A cloud-edge collaborative production line monitoring method, device and system

Abstract

The present invention discloses a cloud-edge collaborative production line monitoring method, device and system, the method comprising: receiving production line data sent by a data acquisition terminal, parsing the production line data, determining standard data, and sending the standard data to the cloud; receiving a configuration file, using the configuration file to determine a workflow task; adding a service node that meets preset conditions in the workflow task to a task queue; storing a production line monitoring model in a mirror warehouse; receiving a production line monitoring model sent by the cloud, using a Docker container and a production line monitoring model; obtaining a task queue in the cloud, polling the task queue using a microservice; executing corresponding tasks according to the polling results, and determining the workflow results; receiving the workflow results sent by the edge terminal, and storing the workflow results. The present invention fully utilizes the advantages of edge-end and cloud-end collaboration, realizes timely monitoring and control of the production line at the edge terminal, and improves data processing efficiency.

Description

Technical Field

[0001] The present invention relates to the field of intelligent production line monitoring workflow technology, and specifically to a cloud-edge collaborative production line monitoring method, device and system. Background Art

[0002] With the development of intelligent manufacturing, factories are deeply integrating emerging technologies such as cloud computing, big data, cloud-edge collaborative production line monitoring, and the Internet of Things with existing systems to further intelligentize factory production systems. The manufacturing industry should innovate its business model and improve the production system architecture to increase flexibility at all levels, enabling production systems to customize products according to individual customer needs. With the development of society, traditional manufacturing systems are unable to meet the market demand for a wide variety of small-batch products, making system flexibility increasingly critical to their survival.

[0003] As a key subsystem of intelligent production systems, intelligent production line monitoring systems monitor and control on-site equipment, production conditions, and product status. They implement functions such as data collection, equipment control, defect and fault detection, and alarms for various signals, providing unified data services for the entire production process and enabling visualization of production data. However, current production line monitoring systems suffer from high coupling between modules, untimely data processing, and insufficient edge computing power. These systems are unable to dynamically adapt the monitoring system workflow, making it difficult to meet the flexible production and product requirements. Summary of the Invention

[0004] In view of this, the embodiments of the present invention provide a cloud-edge collaborative production line monitoring method, device and system, which solve the problems of high coupling degree between modules in the existing methods, untimely data processing, and insufficient computing power on the edge.

[0005] According to a first aspect, an embodiment of the present invention provides a cloud-edge collaborative production line monitoring method, which is applied to an edge end and includes:

[0006] Receive production line data sent by the data acquisition terminal, parse the production line data, determine standard data, and send the standard data to the cloud;

[0007] Receive the production line monitoring model sent by the cloud, and use the Docker container and the production line monitoring model to carry out the production line task execution microservice;

[0008] Obtain the task queue on the cloud, poll the task queue using the microservice, and determine the polling result;

[0009] Execute corresponding tasks according to the polling results, determine the workflow results, and send the workflow results to the cloud.

[0010] In combination with the first aspect, in a first implementation of the first aspect, polling the task queue using the microservice to determine a polling result includes:

[0011] Determine whether there is a task to be executed in the task queue;

[0012] When there is no task to be executed, return to continue to obtain the task queue on the cloud;

[0013] When there are tasks to be performed, a workflow is determined.

[0014] In combination with the first aspect, in the second implementation of the first aspect, sending the standard data to the cloud includes: passing the standard data as a parameter into the interface of the workflow engine service through the HTTP protocol, the workflow engine service stores the configuration file of the workflow in the cloud, and performs file search and matching through a non-relational database.

[0015] In combination with the first embodiment of the first aspect, in the third embodiment of the first aspect, parsing the production line data to determine standard data includes:

[0016] Performing data cleaning and data integration on the production line data to determine pre-processed data;

[0017] By parsing the pre-processed data, the workflow of the device is identified and bound, and standard data is determined.

[0018] In combination with the first aspect, in a fourth implementation of the first aspect, the use of the Docker container and the production line monitoring model to carry out the production line task execution microservice includes:

[0019] Taking the standard data as input parameters, using the input parameters to obtain the corresponding production line monitoring model for monitoring, and determining the monitoring results;

[0020] Utilize the Docker container and the monitoring results to execute the preset production line operation instructions in the workflow to determine the production line task execution microservice.

[0021] According to a second aspect, an embodiment of the present invention provides a cloud-edge collaborative production line monitoring method, which is applied in the cloud and includes:

[0022] receiving a configuration file, and determining a workflow task using the configuration file;

[0023] Using the workflow engine service, adding the service nodes that meet the preset conditions in the workflow task to the task queue;

[0024] Receiving the trained production line monitoring model, storing the production line monitoring model in an image warehouse, and sending the production line monitoring model in the image warehouse to an edge end;

[0025] The workflow results sent by the edge end are received and stored to realize production line monitoring.

