An intelligent control method and system based on edge computing assistance

Abstract

The present invention relates to the field of edge computing intelligent control, and provides an edge computing-assisted intelligent control method and system, including a central cloud generating first configuration information according to production parameters and sending it to a corresponding edge server, and the edge server obtains a target algorithm file and completes the configuration accordingly. During the production process, if the central cloud detects that the load of any edge server is too high, it searches for idle edge servers in the local area network, and sends a second configuration information containing algorithm program configuration information and computing modules allowed to be used to both. The idle edge server obtains the algorithm file and configuration program based on this information, and continues the calculation process to ensure task continuity and integrity, and finally feeds back the calculation results to the busy edge server. The busy edge server controls the production line to perform corresponding operations through the edge node based on the results. The computing efficiency and production flexibility are improved, while avoiding the waste of computing resources.

Description

Technical Field

[0001] The present invention relates to the field of edge computing intelligent control technology, and in particular to an edge computing-assisted intelligent control method and system. Background Art

[0002] The contents in this section merely provide background information related to the present invention and may not constitute prior art.

[0003] The application of edge computing in automated production lines is gradually becoming a mainstream trend in the industry. This technological innovation has brought unprecedented efficiency and real-time performance to industrial automation control by directly sinking data processing and computing capabilities to edge devices close to data generation. On highly automated production lines, massive amounts of real-time data need to be processed and analyzed quickly to ensure the stable operation and optimal efficiency of the production process. The introduction of edge computing technology, through the flexible deployment of computing resources around the production line, has realized the ability to process and analyze data instantly at the source of data transmission. This transformation has greatly shortened the delay of data transmission and significantly improved the response speed and operating efficiency of the entire system.

[0004] With the continuous expansion of production scale, the continuous increase in the number of production lines has become an inevitable trend. However, this trend also brings challenges to the configuration of edge servers. Under the current technical architecture, each edge server operates independently and is responsible for processing the information of multiple edge nodes. For example, a production line control method, device and medium based on edge computing disclosed in publication number CN114004548B adopts the control of multiple edge computing devices in each edge computing device group. This one-to-one or one-to-many computing mode may still meet the needs in the early stage, but with the continuous addition of production lines, its limitations are becoming increasingly prominent. Whenever a new production line needs to be added, in order to ensure the real-time processing and transmission of information, the company has to increase the number of edge servers accordingly. This process not only involves the procurement, installation and debugging of servers, but also requires corresponding adjustments to the network architecture, which leads to an extension of the construction period and an increase in costs. At the same time, when certain specific types of production lines are temporarily closed due to changes in market demand, their corresponding edge servers are often idle and cannot be effectively reused. This not only causes a waste of resources, but also reduces the overall operational efficiency. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide an intelligent control method and system based on edge computing assistance. By dividing the algorithm file into multiple computing modules and configuring different computing modules in multiple different edge servers, the construction period and cost of the new production line can be reduced.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] In a first aspect, an edge computing-assisted intelligent control method is applied to an automated production line, comprising a central cloud, multiple edge servers, and multiple edge nodes; the multiple edge servers are all connected to the central cloud for communication, each edge server is connected to multiple edge nodes for communication, and communication connections are established between the edge servers;

[0008] The above methods include:

[0009] Based on the input production parameters, the central cloud sends the first configuration information to the corresponding edge server according to the product category of the production parameters; a plurality of algorithm files are pre-stored in the central cloud, and each algorithm file is divided into a plurality of independently operable computing modules according to the computing order;

[0010] The edge server obtains the corresponding target algorithm file from the central cloud according to the first configuration information, and configures the algorithm program; the edge server sends a production request to the production server; the edge server pre-stores a plurality of control instructions for controlling the production line;

[0011] In response to the permission of the production request, the edge server starts the detection of computing load and reports the load status data to the central cloud in real time; at the same time, the edge server sends a start instruction to the edge node corresponding to the product category;

[0012] In response to the start instruction, the edge node controls the corresponding number of production lines to start production, and feeds back detection information to the edge server in real time through the edge node.

