the MEC digital twin; based on specific digital twin service rules and policies, generate and schedule tasks during the operation management of the MEC digital twin;

[0076] F. MEC digital twin visualization module: For the MEC digital twin instance, the operating status information, data, and control information of the MEC digital twin are visualized to facilitate interaction with users.

[0077] (3) MEC model providing module: including MEC model instance, MEC model orchestration module and MEC model data storage module. Among them, MEC model instance includes MEC model, MEC model management and control module and MEC model visualization module. The main functions of each module are introduced below:

[0078] A. MEC model instance: refers to the use case of the MEC model that characterizes the MEC physical system from different perspectives;

[0079] B. MEC model orchestration module: provides orchestration and optimization management functions for a single MEC model and / or multiple MEC models in the MEC digital twin; including but not limited to the construction, deletion and optimization management operations for MEC models, as well as associating, combining, integrating and fusing multiple MEC models into a new MEC model;

[0080] C. MEC model data storage module: used to store data information of MEC models, including but not limited to modeling scenario information, platform, code and system key parameter configuration information of MEC models;

[0081] D. MEC model: refers to a MEC model that describes the functions and / or performance of a MEC physical system from a specific perspective. In fact, modeling and simulation / emulation from different perspectives of the MEC system can form different MEC models, such as a MEC delay model formed by simulating and evaluating its delay performance; a MEC computing energy consumption model formed by simulating and evaluating its computing and energy consumption performance; a MEC system virtualization resource management model formed by simulating and evaluating its virtualization resource management, etc.

[0082] E. MEC model management and control module: implements closed-loop optimization control of the MEC model and / or the corresponding MEC physical system by updating, synchronizing and optimizing the MEC model in the MEC model instance in real time;

[0083] F. MEC model visualization module: used to visualize the operating status, data, and control information of the MEC model contained in the MEC model instance to facilitate interaction with users.

[0084] (4) MEC data management module: This module includes MEC data preprocessing submodule, MEC data analysis submodule, MEC data fusion submodule and MEC data storage submodule. The main functions of each module are as follows:

[0085] A. MEC data preprocessing submodule: used to preprocess MEC physical system data and / or MEC model data, including but not limited to cleaning, transforming, aggregating, splicing, etc.

[0086] B. MEC data analysis submodule: used for feature extraction, representation and information mining of MEC physical system data and / or MEC model data;

[0087] C. MEC data fusion submodule: It integrates the data of MEC physical system and / or MEC model according to business requirements and / or the needs of big data analysis;

[0088] D. MEC data storage submodule: used to store the operating data of MEC physical systems, MEC model instances, and MEC digital twin instances.

[0089] (5) Communication module: used to support the communication between the digital twin of the MEC system and its corresponding MEC physical system.

[0090] Furthermore, the functional submodules included in the MEC model management control module according to the embodiment of the present invention are as follows: Figure 5 As shown, it includes: MEC model real-time update submodule, MEC model monitoring submodule, MEC model analysis submodule, MEC model evaluation submodule, MEC model report generation submodule, MEC model policy management submodule, MEC model optimization design submodule and MEC physical system control submodule. The main functions of each submodule are as follows:

[0091] A. MEC model real-time update submodule: used to update the operating status of the MEC model instance in real time based on the real-time operating data of the MEC physical system and / or the MEC model.

[0092] B. MEC model monitoring submodule: monitors the operating status and performance of MEC model instances based on real-time collected operating data.

[0093] C. MEC model analysis submodule: Based on data-driven analysis methods, it analyzes and predicts the operating status and performance of MEC model instances.

[0094] D. MEC model evaluation submodule: Based on the operating status and performance data provided by the MEC model monitoring submodule and the analysis result information output by the MEC model analysis submodule, it evaluates the performance of the MEC model instance and provides the performance evaluation results.

[0095] E.MEC model report generation submodule: Generates an evaluation report on the running status and performance of the MEC model instance based on the performance evaluation results of the MEC model evaluation submodule.

[0096] F. MEC model optimization design submodule: used to optimize the design of the MEC model instance based on the evaluation result information and operation status information of the MEC model instance, and output the optimization design plan for the MEC model instance.

[0097] G. MEC model policy management submodule: Receives information from the MEC digital twin policy management submodule, generates policy management rules for MEC model instances, and performs policy management for the evaluation and optimization design of MEC model instances based on the policy management rules.

[0098] H.MEC physical system control submodule (MEC model level): Based on the communication module, it interacts with the MEC physical system and completes the closed-loop optimization control of the MEC physical system based on the MEC model level.

[0099] Furthermore, the main functional submodules included in the MEC digital twin operation management module according to the embodiment of the present invention are as follows: Figure 6 As shown, it includes: MEC digital twin real-time update submodule, MEC digital twin monitoring submodule, MEC digital twin analysis submodule, MEC digital twin evaluation submodule, MEC digital twin report generation submodule, MEC digital twin optimization design submodule, MEC digital twin task scheduling submodule, MEC digital twin strategy management submodule and MEC physical system control submodule. The specific functions of each submodule are as follows:

[0100] A. MEC digital twin real-time update submodule: used to update the operating status of the MEC digital twin instance in real time based on the real-time operating data of the MEC physical system and / or the MEC digital twin.

[0101] B. MEC digital twin monitoring submodule: monitors the operating status and performance of the MEC digital twin instance based on real-time collected operating data.

[0102] C. MEC digital twin analysis submodule: Analyzes and predicts the operating status and performance of MEC digital twin instances based on data-driven analysis methods.

[0103] D. MEC digital twin evaluation submodule: Based on the operating status and performance data information provided by the MEC digital twin monitoring submodule and the analysis result information output by the MEC digital twin analysis submodule, it evaluates the performance of the MEC digital twin instance and gives the evaluation results.

[0104] E.MEC digital twin report generation submodule: Generates an evaluation report on the operating status and performance of the MEC digital twin instance based on the evaluation results output by the MEC digital twin evaluation submodule.

[0105] F. MEC digital twin optimization design submodule: Based on the evaluation results and operating status of the MEC digital twin, optimize the design of the MEC digital twin instance, generate and output the optimization design plan.