[0026] Compared with the existing technology, the cloud-edge collaborative production line monitoring method provided by the embodiment of the present invention fully utilizes the advantages of edge-end and cloud-end collaboration, realizes timely monitoring and control of the production line at the edge, improves data processing efficiency, uses cloud computing capabilities to train algorithm models in the cloud, and realizes visualization of intelligent production line monitoring data. The present invention is no longer limited to local workflow orchestration for intelligent production line monitoring, and remote workflow orchestration can be achieved through the HTTP interface exposed by the cloud workflow engine. The present invention realizes the access of heterogeneous devices and heterogeneous networks through the edge-end data and protocol parsing module, and after conversion, it is connected to the system with unified data and protocol standards to achieve plug-and-play. Both the edge and cloud microservices are run in Docker containers, which fully utilizes server resources and facilitates system resource management.

[0027] In conjunction with the second aspect, in a first implementation of the second aspect, the workflow engine service includes a workflow execution engine and a workflow registration engine, the configuration file includes an edge-end configuration file and a user-end configuration file, and the received configuration file includes:

[0028] receiving standard data sent by the edge end, and determining the standard data as the edge end configuration file through the workflow execution engine;

[0029] The workflow registration engine is used to register the JSON file of the user terminal as the user terminal configuration file through the HTTP protocol.

[0030] In combination with the second aspect, in a second implementation of the second aspect, the method further includes: changing the configuration file according to changes in demand to update the task queue.

[0031] The cloud-edge collaborative production line monitoring method provided by the embodiment of the present invention, compared with the existing technical methods, can realize the orchestration of intelligent production line monitoring workflows without stopping production or restarting the system, thereby dynamically adapting to rich and varied production business scenarios, and realizing that intelligent production line monitoring can effectively be compatible with product flexibility, process flexibility, and expansion flexibility. When the product is replaced, the production process logic is changed, and the intelligent production line monitoring process is transformed, it is only necessary to upload the changed task module and register it in the workflow engine. Without modifying the source code of other task modules, the production line monitoring workflow can be changed only by rewriting the JSON orchestration file of the workflow. This method realizes the complete decoupling of each task execution module, and there is no dependency on each other, so that each task microservice module is easy to replace and reuse. The workflow engine adopts the inversion control method to obtain the workflow orchestration file, reads the configuration file information during operation to form an execution workflow process, and realizes the ability to dynamically change the existing production monitoring workflow.

[0032] According to a third aspect, an embodiment of the present invention provides a cloud-edge collaborative production line monitoring device, which is applied to an edge end and includes:

[0033] The parsing module is used to receive the production line data sent by the data acquisition terminal, parse the production line data, determine the standard data, and send the standard data to the cloud;

[0034] A monitoring module, configured to receive a production line monitoring model sent from the cloud, and to utilize a Docker container and the production line monitoring model to execute a production line task microservice;

[0035] A polling module is used to obtain the task queue on the cloud, poll the task queue using the microservice, and determine the polling result;

[0036] The execution module is used to execute the corresponding task according to the polling result, determine the workflow result, and send the workflow result to the cloud.

[0037] According to a fourth aspect, an embodiment of the present invention provides a cloud-edge collaborative production line monitoring device, which is applied in the cloud and includes:

[0038] A first processing module is configured to receive a configuration file and determine a workflow task using the configuration file;

[0039] The second processing module is used to use the workflow engine service to add the service nodes that meet the preset conditions in the workflow task to the task queue;

[0040] A third processing module is configured to receive the trained production line monitoring model, store the production line monitoring model in an image warehouse, and send the production line monitoring model in the image warehouse to an edge end;

[0041] The fourth processing module is used to receive the workflow results sent by the edge end and store the workflow results to achieve production line monitoring.

[0042] According to the fifth aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to execute the cloud-edge collaborative production line monitoring method described in the first aspect or any one of the embodiments of the first aspect, or executing the cloud-edge collaborative production line monitoring method described in the second aspect or any one of the embodiments of the second aspect.

[0043] According to the sixth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the cloud-edge collaborative production line monitoring method described in the first aspect or any one of the embodiments of the first aspect, or to execute the cloud-edge collaborative production line monitoring method described in the second aspect or any one of the embodiments of the second aspect.

[0044] According to a seventh aspect, an embodiment of the present invention provides a cloud-edge collaborative production line monitoring system, including:

[0045] At least one edge end, the edge end being configured to execute the cloud-edge collaborative production line monitoring method described in the first aspect or any one of the embodiments of the first aspect;

[0046] A cloud, connected to the at least one edge terminal, the cloud being configured to execute the cloud-edge collaborative production line monitoring method described in the second aspect or any one of the embodiments of the second aspect;

[0047] At least one client is connected to the cloud, and the client is used to determine the JSON file. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 This is a flow chart of the cloud-edge collaborative production line monitoring method applied to the edge according to an embodiment of the present invention;

[0050] Figure 2This is a flowchart of the edge production line monitoring task execution microservice according to a preferred embodiment of the present invention;