[0013] When the central cloud detects that the computing load of any edge server is greater than a preset threshold, it is defined as a busy edge server, and searches for idle edge servers in the local area network; and sends second configuration information to the busy edge server and the idle edge server; the second configuration information includes program configuration information of the target algorithm file, and computing modules allowed to be used by the program in the edge server, and all allowed computing modules perform data transmission in a corresponding computing order according to the address of the server;

[0014] The idle edge server obtains the corresponding algorithm file from the central cloud according to the second configuration information, and configures the algorithm program; the idle edge server obtains the intermediate data from the busy edge server, and calculates in sequence according to the calculation order, and feeds back the calculation results to the busy edge server;

[0015] The busy edge server controls the corresponding production line through the edge node to perform corresponding operations based on the results.

[0016] Further, after searching for idle edge servers in the local area network, the method further includes:

[0017] If the search result is 0, an edge server whose computing load is less than a preset second threshold value in the local area network is searched as an idle edge server.

[0018] Further, after searching for an edge server in the local area network whose computing load is less than a preset second threshold, the method further includes:

[0019] If the search result is 0, the central cloud configures the program of the target algorithm locally according to the target algorithm file, and sends third configuration information to the busy edge server, where the third configuration information includes a computing module allowed to be used by the program in the edge server, and the computing module is a computing module not allowed to be used in the central cloud;

[0020] The central cloud obtains intermediate data from the busy edge servers, and calculates them in sequence according to the corresponding calculation order, and the calculation results are fed back to the busy edge servers.

[0021] Furthermore, before sending the second configuration information to the busy edge server and the idle edge server, the method further includes:

[0022] According to the allocation model, the computing module of the corresponding computing model is allocated to the idle edge server to achieve load balancing; the allocation model includes: obtaining the computing difficulty index of the computing module and the performance index of each edge server; and allocating the computing module to the corresponding idle edge server based on the ratio of the difficulty index to the performance index plus the minimum load of the corresponding server as a constraint condition.

[0023] Furthermore, the distribution model expression is:

[0024]

[0025] in, is the computing performance of the ith server, is the computational difficulty of the j-th model, is the current load of the i-th server, is whether the j-th model is assigned to the i-th server, k is the number of servers, and C is a safety factor to ensure that the server is not overloaded. The server index selected for model j.

[0026] Furthermore, the method further comprises:

[0027] When the central cloud receives new production parameters, it queries the corresponding edge server based on the product category of the production parameters to see if it has been configured; if not, it sends the corresponding configuration information;

[0028] If it has been configured, query the configuration information of the edge server, and configure the corresponding algorithm program and license the corresponding computing module according to the configuration information;

[0029] According to the transmission order of the data in the configuration information, the upstream server and downstream server of the edge server are queried, and update information is sent to the upstream server and downstream server to change the sending address of the upstream server to the central cloud and the receiving address of the downstream server to the central cloud.

[0030] Furthermore, the method further comprises:

[0031] When the central cloud receives the stop command, it sends a command to stop material supply to the production server, and sends a delayed stop command to the edge node through the edge server.

[0032] In a second aspect, the present invention provides an intelligent control system based on edge computing assistance, which is applied to an automated production line, including a central cloud, multiple edge servers and multiple edge nodes; the multiple edge servers are all connected to the central cloud for communication, each edge server is connected to multiple edge nodes for communication, and communication connections are established between the edge servers;

[0033] The system includes:

[0034] The production module, based on the input production parameters, the central cloud sends the first configuration information to the corresponding edge server according to the product category of the production parameters; the central cloud pre-stores a plurality of algorithm files, and each algorithm file is divided into a plurality of independently operable computing modules according to the computing order;

[0035] The edge service module is used for the edge server to obtain the corresponding target algorithm file from the central cloud according to the first configuration information and configure the algorithm program; the edge server sends a production request to the production server; the edge server pre-stores a plurality of control instructions for controlling the production line;

[0036] Load detection module, in response to the permission of the production request, the edge server starts the detection of the computing load and reports the load status data to the central cloud in real time; at the same time, the edge server sends a start instruction to the edge node corresponding to the product category;

[0037] The detection information acquisition module responds to the start instruction, and the edge node controls the corresponding number of production lines to start production, and feeds back the detection information to the edge server in real time through the edge node.