[0106] G. MEC digital twin task scheduling submodule: This submodule schedules and manages the tasks during the operation of the MEC digital twin based on the policy rules given by the MEC digital twin policy management submodule. The operating tasks include but are not limited to the performance analysis, performance evaluation tasks, data synchronization tasks, operation data collection tasks and operation data storage tasks generated during the operation of the MEC digital twin and one or more MEC models contained therein, or one or more MEC models formed by association, combination, integration and fusion.

[0107] H.MEC digital twin policy management submodule: This submodule receives information from the MEC digital twin service description module, generates policy management rules for the MEC digital twin instance, and based on the policy management rules, performs policy management for the task scheduling, monitoring, evaluation, and optimization design of the MEC digital twin instance.

[0108] I. MEC physical system control submodule: includes control functions at the MEC digital twin level and the MEC model level: (I.1) MEC digital twin level: interacts with the MEC physical system through the communication module, and completes closed-loop optimization control of the MEC physical system at the MEC digital twin level; (I.2) MEC model level: based on the communication module, interacts with the MEC physical system, and completes closed-loop optimization control of the MEC physical system at the MEC model level.

[0109] The closed-loop optimization control process for the MEC model level and the MEC digital twin level is as follows:

[0110] (1) Closed-loop optimization control process for MEC model level (MEC-LEVEL-OCP)

[0111] Step 1: The MEC model orchestration module notifies the MEC model management and control module of a certain MEC model instance through the communication module, and starts a closed-loop optimization control process based on the MEC model level for the MEC model instance.

[0112] The MEC model management control module starts the closed-loop optimization control process for the MEC model level. The MEC model real-time update submodule periodically obtains real-time operation data synchronized with the MEC physical system from the MEC data storage submodule in the MEC data management module, and updates and synchronizes the operation status of the MEC model instance in real time.

[0113] Step 2: The MEC model monitoring submodule periodically collects the operation data of the MEC model instance and selectively inputs the data into the MEC model analysis submodule; the MEC model analysis submodule analyzes and predicts the operation status of the MEC model based on the operation data, and outputs the analysis and prediction results to the MEC model evaluation submodule;

[0114] Step 3: The MEC model evaluation submodule evaluates the operating status and performance of the MEC model instance based on the operating data and the analysis and prediction results output by the MEC model analysis submodule, and inputs the output evaluation result information into the MEC model report generation submodule and the MEC model visualization module.

[0115] The MEC model report generation submodule generates the corresponding MEC model evaluation report based on the evaluation result information, and the MEC model visualization module performs a visual display of the evaluation results.

[0116] If the output evaluation result information indicates that the MEC model is running normally, execute step 8;

[0117] If the output evaluation result information is that the MEC model operation status is abnormal and / or the MEC model performance is degraded, the MEC model evaluation submodule sends a report to the MEC model analysis submodule;

[0118] Step 4: The MEC model analysis submodule obtains the historical operation data of the MEC physical system from the MEC data management module based on the evaluation results of the abnormal operation and / or performance degradation of the MEC model, and analyzes and causally infers the causes of the abnormal operation and performance degradation of the MEC model instance based on the data-driven analysis method, and outputs the analysis results to the MEC model visualization module, the MEC model report generation submodule and the MEC model optimization design submodule respectively. The MEC model report generation module generates a MEC model operation status analysis report; the MEC model visualization module displays the corresponding operation status analysis information;

[0119] Step 5: The MEC model optimization design submodule collects the historical operation data of the MEC physical system in the MEC data management module, sets the optimization goals based on the analysis results and the policy management information obtained by interacting with the MEC model policy management submodule, performs orchestration optimization design for the MEC model, and outputs the orchestration optimization plan to the MEC model.

[0120] Step 6: Based on the orchestration optimization results, the MEC model updates the relevant MEC model configuration and completes the simulation / emulation process of the MEC model based on the orchestration optimized MEC model.

[0121] Step 7: The MEC model monitoring submodule periodically collects the operation data of the orchestration-optimized MEC model, stores the operation data in the MEC data storage submodule in the MEC data management module, and selectively inputs the data into the MEC model analysis submodule; the MEC model analysis submodule analyzes and predicts the operation status of the MEC model based on the operation data, and outputs the analysis and prediction results to the MEC model evaluation submodule; the MEC model evaluation submodule evaluates the operation performance of the orchestration-optimized MEC model;

[0122] If the running performance of the MEC model reaches the optimization target, the MEC model evaluation submodule outputs the orchestration optimization success result information to the MEC model optimization design submodule, the MEC model visualization module and the MEC model report generation submodule; the MEC model report generation submodule generates a MEC model orchestration optimization result report, and the MEC model visualization submodule displays the corresponding MEC model orchestration optimization result report information; execute step 8;

[0123] If the running performance of the MEC model does not reach the optimization target, the MEC model evaluation submodule outputs the orchestration optimization failure result information to the MEC model optimization design submodule, the MEC model visualization module and the MEC model report generation submodule; the MEC model report generation submodule outputs the running status result information after the orchestration, and the MEC model visualization submodule visualizes the running status result information after the orchestration; return to step 5;

[0124] Step 8: The MEC model optimization design submodule outputs the optimization design result information to the MEC model orchestration module and the MEC model data storage module through the communication module. The MEC model data storage module stores the model key parameter configuration information related to the MEC model optimization design result.

[0125] (2) Closed-loop optimization control process at the MEC digital twin level (DT-LEVEL-OCP)

[0126] At the MEC digital twin level, the MEC digital twin orchestration module builds one or more MEC models for a certain digital twin instance based on the association, combination, integration and fusion of multiple MEC models to complete the closed-loop optimization control process of the digital twin instance at the MEC digital twin level.

[0127] If the MEC digital twin instance contains only one MEC model, the corresponding closed-loop optimization control process can be completed based on the MEC-LEVEL-OCP process through the interaction between the MEC digital twin operation management module, the MEC model orchestration module and the MEC model instance.

[0128] If the MEC digital twin instance contains multiple MEC models, the MEC digital twin orchestration module and the digital twin operation management module of the digital twin instance complete the corresponding MEC digital twin level closed-loop optimization control process through the following steps:

[0129] Step 1: The MEC digital twin orchestration module interacts with the MEC digital twin service description module to obtain the policy management rule information for the MEC digital twin instance, and sends the policy management rule information to the digital twin policy management submodule of the digital twin instance.