[0051] Figure 3 This is a flowchart of a cloud-edge collaborative production line monitoring method applied to the cloud according to an embodiment of the present invention;

[0052] Figure 4 is a flowchart of a decision task queue service according to an embodiment of the present invention;

[0053] Figure 5 is a flowchart of a cloud-edge collaborative production line monitoring method according to an embodiment of the present invention;

[0054] Figure 6 is another specific flow chart of the cloud-edge collaborative production line monitoring method according to an embodiment of the present invention;

[0055] Figure 7 is a schematic diagram of a cloud-edge collaborative production line monitoring device according to an embodiment of the present invention;

[0056] Figure 8 is another schematic diagram of a cloud-edge collaborative production line monitoring device according to an embodiment of the present invention;

[0057] Figure 9 is a schematic diagram of a cloud-edge collaborative production line monitoring system according to an embodiment of the present invention;

[0058] Figure 10 is another schematic diagram of a cloud-edge collaborative production line monitoring system according to an embodiment of the present invention;

[0059] Figure 11 It is a schematic diagram of the hardware structure of the electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0061] The cloud-edge collaborative production line monitoring method provided by the embodiment of the present invention enables dynamic adjustment and replacement of the production line monitoring workflow. It has the advantages of high process flexibility, convenient structural reconstruction, real-time data processing, and adaptability to rich production business scenarios, solving the problems in the existing technology.

[0062] According to an embodiment of the present invention, an embodiment of a cloud-edge collaborative production line monitoring method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0063] In this embodiment, a cloud-edge collaborative production line monitoring method is provided, which can be used for cloud-edge collaborative production line monitoring devices, such as computers, mobile phones, tablets, etc. Figure 1 Flowchart of the cloud-edge collaborative production line monitoring method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0064] S11, receiving the production line data sent by the data acquisition terminal, analyzing the production line data, determining the standard data, and sending the standard data to the cloud.

[0065] In this embodiment, the data and protocol parsing module parses and converts the collected data. The intelligent production line monitoring system needs to connect to production line equipment, sensors, and other systems from different manufacturers, categories, and interface styles, resulting in a high degree of heterogeneity. To adapt the intelligent production line monitoring system to different equipment and systems and flexibly modify the intelligent production line monitoring workflow, the transmitted data and protocols need to be standardized, ultimately serving as input parameters for the intelligent production line monitoring workflow. This work is performed by the data and protocol parsing module of the edge gateway.

[0066] The edge gateway provides external communication interfaces such as RS485, serial port, USB, and Modbus. The data and protocol parsing module unpacks the transmitted data according to the corresponding transmission protocol and then converts it into programming language entity objects and edge database formats through data mapping for storage. When communication with the cloud or other devices is required, the unified MQTT protocol is used to subscribe to and publish information through the MQTT broker on the EMQX platform. Each MQTT client has a unique client ID and a subscribed topic. When the MQTT client sends information to a designated topic channel, clients subscribed to that topic will receive the information. This enables communication between the edge and the cloud, between devices, and between systems using a unified data format and transmission protocol.

[0067] S12: Receive the production line monitoring model sent from the cloud and use the Docker container and the production line monitoring model to execute the production line task microservice. After receiving the production line monitoring model from the cloud, the edge device parses the standard format data obtained as interface parameters for the cloud-based workflow engine, which then selects the corresponding matching workflow for execution. The workflow event listener senses data preprocessing events and defect detection events. When an event occurs, it pulls the corresponding production line monitoring model image and runs it in a Docker container, executing the production line task microservice.

[0068] S13, obtain the task queue on the cloud, use the microservice to poll the task queue, and determine the polling result.

[0069] The edge production line monitoring task execution microservice registers in the task queue of the cloud workflow engine via HTTP and performs long polling. During the registration process, the number of threads is set to limit the number of worker threads that implement the task on the edge. The task execution microservice that implements the Worker interface is placed as a parameter in the task run configuration function in the form of a queue to complete the connection execution. The task name parameter must be consistent with the task name when registering the JSON configuration file. Figure 2 As shown, after the task execution module is connected, the edge task module will perform long polling on the task queue in the workflow engine. When a task with the same name is found in the task queue, the edge task execution module will execute it according to the parameters passed in the task queue and return the result to the workflow engine.