[0038] The allocation module is used to define a busy edge server as a busy edge server when the central cloud detects that the computing load of any edge server is greater than a preset threshold, search for idle edge servers in the local area network; and send second configuration information to the busy edge server and the idle edge server; the second configuration information includes program configuration information of the target algorithm file, and computing modules allowed to be used by the program in the edge server, and all computing modules allowed to be used perform data transmission in a corresponding computing order according to the address of the server;

[0039] A calculation module is used for the idle edge server to obtain the corresponding algorithm file from the central cloud according to the second configuration information and configure the algorithm program; the idle edge server obtains the intermediate data from the busy edge server, and calculates in sequence according to the calculation order, and feeds back the calculation results to the busy edge server;

[0040] The execution module is used to control the corresponding production line through the edge node to perform corresponding operations according to the results of the busy edge server.

[0041] In a third aspect, the present invention provides an electronic device comprising a processor, a storage medium and a computer program, wherein the computer program is stored in the storage medium and is characterized in that when the computer program is executed by the processor, the control method of the first aspect is implemented.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the control method of the first aspect.

[0043] In summary, the technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0044] The present invention realizes efficient scheduling of production tasks and flexible allocation of computing resources through the collaborative work of the central cloud and the edge server. The central cloud accurately identifies the product category to which the production task belongs according to the production parameters, and generates targeted first configuration information and sends it to the corresponding edge server. The edge server obtains the target algorithm file and completes the configuration accordingly, and then sends a production request to the production server. Once the production request is approved, the edge server starts to monitor the computing load and reports to the central cloud in real time, and sends a start instruction to the corresponding edge node to control the production line to start production. During the production process, the edge node feeds back the detection information in real time to monitor the production status. If the central cloud detects that the load of any edge server is too high, the load balancing strategy is started, the idle edge server in the local area network is searched, and the second configuration information containing the algorithm program configuration information and the computing module allowed to be used is sent to both. The idle edge server obtains the algorithm file and configuration program according to this information, and obtains the intermediate data from the busy edge server, continues the calculation process, ensures the continuity and integrity of the task, and finally feeds back the calculation results to the busy edge server. The busy edge server then accurately controls the production line to perform corresponding operations through the edge node based on these results. This mechanism not only significantly improves computing efficiency and production flexibility, but also effectively avoids the waste of computing resources, providing strong support for factories to quickly respond to changes in production tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of an edge computing-assisted intelligent control method provided by the present invention;

[0046] Figure 2 It is a schematic diagram of the structure of edge nodes, edge servers and central cloud in the present invention;

[0047] Figure 3 A schematic diagram of the allocation of computing tasks in the central cloud to the edge servers in the present invention;

[0048] Figure 4 This is a flow chart of when the central cloud takes over the edge server in the present invention;

[0049] Figure 5 A schematic diagram of the structure of an edge computing-assisted intelligent control system provided by the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions 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 part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0051] The following reference Figures 1 to 4 The present invention is described in further detail.

[0052] Please refer to Figure 1 An edge computing-assisted intelligent control method proposed in an embodiment of the present invention is applied to an automated production line, and is characterized in that: it includes a central cloud, multiple edge servers, and multiple edge nodes; the multiple edge servers are all connected to the central cloud for communication, each edge server is connected to multiple edge nodes for communication, and communication connections are established between the edge servers, such as Figure 2 shown.

[0053] The above methods include:

[0054] S101, based on the input production parameters, the central cloud sends first configuration information to the corresponding edge server according to the product category of the production parameters; the central cloud pre-stores multiple algorithm files, each algorithm file is divided into multiple independently runnable computing modules according to the calculation order, such as Figure 3 shown.

[0055] Specifically, when the system receives input production parameters, such as product type, production quantity, and specific specifications, the central cloud, as the hub for data processing and decision-making, will first accurately identify the product category to which the production task to be executed belongs based on these parameters. This process ensures that the subsequent allocation of computing resources can be accurately connected to the specific production needs of the product. At the same time, the central cloud pre-stores a diverse library of algorithm files. These algorithms cover a variety of aspects, from production scheduling to image recognition and text recognition, and are designed to provide comprehensive technical support for different types of production tasks. What is particularly important is that each algorithm file is carefully designed as multiple independently run computing modules, which are arranged in a specific calculation order, which not only ensures the flexibility of algorithm execution, but also facilitates the splitting and parallel processing of computing tasks when necessary, thereby improving overall computing efficiency.