[0130] Step 2: The MEC digital twin orchestration module notifies the MEC digital twin optimization design submodule to start the closed-loop optimization control process for multiple MEC models in the digital twin instance;

[0131] The MEC digital twin optimization design submodule interacts with the digital twin policy management submodule to obtain policy management rule information for multiple MEC models in this digital twin instance.

[0132] Step 3: Based on the key performance indicators of multiple MEC models and the policy management rules and information for multiple MEC models, the MEC digital twin optimization design submodule transforms the closed-loop optimization control problem into a joint optimization problem with multiple MEC models as optimization sub-objectives, and gives the corresponding constraints and related sub-objective weight parameter setting information.

[0133] Step 4: The MEC digital twin optimization design submodule solves the joint optimization problem.

[0134] If an optimized solution is obtained, the MEC digital twin optimization design submodule outputs the corresponding optimization result information to the MEC digital twin orchestration module, as well as the MEC digital twin data storage module, MEC digital twin report generation submodule and MEC digital twin visualization module of the digital twin instance, wherein the MEC digital twin data storage module is used to store the optimization result information and the system key parameter configuration information corresponding to each MEC model; the MEC digital twin report generation submodule generates an optimization result information report; the MEC digital twin visualization module visualizes the solved optimization result information for interaction with people; execute step 5;

[0135] If no optimized solution is obtained, the MEC digital twin optimization design submodule will output the corresponding optimization failure result information to the MEC digital twin orchestration module, as well as the MEC digital twin data storage module, MEC digital twin report generation submodule and MEC digital twin visualization module of the digital twin instance, wherein the MEC digital twin data storage module stores the relevant information of this optimization failure and the system key parameter configuration information of each MEC model, the MEC digital twin report generation submodule generates an optimization failure information report, and the MEC digital twin visualization module visualizes the optimization failure information for interaction with people.

[0136] The MEC digital twin optimization design submodule adjusts the constraints and key parameters corresponding to the optimization problem manually and / or automatically, and executes step 3;

[0137] Step 5: The closed-loop optimization control process at the MEC digital twin level ends.

[0138] It should be understood that, when performing digital twinning for a MEC physical system, since the MEC physical system can be a constructed MEC physical system or an unconstructed MEC physical system, based on the main functional modules included in the digital twinning of the MEC physical system, when performing digital twinning for a MEC physical system, digital twinning for a constructed and unconstructed MEC physical system is included. The following are specific methods for performing digital twinning for a constructed MEC physical system and for performing digital twinning for an unconstructed MEC physical system, respectively.

[0139] (1) Digital twinning of the constructed MEC physical system:

[0140] Step 1: The MEC digital twin service description module generates a technical configuration file for the digital twin to be built based on the requirements for building the digital twin corresponding to the MEC physical system, as well as the historical data and key system configuration information of the stored MEC physical system. The file includes but is not limited to providing the modeling environment, modeling rules and modeling logic flow corresponding to the digital twin service, as well as key service quality indicators, system performance indicators and key system parameter requirements during simulation / emulation;

[0141] The MEC digital twin service description module sends the above configuration file to the MEC digital twin service deployment and operation module.

[0142] Step 2: The MEC digital twin service deployment and operation module builds the MEC digital twin orchestration module and the MEC digital twin data storage module;

[0143] The MEC digital twin service deployment and operation module sends a MEC digital twin service request to the MEC digital twin orchestration module, requesting to build a digital twin instance for the MEC physical system.

[0144] Step 3: The MEC digital twin orchestration module responds to the MEC digital twin service request and builds the MEC digital twin module, MEC digital twin visualization module and MEC digital twin operation management module corresponding to the digital twin instance at the digital twin level;

[0145] Step 4: The digital twin orchestration module converts the MEC digital twin service configuration file information of the MEC digital twin into the MEC model scenarios, model types and their semantic associations, corresponding service quality indicators, performance indicators and their parameter requirements, and corresponding key system parameter configuration information based on one or more perspectives corresponding to the construction of the digital twin instance by parsing the digital twin service request; as well as the construction-related operation logic corresponding to one or more MEC models and their association, combination, integration and fusion required when generating the digital twin instance;

[0146] The MEC digital twin policy management submodule provides the performance evaluation indicators and operation management policy rules of the constructed MEC digital twin instance at the MEC digital twin level and MEC model level.

[0147] Step 5: The MEC digital twin orchestration module establishes the key control modules required for the MEC model to provide module instances, mainly including the MEC model orchestration module and the MEC model data storage module.

[0148] Step 6: The MEC digital twin orchestration module sends one or more MEC model establishment requests corresponding to the digital twin of the MEC physical system to the MEC model orchestration module; wherein, when there are multiple MEC model establishment requests, each MEC model establishment request includes but is not limited to the scenario information required to build the MEC model, system key parameter configuration information, service quality indicator parameter information and performance evaluation indicator requirement information.

[0149] Step 7: The MEC data management module interacts with the MEC physical system through the communication module, periodically collects data from the MEC physical system, and stores the collected data in the MEC data storage submodule. The collected MEC physical system data includes but is not limited to configuration information and management information data of system-level management and host-level management, MEC platform configuration information and operation data, MEC APP deployment mode data, operation status data, and virtualization resource usage data.

[0150] The data preprocessing submodule in the MEC data management module preprocesses the MEC physical system data, the data analysis submodule analyzes the data, and the data fusion submodule fuses the data. The processed MEC physical system data is stored in the MEC data storage submodule.

[0151] Step 8: At the MEC model level, the MEC model orchestration module receives the request from the MEC digital twin orchestration module to establish the MEC model, and based on the historical data of the MEC physical system obtained from the MEC data storage submodule, builds one or more MEC models according to the perspective requirements of the MEC physical system, including but not limited to a MEC model for latency performance, a MEC model for computing energy consumption, a MEC model for virtualization resource allocation, and a MEC model for minimizing cost.