[0070] The input, output and process structure of the production line monitoring workflow need to be built through the JSON configuration file written by the user and sent to the cloud workflow engine for construction. First, a unified task execution template is built. A unified JSON template is designed by extracting the common features of the task execution microservices, where the service name is

[0071] The "serviceName" and method name "methodName" parameters determine the specific microservice method to call. The fully qualified name of the 5-parameter entity "className" is used as the parsed entity for the incoming parameter entity "params". Finally, the result of the task microservice execution will be stored in the result parameter "resultObject". The following is an example description of the scenario task execution module configuration file. The definition and meaning of the scenario configuration file in this example are as follows: [

[0073] 0{

[0074] #Task name

[0075] "name":"taskTemplate",

[0076] #Task Description

[0077] "description":"Task execution template",

[0078] 5#Reconnection times

[0079] "retryCount":3,

[0080] #Timeout reconnection time

[0081] "timeoutSeconds":1000,

[0082] #Input parameters

[0083] 0"inputKeys":[

[0084] "serviceName",

[0085] "methodName",

[0086] "className",

[0087] "params"

[0088] 5],

[0089] #Output parameters

[0090] "outputKeys":[

[0091] "resultObject"

[0092] ],

[0093] #Timeout reconnection strategy

[0094] "timeoutPolicy":"TIME_OUT_WF",

[0095] "retryLogic":"FIXED",

[0096] "retryDelaySeconds":600,

[0097] "responseTimeoutSeconds":900

[0098] } ]

[0100] The parameter names and parameter entities described above are stored in a key-value format. After uploading the task template, the user writes a JSON file to describe the production line monitoring workflow. The workflow engine uses the workflow relationship description statements provided by the workflow to orchestrate the task workflow. Finally, the file is transferred to the cloud-based non-relational database via HTTP for storage. The workflow engine generates the workflow process structure based on the configuration file and queues the tasks to be executed during operation, waiting for polling and execution by the edge task execution module.

[0101] S14, executing corresponding tasks according to the polling results, determining the workflow results, and sending the workflow results to the cloud.

[0102] After the cloud-edge collaborative production line monitoring device determines the polling result in the above S13, the workflow result callback is the final processing link of the intelligent production line monitoring workflow. This module is also written and uploaded by the user for registration. The workflow result callback module mainly performs customized processing on the results of the workflow. For example, the workflow processing results can be further encapsulated, the processed results can be stored in the edge database for preservation, or they can be published to the cloud through the MQTT client using the MQTT protocol for visual display or further analysis and calculation. This module is also managed by the workflow engine and can be replaced at any time. At the same time, different callback modules can be used according to different workflows.

[0103] Compared with the existing technology, the cloud-edge collaborative production line monitoring method provided in this embodiment fully utilizes the advantages of edge-end and cloud-end collaboration, realizes timely monitoring and control of production lines at the edge, improves data processing efficiency, uses cloud computing capabilities to train algorithm models in the cloud, and realizes visualization of intelligent production line monitoring data. The present invention is no longer limited to local workflow orchestration for intelligent production line monitoring, and remote workflow orchestration can be achieved through the HTTP interface exposed by the cloud workflow engine. The present invention realizes the access of heterogeneous devices and heterogeneous networks through the edge-end data and protocol parsing module, and after conversion, it is connected to the system with unified data and protocol standards to achieve plug-and-play. Both the edge and cloud microservices are run in Docker containers, which fully utilizes server resources and facilitates system resource management.

[0104] This embodiment provides a cloud-edge collaborative production line monitoring method, which includes the following steps:

[0105] S21, receiving the production line data sent by the data acquisition terminal, analyzing the production line data, determining the standard data, and sending the standard data to the cloud.

[0106] Specifically, the above step S21 further includes the following steps:

[0107] S211, performing data cleaning and data integration on the production line data and determining pre-processed data;

[0108] S212 , by parsing the pre-processed data, identifying and binding the workflow of the device, and determining the standard data.

[0109] In this embodiment, the data and protocol parsing module parses and converts data collected by monitoring devices and sensors. This data is then used to identify and bind product, device, and execution process information. This information is then passed as parameters to the workflow engine's interface via HTTP. The workflow engine then searches and matches workflow configuration files in a non-relational database stored in the cloud. Based on the matching workflow configuration files, a solution is selected in the decision queue service, and the task modules to be executed are placed into the task queue according to the solution.

[0110] S22 receives the production line monitoring model sent from the cloud, uses the Docker container and the production line monitoring model, and carries out the production line task execution microservice.

[0111] Specifically, the above step S22 further includes the following steps:

[0112] S221, using the standard data as input parameters, obtaining the corresponding production line monitoring model using the input parameters to perform monitoring and determine the monitoring results;

[0113] S222: Utilize the Docker container and monitoring results to execute the preset production line operation instructions in the workflow to determine the production line task execution microservice.

[0114] For details, please see Figure 1 S12 of the illustrated embodiment will not be described in detail here.

[0115] S23: Obtain the task queue on the cloud, poll the task queue using the microservice, and determine the polling result. Determine whether there are pending tasks in the task queue. If no pending tasks exist, return to obtaining the task queue on the cloud. If there are pending tasks, determine the workflow.

[0116] For details, please see Figure 1 S13 of the illustrated embodiment will not be described in detail here.

[0117] S24, executing corresponding tasks according to the polling results, determining the workflow results, and sending the workflow results to the cloud.

[0118] For details, please see Figure 1 S14 of the illustrated embodiment will not be described in detail here.