[0056] After accurately identifying the product category to which the production task belongs, the central cloud will quickly select and generate the first configuration information required for the production of this type of product based on this category information. These configuration information specifies in detail the key elements such as parameter settings, algorithm selection, and execution order that the edge server should follow when performing computing tasks. Subsequently, the central cloud sends this configuration information to the corresponding edge server, realizing the precise deployment and instant scheduling of computing tasks for the edge server.

[0057] S102, the edge server obtains the corresponding target algorithm file from the central cloud according to the first configuration information, and configures the algorithm program; the edge server sends a production request to the production server; the edge server pre-stores a plurality of control instructions for controlling the production line.

[0058] Specifically, after receiving the first configuration information, the edge server will obtain the corresponding target algorithm file from the central cloud and configure the algorithm program. This process ensures that the edge server can perform production tasks using the correct algorithms and parameters in accordance with the instructions of the central cloud. After the configuration is completed, the edge server will send a production request to the production server. This step is intended to facilitate manual operation confirmation to avoid errors in the issuance of production tasks. At the same time, the edge server pre-stores multiple control instructions for controlling the production line. These instructions include but are not limited to starting, stopping, adjusting production speed, etc., which provide a basis for the automated control of the production line.

[0059] S103, in response to the permission of the production request, the edge server starts the detection of computing load and reports the load status data to the central cloud in real time; at the same time, the edge server sends a start instruction to the edge node corresponding to the product category. Among them, the detection of computing load of the edge server mainly monitors the load of the processor such as the detector CPU in real time.

[0060] Specifically, once a production request is approved, the edge server will start monitoring the computing load and report the load status data to the central cloud in real time. This step is crucial for timely detection and handling of overload conditions on the edge server. At the same time, the edge server will send a startup instruction to the edge nodes corresponding to the product category, and these edge nodes are responsible for controlling the corresponding number of production lines to start production.

[0061] S104, in response to the start instruction, the edge node controls the corresponding number of production lines to start production, and feeds back detection information to the edge server in real time through the edge node. The feedback detection information includes the equipment status information of the production line and the detection information of the workpiece on the production line, so as to detect the status of each process on the production line.

[0062] S105, when the central cloud detects that the computing load of any edge server is greater than a preset threshold, it is defined as a busy edge server, and searches for idle edge servers in the local area network; and sends second configuration information to the busy edge server and the idle edge server; the second configuration information includes program configuration information of the target algorithm file, and computing modules allowed to be used by the program in the edge server, and all allowed computing modules perform data transmission in the corresponding computing order according to the server address.

[0063] Specifically, when the number of production lines for a specific product category is increased in a factory, the edge servers responsible for processing the data of these production lines may be overloaded due to the sharp increase in computing power. In this case, if other edge servers are idle, it will cause a waste of computing resources.

[0064] In order to solve this problem, the present invention designs a collaborative working mechanism between edge servers. Under this mechanism, a stable communication connection is established between the edge servers so that information sharing and task scheduling can be quickly performed when needed. When the central cloud detects that the computing load of an edge server (i.e., a busy edge server) exceeds a preset threshold, the load balancing strategy is immediately started. The core of this strategy is to search other edge servers in the local area network to find those servers that are currently idle or have a low computing load (i.e., idle edge servers).