[0152] The MEC model orchestration module controls the generation of one or more corresponding MEC model instances based on the system key parameter configuration information related to the construction of the MEC model, and completes the related operations of building, deleting, arranging, and managing one or more MEC models, wherein the MEC model, MEC model management control module, and MEC model visualization module are generated for each MEC model instance. Among them, the MEC model management control module synchronizes, updates in real time, and optimizes the MEC model based on the operation data of the MEC physical system of the MEC data management module, and realizes the closed-loop optimization control of the MEC physical system; the MEC model visualization module visualizes the operation status and performance of the MEC model and provides interaction with the user; the MEC model data storage module is used to store the data information of the MEC model instance, including but not limited to the modeling scenario information, platform, code, and system key parameter configuration information of the MEC model instance.

[0153] The MEC model management and control module of each MEC model instance interacts with its MEC model data storage module to complete the storage of the system key parameter configuration information of the MEC model; through the interaction between each MEC model management and control module and the MEC data storage module, the MEC data storage module allocates storage resources for each MEC model instance to store its operating data.

[0154] After all MEC model instances are built, the MEC model orchestration module returns a successful establishment response of one or more corresponding MEC models to the MEC digital twin orchestration module in response to one or more MEC model establishment requests sent by the MEC digital twin orchestration module; the MEC model orchestration module notifies the MEC digital twin orchestration module and the MEC digital twin operation management module that the MEC model instance construction operation at the MEC model level is completed.

[0155] Step 9: At the digital twin level, the MEC digital twin operation management module interacts with the MEC digital twin data storage module, and the MEC digital twin data storage module completes the operation of storing the key information of the MEC digital twin instance, including but not limited to the MEC model contained in the digital twin and its semantic association; the modeling scenario information, platform, code of the digital twin instance and the system key parameter configuration information of the MEC model instance contained therein.

[0156] Step 10: The MEC digital twin orchestration module returns a response to the digital twin service request to the MEC digital twin service deployment and operation module to complete the construction of the digital twin instance.

[0157] Step 11: The MEC digital twin and its MEC models start running, providing digital twin services for the MEC physical system.

[0158] Step 12: The MEC digital twin task scheduling submodule generates a task scheduling table and its task execution order for operation management of the MEC digital twin based on the performance evaluation indicators and operation management policy rules at the MEC digital twin level and the MEC model level provided by the MEC digital twin policy management submodule, wherein the tasks include but are not limited to the performance analysis, performance evaluation tasks, data synchronization tasks, operation data collection tasks and operation data storage tasks for the MEC model and / or MEC digital twin generated when the MEC digital twin and one or more MEC models contained therein are characterized, or one or more MEC models formed by association, combination, integration and fusion are run;

[0159] Step 13: The MEC digital twin task scheduling submodule monitors the running status and running results corresponding to the performance analysis and performance evaluation tasks in the operation management scheduling task table based on the execution order of the task sequence number identifier in the operation management scheduling task table;

[0160] Step 14: Read the level category to which the task belongs and the result information of the analysis and / or evaluation task operation;

[0161] If the running status information of the analysis and / or evaluation task is normal, execute step 16; otherwise,

[0162] If the running result of the analysis and / or evaluation task is performance degradation / deterioration, and the category of the task is MEC level, then by notifying the corresponding MEC model orchestration module, the logical process (MEC-LEVEL-OCP) for implementing closed-loop continuous optimization control of the MEC physical system based on the MEC model level is called to complete the closed-loop optimization control process for the MEC model instance; based on the process, the MEC physical system control submodule in the MEC model management control module converts the optimization control problem into an optimization problem at the MEC model level that takes key service indicators / system performance indicators and / or system key parameters as optimization objectives and / or constraints based on modeling and / or heuristic and / or data-driven optimization methods, solves the optimization problem, generates optimization control information for the MEC physical system at the MEC model level, and sends the optimization control information to the MEC physical system through the communication module to complete the optimization control of the MEC physical system at the MEC model level; turn to step 15;

[0163] If the running result of the task is performance degradation / deterioration, and the category of the task is the MEC digital twin level, then by notifying the corresponding MEC digital twin orchestration module and / or MEC model orchestration module, the closed-loop optimization control process (DT-LEVEL-OCP) based on the MEC digital twin level is called to complete the closed-loop optimization control process corresponding to the task at the MEC digital twin level; based on the process, the MEC physical system control submodule in the MEC digital twin operation management module converts the optimization control problem into an optimization problem at the MEC digital twin level that takes key service indicators and / or system performance indicators and / or system key parameters as optimization objectives and / or constraints based on modeling and / or heuristics and / or data-driven methods, solves the optimization problem, and generates optimization control information for the MEC physical system at the MEC digital twin level, and sends the optimization control information to the MEC physical system through the communication module to complete the optimization control of the MEC physical system at the MEC digital twin level; turn to step 15;

[0164] Step 15: The MEC physical system receives the corresponding optimization control information; the MEC physical system completes the corresponding optimization operation process based on the optimization control information.

[0165] Step 16: Go to the next analysis and / or evaluation task in the management scheduling task list and return to step 13.

[0166] (2) Digital twinning of unbuilt MEC physical systems:

[0167] Before establishing the actual MEC physical system, the digital twin of the MEC system builds a MEC digital twin based on the MEC system design plan, simulates / emulates the operating status of the MEC physical system through the MEC digital twin, and gives the corresponding simulation / emulation evaluation results of the function and / or performance. When the simulation / emulation function and / or performance evaluation results of the MEC digital twin meet the required service quality indicators and system key parameter requirements, the corresponding actual MEC physical system is constructed based on the simulation / emulation results of the MEC digital twin instance and its corresponding system key parameter configuration information, and the digital twin service is performed on the constructed MEC physical system.

[0168] The following are the main steps for developing a digital twin for an unbuilt MEC physical system:

[0169] Step 1: The MEC digital twin service description module generates a technical configuration file for the digital twin to be built based on the requirements for building the digital twin corresponding to the MEC physical system. The configuration file includes but is not limited to the modeling environment, modeling rules and modeling logic flow corresponding to the digital twin service, as well as key service quality indicators, system performance indicators and system key parameter requirements when simulating / emulating the MEC system;

[0170] The MEC digital twin service description module sends the above configuration file to the MEC digital twin service deployment and operation module.