[0119] In this embodiment, a cloud-edge collaborative production line monitoring method is provided, which can be used in the cloud, such as computers, mobile phones, tablets, etc. Figure 3 Flowchart of the cloud-edge collaborative production line monitoring method according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0120] S31, receiving a configuration file, and determining a workflow task using the configuration file.

[0121] The intelligent production line monitoring task execution module microservice and monitoring workflow are registered. The intelligent production line monitoring task modules are hosted using Docker containers, achieving task resource and process isolation. Since the monitoring task microservices have no dependencies, they can be written independently. These microservices are then mirrored using Dockerfiles and uploaded to an image repository. The corresponding images are pulled as needed and run as containers on the edge. Each intelligent production line monitoring task microservice module uses Nacos as its configuration center, enabling remote configuration and hot deployment updates of code configuration via Nacos.

[0122] S32, using the workflow engine service, adding the service nodes in the workflow task that meet the preset conditions to the task queue.

[0123] A JSON file is written locally based on the intelligent production line monitoring process and product status, and then sent to the cloud-based workflow orchestration interface. The workflow engine searches for a matching item in the workflow configuration file's non-relational database using the workflow name in the JSON file. If no matching item exists, a new workflow is created. If a matching item exists, the content in the JSON file is updated to the matching item's configuration file.

[0124] After the data and protocol parsing module parses the intelligent production line monitoring data and identifies the corresponding product information and workflow name based on the parsed data, it executes the intelligent production line monitoring workflow with the requested name through an HTTP POST request, using the workflow name as the request name. The remaining parameters are passed as parameters to the first task execution module in the workflow orchestration. If the POST request returns a status code of "200", the workflow execution request is successful. Otherwise, the execution error requires investigation.

[0125] like Figure 2 As shown in , the workflow engine will start task decision scheduling after executing the requested workflow configuration file through inversion of control. Figure 4As shown, the decision task queue service selects a task execution microservice from the task queue with the same name for execution. The task scheduler sorts and calls the task execution microservice based on the CPU usage of the edge server where the task execution microservice is located. The lower the overall CPU usage of the edge server where the same task execution module microservice is located, the greater the probability of being called for execution.

[0126] S33, receiving the trained production line monitoring model, storing the production line monitoring model in the image warehouse, and sending the production line monitoring model in the image warehouse to the edge end.

[0127] First, based on the incoming workflow data type, a determination is made as to whether data preprocessing is required. If so, a further determination is made as to which data preprocessing model should be executed. This determination of whether data preprocessing is required is made in real time by the workflow engine's event monitoring module. The decision on which model to use utilizes the workflow engine's Decide function. Based on the incoming parameters, the Value of the decision parameter caseValueParam is assigned. The decision is then matched against the decisionCases parameter of the workflow engine's Switch selection function, selecting the corresponding task execution module or workflow branch for execution, thus enabling ongoing decision-making and workflow diversion.

[0128] After data preprocessing is complete, the results are returned to the workflow engine and used as input parameters for the next task execution module. The same evaluation is then performed on the input parameters. The data type of the input to the monitoring and compression algorithm model execution module determines whether the data should be run through the monitoring algorithm model. For example, if the data is determined to be for product defect monitoring or equipment fault inspection, the task module will be executed. If so, the task module will be further determined to determine which monitoring algorithm model to execute. Due to limited computing power on the edge, algorithm model training is performed in the cloud. After model training, the compressed model is packaged into an image file using a Dockerfile and stored in an image repository. When a new image is created or updated, the cloud will proactively push the latest version of the algorithm model image to the local image repository on the edge for storage. After determining which monitoring algorithm model to execute, the edge first determines whether the model container is already running. If so, it will directly invoke it. If not, it will pull the latest image of the model and run the container.

[0129] S34, receiving the workflow results sent by the edge end and storing the workflow results to achieve production line monitoring.

[0130] When the monitoring results come out, they enter the production line feedback control link, that is, the monitoring data results reach the production line operation instruction module, which is written and uploaded by the user. The production line operation instruction module will adjust the production line process parameters and process execution flow according to product changes, abnormal monitoring results, etc. This module can be composed of multiple sub-microservices. Users can upload production line operation task execution microservices under different circumstances according to their needs, and register tasks in the above manner. Finally, only the original workflow JOSN file needs to be adjusted without changing the source code of other microservices to implement changes in the intelligent production line monitoring workflow. Since the workflow engine reads the configuration file through the inversion control method, users can replace and upgrade the workflow task execution module and task execution sequence without stopping production or restarting the system, thereby achieving flexible adaptation to a rich and diverse production business environment.