[0065] Once an idle edge server is found, the central cloud will send the second configuration information to the busy edge server and the idle edge server. This configuration information not only contains the program configuration information of the target algorithm file, but also clearly indicates the computing modules that the program in the edge server is allowed to use. Since the algorithm file has been divided into multiple independently run computing modules at the beginning of the design, these modules can be flexibly scheduled and executed on different edge servers. In other words, the idle edge server can take over part of the computing tasks from the busy edge server according to the instructions of the second configuration information, and perform the calculations in sequence according to a specific calculation order, such as Figure 3 As shown. Among them, whether the computing module is allowed to be used is determined according to the allocation model, that is, the computing module of the corresponding computing model is allocated to the idle edge server according to the allocation model to achieve load balancing; the allocation model includes: obtaining the computing difficulty index of the computing module: a comprehensive score of the number of parallel codes, the time required for model operation, and resource consumption. The performance indicators of each edge server: a comprehensive score of the number of CPU cores, memory size, processing speed, etc. Then, the computing module is allocated to the corresponding idle edge server based on the ratio of the difficulty index to the performance index plus the minimum load of the corresponding server as a constraint. The expression is:

[0066]

[0067] in, is the computing performance of the ith server, is the computational difficulty of the j-th model, is the current load of the i-th server, is whether the j-th model is assigned to the i-th server, k is the number of servers, and C is a safety factor to ensure that the server is not overloaded. The server index selected for model j.

[0068] This collaborative computing method not only significantly improves computing efficiency, but also effectively avoids idle edge servers. More importantly, it provides factories with the ability to respond quickly when production tasks suddenly increase. When certain products need to be produced urgently and production lines need to be added, the new production lines can be directly connected to the existing edge servers without the need to connect to new edge servers separately. This not only saves the time of connecting the edge servers to the production lines separately, but also improves the flexibility and efficiency of the entire production system.

[0069] In addition, if the computing load of all edge servers exceeds the preset threshold, the factory can also quickly add new edge servers based on this collaborative computing method and connect them to the original edge servers. In this way, the new edge servers can immediately participate in the computing tasks, thereby further alleviating the computing pressure and ensuring the smooth completion of production tasks.

[0070] Furthermore, when there is no edge server in a completely idle state in the local area network, the present invention adds an additional search condition, specifically, if the search result is 0, the edge server whose computing load is less than the preset second threshold in the local area network is searched as an idle edge server. That is, if no completely idle server is found, the system will immediately start a broader search strategy. The goal of this strategy is to find edge servers whose computing load is still lower than the preset second threshold although they are running computing tasks. Although these servers are not completely idle, they are still capable of accepting and processing additional computing tasks due to their low load, so they can also be used as idle servers for computing. Thereby improving the flexibility and efficiency of resource utilization.

[0071] Furthermore, after searching for edge servers whose number is less than the preset second threshold, if the search result is still 0, the central cloud configures the program of the target algorithm locally according to the target algorithm file, and sends third configuration information to the busy edge server, where the third configuration information includes a computing module allowed to be used by the program in the edge server, and the computing module is a computing module that is not allowed to be used in the central cloud; the central cloud obtains intermediate data from the busy edge server, and calculates it in sequence according to the corresponding calculation order, and the calculation results are fed back to the busy edge server.

[0072] That is to say, when there is neither a completely idle edge server nor a relatively idle server whose computing load is lower than the preset second threshold in the local area network, the central cloud will take on part of the computing task by directly configuring the program of the target algorithm and sending the third configuration information to the busy edge server to achieve effective scheduling and utilization of computing resources. The central cloud will select those computing modules that have not been used in the central cloud according to the actual load of the current busy edge server and the specific requirements of the target algorithm. These computing modules are then encapsulated in the third configuration information and sent to the busy edge server. The key to this step is that it allows the busy edge server to offload some computing tasks that were originally unable to be processed to the central cloud for processing while maintaining the existing computing tasks.

[0073] After receiving the intermediate data from the busy edge servers, the central cloud will process the data in sequence according to the calculation order specified in the algorithm file. Thanks to the powerful computing power and efficient resource management of the central cloud, it can quickly complete these computing tasks and feed back the results to the busy edge servers in a timely manner. In this way, even when there are no idle or relatively idle edge servers available, the system can still maintain efficient operation and ensure the smooth completion of production tasks.

[0074] S106, the idle edge server obtains the corresponding algorithm file from the central cloud according to the second configuration information, and configures the algorithm program; these algorithm files are necessary for executing the task, and they contain the instructions and logic required to complete the specific computing task. By obtaining the algorithm file from the central cloud, the idle edge server can ensure that it has all the resources required to perform the computing task. The idle edge server obtains the intermediate data from the busy edge server and calculates it in sequence according to the calculation order, so that the idle edge server can continue the previous calculation process, ensuring the continuity and integrity of the computing task. Finally, the calculation results are fed back to the busy edge server.