[0171] Step 2: The MEC digital twin service deployment and operation module builds the MEC digital twin orchestration module and the MEC digital twin data storage module;

[0172] The MEC digital twin service deployment and operation module sends a MEC digital twin service request to the MEC digital twin orchestration module, requesting to build a digital twin instance for the MEC physical system.

[0173] Step 3: The MEC digital twin orchestration module responds to the MEC digital twin service request and builds the MEC digital twin module, MEC digital twin visualization module and MEC digital twin operation management module corresponding to the digital twin instance at the digital twin level;

[0174] Step 4: The digital twin orchestration module converts the MEC digital twin service configuration file information of the MEC digital twin into the MEC model scenarios, model types and their semantic associations, corresponding service quality indicators, performance indicator requirements and parameter requirements, and corresponding system key configuration parameter information based on one or more perspectives corresponding to the construction of the digital twin instance by parsing the digital twin service request; and one or more MEC models required when generating the digital twin instance and their association, combination, integration and fusion corresponding MEC model instance construction operation logic;

[0175] The MEC digital twin policy management submodule provides the performance evaluation indicators and operation management policy rules of the constructed MEC digital twin instance at the MEC digital twin level and MEC model level.

[0176] Step 5: The MEC digital twin orchestration module establishes the key control modules corresponding to the MEC model provision module instance, which mainly includes the MEC model orchestration module and the MEC model data storage module;

[0177] Step 6: The MEC digital twin orchestration module sends one or more MEC model establishment requests corresponding to the digital twin of the MEC physical system to the MEC model orchestration module; wherein, when there are multiple MEC model establishment requests, each MEC model establishment request includes but is not limited to the scenario information of the MEC model, system key parameter configuration information, service quality indicators and / or performance indicators and their parameter requirements.

[0178] Step 7: At the MEC model level, the MEC model orchestration module receives the request from the MEC digital twin orchestration module to establish the MEC model, and builds one or more MEC model instances based on the perspective requirements of the MEC physical system, including but not limited to MEC model instances for latency performance, MEC model instances for computing energy consumption, MEC model instances for virtualization resource allocation, and MEC model instances for minimizing costs.

[0179] The MEC model orchestration module provides the system key parameter configuration information related to the construction of the MEC model, controls the generation of one or more corresponding MEC model instances, completes the construction, deletion, orchestration, and management of one or more MEC models, and generates the MEC model, MEC model management control module, and MEC model visualization module contained in each MEC model instance. Among them, the MEC model management control module is used to synchronize, update and optimize the MEC model in real time to achieve closed-loop optimization control of the MEC physical system; the MEC model visualization module is used to visualize the operating status and performance information of the MEC model and provide interaction with the user; the MEC model data storage module is used to store the information of the MEC model, including but not limited to the modeling scenario information, platform, code, and system key parameter configuration information of the MEC model.

[0180] The MEC model management and control module of each MEC model instance interacts with its MEC model data storage module to complete the storage of key system parameter configuration information for the MEC model; through the interaction between each MEC model management and control module and the MEC data storage module, the MEC data storage module allocates storage resources for each MEC model instance to store its operating data.

[0181] After all MEC model instances are built, the MEC model orchestration module returns a corresponding one or more MEC model construction success response to the MEC digital twin orchestration module in response to one or more MEC model establishment requests sent by the MEC digital twin orchestration module; at the same time, the MEC model orchestration module notifies the MEC digital twin orchestration module and the MEC digital twin operation management module that the MEC model instance construction operation at the MEC model level is completed.

[0182] Step 8: The MEC digital twin operation management module starts the MEC digital twin instance construction operation at the digital twin level. The MEC digital twin operation management module interacts with the MEC digital twin data storage module, and the MEC digital twin data storage module completes the operation of storing the key information of the MEC digital twin instance, including but not limited to the MEC model contained in the digital twin and its semantic association; the modeling scenario information, platform, code of the digital twin and the system key parameter configuration information of the MEC model instance contained therein.

[0183] Step 9: The MEC digital twin orchestration module returns a response to the digital twin service request to the MEC digital twin service deployment and operation module to complete the construction of the MEC digital twin instance.

[0184] Step 10: The MEC digital twin starts running and provides digital twin services for the MEC physical system.

[0185] Step 11: The MEC data management module interacts with the MEC digital twin through the communication module, periodically collects the operation data of the MEC digital twin instance, and stores the collected operation data in the MEC data storage submodule. The collected data of the MEC digital twin instance includes but is not limited to the configuration information and management information data of the system-level management and host-level management, MEC platform configuration information and operation data, MEC APP deployment mode data, operation data and virtualization resource usage data.

[0186] Step 12: The MEC digital twin optimization design submodule obtains the performance evaluation indicators and operation policy management rule information at the MEC digital twin level and the MEC model level from the MEC digital twin policy management submodule, and based on the policy management rule information, gives the performance evaluation indicators and key parameter information at the MEC digital twin level and the MEC model level respectively.

[0187] Step 13: The MEC digital twin optimization design submodule uses the DT-LEVEL-OCP process to start the closed-loop optimization control process for the digital twin instance, and based on the operation result information of the DT-LEVEL-OCP process, outputs the optimization design results of the MEC digital twin and its corresponding system key parameter configuration information to the MEC digital twin orchestration module and the MEC digital twin storage module. The MEC digital twin data storage module completes the operation of storing the key information of the MEC digital twin instance, including but not limited to the MEC model contained in the digital twin and its semantic association; the modeling scenario information, platform, code of the digital twin instance and the system key parameter configuration information of the MEC model instance contained therein.

[0188] Step 14: The MEC digital twin orchestration module evaluates the optimized design results.

[0189] If the functions and performance of the MEC digital twin meet the service quality indicators and system parameter requirements corresponding to the digital twin service request, the MEC digital twin orchestration module returns a successful response to the service request to the MEC digital twin service deployment and operation module, and the MEC digital twin service deployment and operation module interacts with the MEC digital twin data storage module to obtain the system optimization key configuration information for the digital twin corresponding to the MEC physical system to be constructed, and executes step 16;

[0190] If the functions and performance of the MEC digital twin do not meet the service quality indicators and system parameter requirements corresponding to the digital twin service request, the MEC digital twin orchestration module will notify the MEC digital twin optimization design submodule based on the policy management rule information, and visualize the result through the digital twin visualization module; at the same time, based on human-computer interaction, it can choose to use automatic and / or manual methods to adjust key parameters of the MEC digital twin.