[0131] Compared with the existing technical methods, the cloud-edge collaborative production line monitoring method provided in this embodiment can realize the orchestration of intelligent production line monitoring workflows without stopping production or restarting the system, thereby dynamically adapting to rich and varied production business scenarios, and realizing that intelligent production line monitoring can effectively be compatible with product flexibility, process flexibility, and expansion flexibility. When the product is replaced, the production process logic is changed, and the intelligent production line monitoring process is transformed, it is only necessary to upload the changed task module and register it in the workflow engine. Without modifying the source code of other task modules, the production line monitoring workflow can be changed only by rewriting the JSON orchestration file of the workflow. This method realizes the complete decoupling of each task execution module, and there is no dependency on each other, so that each task microservice module is easy to replace and reuse. The workflow engine adopts the inversion control method to obtain the workflow orchestration file, reads the configuration file information during operation to form the execution workflow process, and realizes the ability to dynamically change the existing production monitoring workflow.

[0132] This embodiment provides a cloud-edge collaborative production line monitoring method that can be applied to the cloud. The process includes the following steps:

[0133] S41, receiving a configuration file, and determining a workflow task using the configuration file.

[0134] The workflow engine service includes a workflow execution engine and a workflow registration engine. The configuration files include an edge configuration file and a user configuration file. Specifically, the above step S41 further includes the following steps:

[0135] S411, receiving standard data sent by the edge end, and determining the standard data as an edge end configuration file through the workflow execution engine;

[0136] S412: The workflow registration engine is used to register the JSON file on the user side as a user side configuration file via the HTTP protocol.

[0137] For details, please see Figure 3 S31 of the illustrated embodiment will not be described in detail here.

[0138] In this embodiment, after receiving the configuration file, the configuration file is modified according to the change in demand to achieve the update of the task queue.

[0139] S42: Using the workflow engine service, add the service nodes in the workflow task that meet the preset conditions to the task queue.

[0140] For details, please see Figure 3 S32 of the illustrated embodiment will not be described in detail here.

[0141] S43, receiving the trained production line monitoring model, storing the production line monitoring model in the image warehouse, and sending the production line monitoring model in the image warehouse to the edge end.

[0142] For details, please see Figure 3 S33 of the illustrated embodiment will not be described in detail here.

[0143] S44, receiving the workflow results sent by the edge end and storing the workflow results to achieve production line monitoring.

[0144] For details, please see Figure 3 S34 of the illustrated embodiment will not be described in detail here.

[0145] In this embodiment, a cloud-edge collaborative production line monitoring method is provided. Figure 5 As shown in the figure, data communication and data processing are carried out through the cloud and edge to realize cloud-edge collaborative production line monitoring.

[0146] S1, receives the production line data sent by the data acquisition terminal, analyzes the production line data, determines the standard data, and sends the standard data to the cloud; see Figure 1 S11 of the illustrated embodiment will not be described in detail here.

[0147] S2, receiving the configuration file and using it to determine the workflow task; the configuration file is sent by the user end; see Figure 3 S31 of the illustrated embodiment will not be described in detail here.

[0148] S3, using the workflow engine service, adds the service nodes that meet the preset conditions in the workflow task to the task queue; see Figure 3 S32 of the illustrated embodiment will not be described in detail here.

[0149] S4, receiving the trained production line monitoring model, storing the production line monitoring model in the image warehouse, and sending the production line monitoring model in the image warehouse to the edge end; wherein the model training process is performed on the user end, and the training process is an existing model training technology, which is not discussed in this embodiment; for details, please refer to Figure 3 S33 of the illustrated embodiment will not be described in detail here.

[0150] S5 receives the production line monitoring model sent by the cloud, uses the Docker container and the production line monitoring model, and deploys the production line task execution microservice. For details, see Figure 1 S12 of the illustrated embodiment will not be described in detail here.

[0151] S6, obtains the task queue on the cloud, uses microservices to poll the task queue, and determines the polling result; see Figure 1 S13 of the illustrated embodiment will not be described in detail here.

[0152] S7, executes the corresponding task according to the polling result, determines the workflow result, and sends the workflow result to the cloud; see Figure 1 S14 of the illustrated embodiment will not be described in detail here.

[0153] S8, receives the workflow results sent by the edge end and stores the workflow results to achieve production line monitoring; see Figure 3 S34 of the illustrated embodiment will not be described in detail here.

[0154] In practical applications, please refer to Figure 6 , wherein the perception layer mainly includes acquisition equipment, sensors, etc. The specific implementation process is detailed in the above method embodiment, and will not be repeated in this embodiment.

[0155] In this embodiment, a cloud-edge collaborative production line monitoring device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation methods, and will not be repeated here. As used below, the term "module" can implement a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0156] This embodiment provides a cloud-edge collaborative production line monitoring device, such as Figure 7As shown, it includes: a parsing module 1, which is used to receive the production line data sent by the data acquisition end, parse the production line data, determine the standard data, and send the standard data to the cloud; a monitoring module 2, which is used to receive the production line monitoring model sent by the cloud, use the Docker container and the production line monitoring model, and carry out the production line task execution microservice; a polling module 3, which is used to obtain the task queue on the cloud, use the microservice to poll the task queue, and determine the polling result; an execution module 4, which is used to execute the corresponding task according to the polling result, determine the workflow result, and send the workflow result to the cloud.