[0075] S107, the busy edge server controls the corresponding production line to perform corresponding operations through the edge node according to the result. The operations performed by the production line include but are not limited to the start, stop, and timing operations of each device (such as motors and robotic arms) of the production line.

[0076] In some embodiments of the present invention, when some idle edge servers are coordinating with busy edge servers, once the production line corresponding to the idle server needs to be started, the central cloud is used to take over its original computing work, such as Figure 4 As shown, the specific principles are as follows:

[0077] S201, when the central cloud receives new production parameters, the central cloud queries whether the corresponding edge server has been configured according to the product category of the production parameters; if not, the corresponding configuration information is sent.

[0078] Specifically, when the central cloud receives new production parameters from the production line or management system, these parameters usually contain key information such as the category, specifications and quantity of the products to be produced. The central cloud then starts a query process to check whether the corresponding edge server has been pre-configured with the necessary computing resources and software environment according to the product category. If the query result shows that the edge server has not been configured for this type of product, the central cloud will immediately generate and send the corresponding configuration information to the edge server. This information includes the required algorithm program files, computing resource allocation plans, and necessary computing module licenses, etc., to ensure that the edge server can quickly adapt to new production tasks.

[0079] S202: If it has been configured, query the configuration information of the edge server, and configure the corresponding algorithm program for the central cloud and authorize the corresponding computing module according to the configuration information.

[0080] Specifically, if the edge server has been configured for the product category, the central cloud will further query the specific configuration information of the server. Based on this information, the central cloud will perform two key operations: first, according to the hardware configuration and software environment of the edge server, the central cloud will configure the corresponding algorithm program to ensure that the algorithm can run efficiently; second, it will license the corresponding computing module; that is, if the central cloud wants to replace the role of the original edge server, it needs to license or activate the same computing module as the original computing module of the edge server.

[0081] S203, according to the transmission order of the data in the configuration information, query the upstream server and downstream server of the edge server, and send update information to the upstream server and downstream server to change the sending address of the upstream server to the central cloud, and change the receiving address of the downstream server to the central cloud. Among them, after replicating the configuration environment of the original edge server, the central cloud also needs to deal with the update of the data transmission path. According to the data transmission order specified in the configuration information, the central cloud will query and record the upstream server (responsible for sending data to it) and the downstream server (responsible for receiving its processed data) of the edge server. Subsequently, the central cloud sends update information to these upstream and downstream servers, instructing them to change the address for sending and receiving data from the original edge server to the central cloud. The continuity of the data flow is ensured, so that even if the edge server is taken over by the central cloud, the data on the production line can be seamlessly connected to ensure the smooth progress of the production process.

[0082] Furthermore, in the business operation process, in order to ensure the continuity of the production process and the optimal use of resources, when the central cloud receives a stop command, it sends a command to stop the material supply to the production server, cutting off the source of new materials entering the production line, and preventing waste or confusion that may be caused by continued supply of materials. The edge server also sends a delayed stop command to the edge node to ensure that the semi-finished products on the production line have enough time to complete the current process and smoothly transition to the next process until the processing is finally completed. This delayed stop strategy not only avoids the problem of semi-finished products being unable to continue processing due to process interruptions, but also minimizes waste in the production process and improves overall production efficiency and resource utilization.