[0191] The MEC digital twin optimization design submodule adopts the DT-LEVEL-OCP process to optimize the MEC digital twin instance until the operating performance of the MEC model contained in the MEC digital twin instance and / or the associated, combined, integrated and fused MEC model meets the corresponding service quality indicators and system parameter requirements.

[0192] The MEC digital twin optimization design submodule outputs the system key parameter configuration information of the optimized MEC digital twin instance to the MEC digital twin orchestration module and the MEC digital twin data storage module; the MEC digital twin data storage module stores the corresponding optimization design results and their corresponding system key parameter configuration information.

[0193] Step 15: The MEC digital twin orchestration module returns a successful response to the service request to the MEC digital twin service deployment and operation module. The MEC digital twin service deployment and operation module interacts with the MEC digital twin data storage module to obtain the system key configuration information of the MEC physical system corresponding to the MEC digital twin instance after orchestration optimization.

[0194] Step 16: The user establishes the actual MEC physical system based on the optimized key configuration information of the digital twin to be built output by the MEC digital twin.

[0195] Step 17: The MEC data management module periodically collects the operation data from the MEC physical system and stores the operation data in the MEC data storage submodule.

[0196] Step 18: The MEC digital twin operation management module completes the data synchronization between the operation status of the MEC digital twin instance and the operation status of the MEC physical system based on the operation data of the MEC physical system obtained from the MEC data management module.

[0197] Step 19: The MEC digital twin task scheduling submodule generates a task scheduling table and its task execution order for operation management of the MEC digital twin based on the performance evaluation indicators and operation management policy rules at the MEC digital twin level and the MEC model level provided by the MEC digital twin policy management submodule, wherein the tasks include but are not limited to the performance analysis, performance evaluation tasks, data synchronization tasks, operation data collection tasks and operation data storage tasks for the MEC model and / or MEC digital twin generated when one or more MEC models formed by association, combination, integration and fusion are run;

[0198] Step 20: The MEC digital twin task scheduling submodule monitors the running status and running results corresponding to the performance analysis and performance evaluation tasks in the operation management scheduling task table based on the execution order of the task sequence number identifier in the operation management scheduling task table;

[0199] Step 21: Read the level category to which the task belongs and the result information of the analysis and / or evaluation task operation;

[0200] If the running status information of the analysis and / or evaluation task is normal, execute step 23; otherwise,

[0201] If the running result of the analysis and / or evaluation task is performance degradation, and the category of the task is MEC level, then by notifying the corresponding MEC model orchestration module, the logical process (MEC-LEVEL-OCP) for implementing closed-loop optimization control of the MEC physical system based on the MEC model level is called to complete the closed-loop optimization control process for the MEC model instance; based on this process, the MEC physical system control submodule in the MEC model management control module converts the optimization control problem into an optimization problem at the MEC model level that takes key service quality indicators and / or system performance indicators and / or system key parameters as optimization objectives and / or constraints based on modeling and / or heuristic and / or data-driven optimization methods, solves the optimization problem, and generates optimization control information for the MEC physical system at the MEC model level, and sends the optimization control information to the MEC physical system through the communication module to complete the optimization control of the MEC physical system at the MEC model level. Go to step 22;

[0202] If the running result of the task is performance degradation / deterioration, and the category of the task is the MEC digital twin level, then by notifying the corresponding MEC digital twin orchestration module and / or MEC model orchestration module, the closed-loop optimization control process (DT-LEVEL-OCP) based on the MEC digital twin level is called to complete the closed-loop optimization control process corresponding to the task at the MEC digital twin level; based on this process, the MEC physical system control submodule in the MEC digital twin operation management module converts the optimization control problem into an optimization problem at the MEC digital twin level that takes the service quality indicators and / or system performance indicators and / or system key parameters as optimization objectives and / or constraints based on modeling and / or heuristic and / or data-driven optimization methods, solves the optimization problem, generates optimization control information for the MEC physical system at the MEC digital twin level, and sends the optimization control information to the MEC physical system through the communication module to complete the optimization control of the MEC physical system at the MEC digital twin level. Go to step 22;

[0203] Step 22: The MEC physical system receives the corresponding optimization control information. The MEC physical system completes the optimization operation process based on the optimization control information.

[0204] Step 23: Go to the next analysis and / or evaluation task in the management scheduling task list and return to step 20.

[0205] The digital twin method of edge computing according to the embodiment of the present invention can optimize the MEC physical system, improve the network management efficiency and resource utilization efficiency of the system, and reduce the operating cost and investment expenditure cost of MEC services. The present invention constructs a digital twin of the MEC system based on multi-perspective and multi-instance function and performance simulation / emulation of the MEC physical system, and completes the state perception, monitoring and closed-loop optimization control process of the MEC physical system based on the MEC digital twin.

[0206] Based on the above embodiments, Figure 7 As shown, this embodiment also provides a digital twin system for edge computing, which includes:

[0207] The external network entity 100 is used to use the external network to network the MEC physical system with the digital twin of the MEC system, so that the MEC physical system and the digital twin of the MEC system can exchange information;

[0208] The MEC physical system 200 is used to communicate with the MEC physical system through an external network and based on information interaction, so that the MEC physical system is mirrored in the digital twin of the MEC system; and

[0209] The digital twin 300 of the MEC system is used to monitor the MEC physical system through an external network, based on simulation and / or emulation, through data collection and storage, to obtain monitoring results, and analyze and evaluate the monitoring results to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and to perform digital twinning on the constructed MEC physical system and complete the closed-loop optimization control of the MEC physical system based on the evaluation results; to perform digital twinning on the unconstructed MEC physical system and complete the optimization design of the MEC physical system based on the evaluation results.