[0157] This embodiment provides a cloud-edge collaborative production line monitoring device, such as Figure 8 As shown, it includes: a first processing module 01, which is used to receive a configuration file and use the configuration file to determine a workflow task; a second processing module 02, which is used to use the workflow engine service to add the service nodes that meet the preset conditions in the workflow task to the task queue; a third processing module 03, which is used to receive the trained production line monitoring model, store the production line monitoring model in the image warehouse, and send the production line monitoring model in the image warehouse to the edge end; a fourth processing module 04, which is used to receive the workflow results sent by the edge end and store the workflow results to realize production line monitoring.

[0158] The cloud-edge collaborative production line monitoring device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0159] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0160] This embodiment also provides a cloud-edge collaborative production line monitoring system, such as Figure 9 As shown, including:

[0161] At least one edge end 010, the edge end 010 is used to execute the cloud-edge collaborative production line monitoring method in the first aspect or any one embodiment of the first aspect.

[0162] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0163] The cloud 020 is connected to at least one edge end 010, and the cloud 020 is used to execute the cloud-edge collaborative production line monitoring method in the second aspect or any one of the embodiments of the second aspect.

[0164] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0165] At least one client 030 is connected to the cloud 020, and the client 030 is used to determine the JSON file.

[0166] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0167] Specifically, if Figure 10 As shown, this embodiment provides an intelligent production line monitoring workflow orchestration system based on edge-cloud collaboration, comprising the following components: edge and cloud. The edge microservices of this intelligent production line monitoring workflow orchestration system include: a data and protocol parsing module, a data preprocessing module, a monitoring compression algorithm model execution module, a production line instruction operation module, and a workflow result callback module. The cloud microservices of this intelligent production line monitoring workflow orchestration system include: a workflow engine module, a monitoring algorithm compression model image repository, a non-relational database for workflow configuration files, and a data visualization panel module.

[0168] The data and protocol parsing module identifies and converts the data and transmission protocols collected by sensors, equipment, and other systems connected to the intelligent production line monitoring system, and outputs them in a standard data format and unified protocol. The data preprocessing module performs data cleansing and integration based on the data parsed by the data and protocol parsing module. The monitoring compression algorithm model execution module uses the cleansed and integrated data as input parameters to pull the corresponding product and equipment monitoring algorithm model images for defect and fault detection. The production line instruction operation module decides and selects the production line operation instruction code written and uploaded by the user in the workflow based on the monitoring results. The workflow result callback module processes the results of the workflow execution and publishes them to the cloud through the MQTT client.

[0169] The workflow engine module provides operations for creating, orchestrating, managing, and maintaining intelligent production line monitoring workflows, while also exposing an HTTP interface for external operations. The monitoring algorithm compression model image repository stores trained production line monitoring algorithms and pushes the latest version of the monitoring algorithm compression model image to the edge based on version updates. A non-relational database stores intelligent production line monitoring workflow configuration files. The data visualization panel module visualizes intelligent production line monitoring information pushed to the edge via the MQTT client.

[0170] The embodiment of the present invention also provides a mobile terminal having the above Figure 11 The cloud-edge collaborative production line monitoring device shown.

[0171] See also Figure 11 , Figure 11 is a schematic diagram of the structure of a terminal provided by an optional embodiment of the present invention, such as Figure 11As shown, the terminal may include: at least one processor 601, such as a CPU (Central Processing Unit), at least one communication interface 603, a memory 604, and at least one communication bus 602. The communication bus 602 is used to realize the connection and communication between these components. The communication interface 603 may include a display screen (Display), a keyboard (Keyboard), and the optional communication interface 603 may also include a standard wired interface and a wireless interface. The memory 604 may be a high-speed RAM memory (Random Access Memory, volatile random access memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 604 may optionally be at least one storage device located away from the aforementioned processor 601. The processor 601 may be combined with the above-mentioned device, the memory 604 stores an application program, and the processor 601 calls the program code stored in the memory 604 to execute any of the above-mentioned method steps.

[0172] The communication bus 602 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 602 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 11 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0173] Among them, the memory 604 may include volatile memory (English: volatile memory), such as random-access memory (English: random-access memory, abbreviated: RAM); the memory may also include non-volatile memory (English: non-volatile memory), such as flash memory (English: flash memory), hard disk drive (English: hard disk drive, abbreviated: HDD) or solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 604 may also include a combination of the above types of memory.

[0174] The processor 601 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.

[0175] The processor 601 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0176] Optionally, the memory 604 is also used to store program instructions. The processor 601 can call the program instructions to implement the application Figure 1 and 3 The method shown in the examples.