[0083] Based on the same concept, Figure 5 As shown, the present invention provides an intelligent control system based on edge computing assistance, which is applied to an automated production line, including a central cloud, multiple edge servers and multiple edge nodes; the multiple edge servers are all connected to the central cloud for communication, each edge server is connected to multiple edge nodes for communication, and communication connections are established between the edge servers;

[0084] The system includes:

[0085] Production module 301, based on the input production parameters, the central cloud sends first configuration information to the corresponding edge server according to the product category of the production parameters; multiple algorithm files are pre-stored in the central cloud, and each algorithm file is divided into multiple independently run computing modules according to the calculation order;

[0086] The edge service module 302 is used for the edge server to obtain the corresponding target algorithm file from the central cloud according to the first configuration information and configure the algorithm program; the edge server sends a production request to the production server; the edge server pre-stores a plurality of control instructions for controlling the production line;

[0087] Load detection module 303, in response to the permission of the production request, the edge server starts the detection of the computing load and reports the load status data to the central cloud in real time; at the same time, the edge server sends a start instruction to the edge node corresponding to the product category;

[0088] The detection information acquisition module 304, in response to the start instruction, controls the corresponding number of production lines to start production, and feeds back the detection information to the edge server in real time through the edge node.

[0089] The allocation module 305 is used to define any edge server as a busy edge server when the central cloud detects that the computing load of any edge server is greater than a preset threshold, search for idle edge servers in the local area network; and send second configuration information to the busy edge server and the idle edge server; the second configuration information includes program configuration information of the target algorithm file, and computing modules allowed to be used by the program in the edge server, and all computing modules allowed to be used perform data transmission in a corresponding computing order according to the address of the server;

[0090] The calculation module 306 is used for the idle edge server to obtain the corresponding algorithm file from the central cloud according to the second configuration information and configure the algorithm program; the idle edge server obtains the intermediate data from the busy edge server, and calculates in sequence according to the calculation order, and feeds back the calculation results to the busy edge server;

[0091] The execution module 307 is used to enable the busy edge server to control the corresponding production line to perform corresponding operations through the edge node according to the result.

[0092] Based on the same concept, the present invention provides an electronic device, which includes a processor, a storage medium and a computer program, wherein the computer program is stored in the storage medium, and is characterized in that when the computer program is executed by the processor, an intelligent control method based on edge computing assistance is implemented.

[0093] Based on the same concept, the present invention provides a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, an intelligent control method based on edge computing assistance is implemented.

[0094] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An edge computing-assisted intelligent control method, applied to an automated production line, characterized in that: It includes a central cloud, multiple edge servers and multiple edge nodes; the multiple edge servers are all connected to the central cloud in communication, each edge server is connected to multiple edge nodes in communication, and communication connections are established between the edge servers; The method comprises: Based on the input production parameters, the central cloud sends first configuration information to the corresponding edge server according to the product category of the production parameters; the central cloud pre-stores a plurality of algorithm files, and each algorithm file is divided into a plurality of independently operable computing modules according to a computing order; The edge server obtains the corresponding target algorithm file from the central cloud according to the first configuration information, and performs algorithm program configuration; the edge server sends a production request to the production server; the edge server pre-stores a plurality of control instructions for controlling the production line; In response to the permission of the production request, the edge server starts computing load detection and reports load status data to the central cloud in real time; at the same time, the edge server sends a start instruction to the edge node corresponding to the product category; In response to the start instruction, the edge node controls a corresponding number of production lines to start production, and feeds back detection information to the edge server in real time through the edge node; When the central cloud detects that the computing load of any edge server is greater than a preset threshold, it is defined as a busy edge server, and searches for an idle edge server in the local area network; and sends second configuration information to the busy edge server and the idle edge server; the second configuration information includes program configuration information of the target algorithm file, and computing modules allowed to be used by the program in the edge server, and all computing modules allowed to be used perform data transmission in a corresponding computing order according to the address of the server; The idle edge server obtains the corresponding algorithm file from the central cloud according to the second configuration information, and performs algorithm program configuration; the idle edge server obtains the intermediate data from the busy edge server, and performs calculations in sequence according to the calculation order, and feeds back the calculation results to the busy edge server; The busy edge server controls the corresponding production line to perform corresponding operations through the edge node according to the result.

2. The intelligent control method according to claim 1, characterized in that: After searching for idle edge servers in the local area network, the method further includes: If the search result is 0, an edge server whose computing load is less than a preset second threshold value in the local area network is searched as an idle edge server.