[0210] Furthermore, at the MEC digital twin level, the communication module interacts with the MEC physical system to complete the closed-loop optimization control of the MEC physical system; at the MEC model level, the communication module interacts with the MEC physical system to complete the closed-loop optimization control of the MEC physical system.

[0211] The edge computing digital twin system according to the embodiment of the present invention can optimize the MEC physical system, improve the system's network management efficiency and resource utilization efficiency, and reduce the operating cost and investment expenditure cost of MEC services. The present invention constructs a digital twin of the MEC system based on multi-perspective, multi-instance function and performance simulation / emulation of the MEC physical system, and completes the state perception, monitoring and closed-loop continuous optimization control process of the MEC physical system based on the MEC digital twin.

[0212] It should be noted that the above explanation of the embodiment of the digital twin method for edge computing is also applicable to the digital twin system for edge computing of this embodiment and will not be repeated here.

[0213] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A digital twin method for edge computing, characterized in that: include: Step S1, using an external network to network the MEC physical system with the digital twin of the MEC system, so that information is exchanged between the MEC physical system and the digital twin of the MEC system; Step S2, the MEC physical system is mirrored in the digital twin of the MEC system through the external network and based on the interaction of the information; as well as, Step S3, the digital twin of the MEC system is used through the external network to monitor the MEC physical system based on simulation and / or emulation, through data collection and storage, to obtain monitoring results, and the monitoring results are analyzed and evaluated to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and, the constructed MEC physical system is digitally twinned, and the closed-loop optimization control of the MEC physical system is completed based on the evaluation results; the unconstructed MEC physical system is digitally twinned, and the optimization design of the MEC physical system is completed based on the evaluation results; For the MEC digital twin level, the communication module is used to interact with the MEC physical system, and based on the MEC digital twin level, the closed-loop optimization control of the MEC physical system is completed; for the MEC model level, the communication module is used to interact with the MEC physical system, and based on the MEC model level, the closed-loop optimization control of the MEC physical system is completed; The digital twinning of the constructed MEC physical system includes: (1) The MEC digital twin service description module generates a configuration file of the digital twin to be constructed, and sends the configuration file to the MEC digital twin service deployment and operation module; (2) The MEC digital twin service deployment and operation module constructs the MEC digital twin orchestration module and the MEC digital twin data storage module, and sends a MEC digital twin service request to request the construction of a digital twin instance for the MEC physical system; (3) The MEC digital twin orchestration module responds to the MEC digital twin service request and constructs the MEC digital twin module, MEC digital twin visualization module and MEC digital twin operation management module corresponding to the digital twin instance at the digital twin level; (4) The digital twin orchestration module converts the service configuration file information into the configuration content corresponding to the construction of the digital twin instance by parsing the digital twin service request; (5) The MEC digital twin orchestration module establishes the key control modules required for the MEC model to provide module instances, including the MEC model orchestration module and the MEC model data storage module; (6) The MEC digital twin orchestration module sends one or more MEC model establishment requests corresponding to the digital twin of the MEC physical system to the MEC model orchestration module; The digital twinning of the constructed MEC physical system also includes: (7) The MEC data management module interacts with the MEC physical system through the communication module and collects data of the MEC physical system, and stores the collected data in the MEC data storage submodule. The data preprocessing submodule in the MEC data management module preprocesses the data of the MEC physical system, and uses the data analysis submodule to analyze the data and uses the data fusion submodule to fuse the data, and stores the processed MEC physical system data in the MEC data storage submodule; (8) At the MEC model level, the MEC model orchestration module receives a request from the MEC digital twin orchestration module to establish a MEC model, and constructs one or more MEC models based on obtaining historical data of the MEC physical system from the MEC data storage submodule; The MEC model orchestration module controls the generation of one or more corresponding MEC model instances based on the system key parameter configuration information related to the construction of the MEC model, including the generation of the MEC model for each MEC model instance, the MEC model management and control module, and the MEC model visualization module; The MEC model management and control module of each MEC model instance interacts with its MEC model data storage module to complete the storage of the system key parameter configuration information of the MEC model; through the interaction between each MEC model management and control module and the MEC data storage module, the MEC data storage module allocates storage resources for each MEC model instance to store its operating data; After all MEC model instances are built, the MEC model orchestration module returns a successful establishment response of one or more corresponding MEC models to the MEC digital twin orchestration module for one or more MEC model establishment requests sent by the MEC digital twin orchestration module; the MEC model orchestration module notifies the MEC digital twin orchestration module and the MEC digital twin operation management module that the MEC model instance construction operation at the MEC model level is completed; (9) At the digital twin level, the MEC digital twin operation management module interacts with the MEC digital twin data storage module, and the MEC digital twin data storage module completes the operation of storing the key information of the MEC digital twin instance; (10) The MEC digital twin orchestration module returns a response to the digital twin service request to the MEC digital twin service deployment and operation module to complete the construction of the digital twin instance; (11) Each MEC model to which the MEC digital twin belongs starts running, providing digital twin services for the MEC physical system; The digital twinning of the constructed MEC physical system also includes: (12.1) The MEC digital twin task scheduling submodule generates a task scheduling table and its task execution order for operation management of the MEC digital twin based on the performance evaluation indicators and operation management strategy rules at the MEC digital twin level and MEC model level provided by the MEC digital twin strategy management submodule; (13.1) The MEC digital twin task scheduling submodule monitors the running status and running results corresponding to the performance analysis and performance evaluation tasks in the operation management scheduling task table based on the task execution order identified by the task sequence number in the operation management scheduling task table; (14.1) Read the level category to which the task belongs and the result information of the analysis and / or evaluation task; If the running status information of the analysis and / or evaluation task is normal, execute step 16.1; otherwise, if the running result of the analysis and / or evaluation task is performance degradation / deterioration, and the category of the task is MEC model level, then call the closed-loop optimization control process of the MEC physical system based on the MEC model to complete the closed-loop optimization control corresponding to the task at the MEC model level. Specifically, the MEC physical system control submodule in the MEC model management control module converts the optimization control problem into an optimization problem at the MEC model level that takes the service quality index and / or system performance index and / or system key parameters as the optimization target and / or constraint condition based on the modeling and / or heuristic and / or data-driven optimization method, solves the optimization problem, generates information for optimizing the control of the MEC physical system at the MEC model level, and sends the optimization control information to the MEC physical system through the communication module, completes the optimization control of the MEC physical system at the MEC model level, and turns to step 15.1; If the running result of the task is performance degradation / deterioration, and the category of the task is the MEC digital twin level, then the closed-loop optimization control process based on the MEC digital twin level is called to complete the closed-loop optimization control at the MEC digital twin level corresponding to the task. Specifically, the MEC physical system control submodule in the MEC digital twin operation management module converts the optimization control problem into an optimization problem at the MEC digital twin level that takes the service quality index and / or system performance index and / or system key parameters as optimization objectives and / or constraints based on modeling and / or heuristics and / or data-driven methods, solves the optimization problem, generates optimization control information for the MEC physical system at the MEC digital twin level, and sends the optimization control information to the MEC physical system through the communication module, completes the optimization control of the MEC physical system at the MEC digital twin level, and turns to step 15.1; (15.1) The MEC physical system receives the corresponding optimization control information, and the MEC physical system completes the optimization operation process based on the optimization control information; (16.1), turn to the next analysis and / or evaluation task in the management scheduling task table, and return to step 13.1; The digital twinning of the unbuilt MEC physical system includes: First, by executing the above steps (1) to (6), the construction of key functional modules at the digital twin level and the MEC model level is completed. Secondly, by executing steps (8) to (11), the construction of the digital twin at the MEC model level and the MEC digital twin level is completed respectively. Thirdly, by executing step (7), the data collection, data processing and data storage during the normal operation of the constructed digital twin are completed. After that, It also includes performing the following steps: (12.2) The MEC digital twin optimization design submodule obtains the performance evaluation indicators and operation strategy management rule information at the MEC digital twin level and the MEC model level from the MEC digital twin strategy management submodule, and based on the strategy management rule information, gives the performance evaluation indicators and key parameter information at the MEC digital twin level and the MEC model level respectively; (13.2) The MEC digital twin optimization design submodule adopts the closed-loop optimization control process based on the MEC digital twin level to start the closed-loop optimization control process for the digital twin instance, and outputs the optimization design results of the MEC digital twin and its corresponding system key parameter configuration information to the MEC digital twin orchestration module and the MEC digital twin storage module according to the operation result information of the closed-loop optimization control process based on the MEC digital twin level; The digital twinning of the unbuilt MEC physical system also includes: (14.2) The MEC digital twin orchestration module evaluates the optimization design results; If the functions and performance of the MEC digital twin meet the service quality indicators and system parameter requirements corresponding to the digital twin service request, the MEC digital twin orchestration module returns a successful response to the service request to the MEC digital twin service deployment and operation module, and the MEC digital twin service deployment and operation module interacts with the MEC digital twin data storage module to obtain the optimized key configuration information of the digital twin corresponding to the MEC physical system to be constructed, and executes step 16.2; If the functions and performance of the MEC digital twin do not meet the service quality indicators and system parameter requirements corresponding to the digital twin service request, the MEC digital twin orchestration module notifies the MEC digital twin optimization design submodule based on the policy management rule information, and visualizes the result information through the digital twin visualization module; at the same time, based on human-computer interaction, choose to use automatic and / or manual methods to adjust key parameters of the MEC digital twin; The MEC digital twin optimization design submodule adopts the closed-loop optimization control process based on the MEC digital twin level to optimize the MEC digital twin instance until the operating performance of the MEC model contained in the MEC digital twin instance and / or the associated, combined, integrated and fused MEC model meets the corresponding service quality indicators and system parameter requirements; The MEC digital twin optimization design submodule outputs the system key configuration information of the optimized MEC digital twin instance to the MEC digital twin orchestration module and the MEC digital twin data storage module; the MEC digital twin data storage module stores the corresponding optimization design results and their corresponding system key parameter configuration information; (15.2) The MEC digital twin orchestration module returns a successful response to the service request to the MEC digital twin service deployment and operation module. The MEC digital twin service deployment and operation module interacts with the MEC digital twin data storage module to obtain the key system configuration information of the MEC physical system corresponding to the orchestration optimized MEC digital twin instance; (16.2) The user establishes the actual MEC physical system based on the optimized key configuration information of the digital twin to be built output by the MEC digital twin; (17.2) The MEC data management module collects system operation data from the MEC physical system and stores the operation data in the MEC data storage submodule; (18.2) The MEC digital twin operation management module completes the data synchronization between the operation status of the MEC digital twin instance and the operation status of the MEC physical system based on the operation data of the MEC physical system obtained from the MEC data storage submodule.