[0177] An embodiment of the present invention further provides a non-transitory computer storage medium storing computer-executable instructions, which can execute the cloud-edge collaborative production line monitoring method in any of the above method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); the storage medium can also include a combination of the above types of memory.

[0178] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A cloud-edge collaborative production line monitoring method, characterized in that: Applied to the edge, the method includes: Receive production line data sent by the data acquisition terminal, parse the production line data, determine standard data, and send the standard data to the cloud; Receive the production line monitoring model sent by the cloud, and use the Docker container and the production line monitoring model to carry out the production line task execution microservice; Obtain the task queue on the cloud, poll the task queue using the microservice, and determine the polling result; Execute corresponding tasks according to the polling results, determine workflow results, and send the workflow results to the cloud; The sending of the standard data to the cloud comprises: passing the standard data as a parameter to an interface of a workflow engine service via an HTTP protocol, wherein the workflow engine service stores a configuration file of the workflow in the cloud and performs file search and matching via a non-relational database; The step of analyzing the production line data to determine standard data includes: Performing data cleaning and data integration on the production line data to determine pre-processed data; By parsing the pre-processed data, the workflow of the device is identified and bound, and standard data is determined.

2. The method according to claim 1, characterized in that The polling the task queue using the microservice to determine the polling result includes: Determine whether there is a task to be executed in the task queue; When there is no task to be executed, return to continue to obtain the task queue on the cloud; When there are tasks to be performed, a workflow is determined.

3. The method according to claim 1, characterized in that The Docker container and the production line monitoring model are used to carry out the production line task execution microservice, including: Taking the standard data as input parameters, using the input parameters to obtain the corresponding production line monitoring model for monitoring, and determining the monitoring results; Utilize the Docker container and the monitoring results to execute the preset production line operation instructions in the workflow to determine the production line task execution microservice.

4. A cloud-edge collaborative production line monitoring method, characterized in that: Applied to the cloud, the method includes: receiving a configuration file, and determining a workflow task using the configuration file; Using the workflow engine service, adding the service nodes that meet the preset conditions in the workflow task to the task queue; Receiving the trained production line monitoring model, storing the production line monitoring model in an image warehouse, and sending the production line monitoring model in the image warehouse to an edge end; Receiving the workflow results sent by the edge end and storing the workflow results to achieve production line monitoring; The workflow engine service includes a workflow execution engine and a workflow registration engine. The configuration file includes an edge-end configuration file and a user-end configuration file. The received configuration file includes: receiving standard data sent by the edge end, and determining the standard data as the edge end configuration file through the workflow execution engine; The workflow registration engine is used to register the JSON file of the user terminal as the user terminal configuration file through the HTTP protocol.

5. A cloud-edge collaborative production line monitoring device, characterized in that: include: The parsing module is used to receive the production line data sent by the data acquisition terminal, parse the production line data, determine the standard data, and send the standard data to the cloud; The sending of the standard data to the cloud includes: passing the standard data as a parameter to an interface of a workflow engine service via the HTTP protocol, wherein the workflow engine service stores a configuration file of the workflow in the cloud and performs file search and matching via a non-relational database; the parsing of the production line data to determine the standard data includes: performing data cleaning and data integration on the production line data to determine pre-processed data; and identifying and binding the workflow of the equipment by parsing the pre-processed data to determine the standard data; A monitoring module, configured to receive a production line monitoring model sent from the cloud, and to utilize a Docker container and the production line monitoring model to execute a production line task microservice; A polling module is used to obtain the task queue on the cloud, poll the task queue using the microservice, and determine the polling result; The execution module is used to execute the corresponding task according to the polling result, determine the workflow result, and send the workflow result to the cloud.

6. A cloud-edge collaborative production line monitoring device, characterized in that: include: A first processing module is configured to receive a configuration file and determine a workflow task using the configuration file, wherein the configuration file includes an edge-end configuration file and a user-end configuration file, wherein receiving the configuration file includes: receiving standard data sent by the edge-end, and determining the standard data as the edge-end configuration file through the workflow execution engine; and determining the user-end JSON file as the user-end configuration file through the workflow registration engine via the HTTP protocol; A second processing module is configured to add service nodes in the workflow task that meet preset conditions to a task queue by using a workflow engine service, wherein the workflow engine service includes a workflow execution engine and a workflow registration engine; A third processing module is configured to receive the trained production line monitoring model, store the production line monitoring model in an image warehouse, and send the production line monitoring model in the image warehouse to an edge end; The fourth processing module is used to receive the workflow results sent by the edge end and store the workflow results to achieve production line monitoring.

7. A cloud-edge collaborative production line monitoring system, characterized in that: include: At least one edge end, the edge end being used to execute the cloud-edge collaborative production line monitoring method according to any one of claims 1 to 3; A cloud, connected to the at least one edge terminal, and the cloud is used to execute the cloud-edge collaborative production line monitoring method described in claim 4; at least one user terminal, connected to the cloud, and the user terminal is used to determine the JSON file.

Images