3. The intelligent control method according to claim 2, characterized in that: After searching for an edge server in the local area network whose computing load is less than a preset second threshold, the method further includes: If the result of the search is 0, the central cloud configures the program of the target algorithm locally according to the target algorithm file, and sends third configuration information to the busy edge server, wherein the third configuration information includes a computing module that is allowed to be used by the program in the edge server, and the computing module is a computing module that is not allowed to be used in the central cloud; The central cloud obtains the intermediate data from the busy edge server, and calculates in sequence according to the corresponding calculation order, and feeds back the calculation results to the busy edge server.

4. The intelligent control method according to claim 1, characterized in that: Before sending the second configuration information to the busy edge server and the idle edge server, the method further includes: According to the allocation model, the computing module of the corresponding computing model is allocated to the idle edge server to achieve load balancing; the allocation model includes: obtaining the computing difficulty index of the computing module and the performance index of each edge server; and allocating the computing module to the corresponding idle edge server with the ratio of the difficulty index to the performance index plus the minimum corresponding server load as a constraint condition.

5. The intelligent control method according to claim 4, characterized in that: The distribution model expression is: in, is the computing performance of the ith server, is the computational difficulty of the jth module, is the current load of the i-th server, is whether the jth module is assigned to the ith server, k is the number of servers, and C is a safety factor to ensure that the server is not overloaded. The server index selected for module j, is the number of computing modules.

6. The intelligent control method according to claim 1, characterized in that: The method further comprises: When the central cloud receives new production parameters, the central cloud queries whether the corresponding edge server has been configured according to the product category of the production parameters; if not, the central cloud sends the corresponding configuration information to the edge server; If it has been configured, query the configuration information of the edge server, and configure the corresponding algorithm program for the central cloud and authorize the corresponding computing module according to the configuration information; According to the transmission order of the data in the configuration information, the upstream server and the downstream server of the edge server are queried, and update information is sent to the upstream server and the downstream server to change the sending address of the upstream server to the central cloud and the receiving address of the downstream server to the central cloud.

7. The intelligent control method according to claim 1, characterized in that: The method further comprises: When the central cloud receives the stop command, the central cloud sends a command to stop the material supply to the production server, and sends a delayed stop command to the edge node through the edge server.

8. An edge computing-assisted intelligent control system, applied to an automated production line, characterized in that: It includes a central cloud, multiple edge servers and multiple edge nodes; the multiple edge servers are all connected to the central cloud in communication, each edge server is connected to multiple edge nodes in communication, and communication connections are established between the edge servers; The system comprises: The production module, based on the input production parameters, the central cloud sends the first configuration information to the corresponding edge server according to the product category of the production parameters; the central cloud pre-stores a plurality of algorithm files, each algorithm file is divided into a plurality of independently operable computing modules according to the computing order; An edge service module, which is used for the edge server to obtain the corresponding target algorithm file from the central cloud according to the first configuration information and configure the algorithm program; the edge server sends a production request to the production server; the edge server pre-stores a plurality of control instructions for controlling the production line; A load detection module, in response to the permission of the production request, the edge server starts the detection of the computing load and reports the load status data to the central cloud in real time; at the same time, the edge server sends a start instruction to the edge node corresponding to the product category; The detection information acquisition module responds to the start instruction, and the edge node controls the corresponding number of production lines to start production, and feeds back the detection information to the edge server in real time through the edge node. The allocation module is used to define any edge server as a busy edge server when the central cloud detects that the computing load of any edge server is greater than a preset threshold, search for idle edge servers in the local area network; and send second configuration information to the busy edge server and the idle edge server; the second configuration information includes program configuration information of the target algorithm file, and computing modules allowed to be used by the program in the edge server, and all computing modules allowed to be used perform data transmission in a corresponding computing order according to the address of the server; A calculation module, configured for the idle edge server to obtain the corresponding algorithm file from the central cloud according to the second configuration information and to configure the algorithm program; the idle edge server obtains the intermediate data from the busy edge server, and calculates in sequence according to the calculation order, and feeds back the calculation results to the busy edge server; An execution module is used for the busy edge server to control the corresponding production line to perform corresponding operations through the edge node according to the result.

9. An electronic device comprising a processor, a storage medium and a computer program, wherein the computer program is stored in the storage medium, wherein: When the computer program is executed by a processor, the control method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the control method according to any one of claims 1 to 7 is implemented.

Images