2. The digital twin method for edge computing according to claim 1, characterized in that: The method of performing a digital twin for the unbuilt MEC physical system, after executing step (18.2), further includes: Based on the constructed MEC digital twin, by executing the above steps (12.1)-(16.1), the closed-loop optimization control process of the digital twin at the digital twin level and the MEC model level is completed.

3. A digital twin system using edge computing according to the method of claim 1, characterized in that: include: An external network entity, used to use an external network to network the MEC physical system with the digital twin of the MEC system, so that information can be exchanged between the MEC physical system and the digital twin of the MEC system; The MEC physical system is used to communicate with the MEC physical system through the external network and based on the interaction of the information, so that the MEC physical system is mirrored in the digital twin of the MEC system; as well as, The digital twin of the MEC system is used to monitor the MEC physical system through the external network, based on simulation and / or emulation, through data collection and storage, to obtain monitoring results, and analyze and evaluate the monitoring results to obtain evaluation results, so as to complete the mirroring of the MEC physical system; and to perform digital twinning on the constructed MEC physical system, and complete closed-loop optimization control of the MEC physical system based on the evaluation results; to perform digital twinning on the unconstructed MEC physical system, and complete the optimization design of the MEC physical system based on the evaluation results.

4. The edge computing digital twin system according to claim 3, characterized in that: For the MEC digital twin level, the communication module is used to interact with the MEC physical system, and based on the MEC digital twin level, the closed-loop optimization control of the MEC physical system is completed; for the MEC model level, the communication module is used to interact with the MEC physical system, and based on the MEC model level, the closed-loop optimization control of the MEC physical system is completed.

Images