CN117590801B - Cloud edge cooperative 5G edge control device - Google Patents
Cloud edge cooperative 5G edge control device Download PDFInfo
- Publication number
- CN117590801B CN117590801B CN202410080383.1A CN202410080383A CN117590801B CN 117590801 B CN117590801 B CN 117590801B CN 202410080383 A CN202410080383 A CN 202410080383A CN 117590801 B CN117590801 B CN 117590801B
- Authority
- CN
- China
- Prior art keywords
- protocol
- data
- model
- edge
- cloud
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004458 analytical method Methods 0.000 claims abstract description 71
- 238000004891 communication Methods 0.000 claims abstract description 35
- 238000007405 data analysis Methods 0.000 claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 28
- 239000002131 composite material Substances 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 35
- 238000012549 training Methods 0.000 claims description 28
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000013500 data storage Methods 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 7
- 238000003064 k means clustering Methods 0.000 claims description 6
- 238000013480 data collection Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 abstract description 2
- 238000012545 processing Methods 0.000 description 17
- 238000003860 storage Methods 0.000 description 16
- 238000007726 management method Methods 0.000 description 12
- 238000004519 manufacturing process Methods 0.000 description 11
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000000737 periodic effect Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000001343 mnemonic effect Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/05—Programmable logic controllers, e.g. simulating logic interconnections of signals according to ladder diagrams or function charts
- G05B19/056—Programming the PLC
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/10—Plc systems
- G05B2219/13—Plc programming
- G05B2219/13004—Programming the plc
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Communication Control (AREA)
Abstract
The cloud edge cooperative 5G edge control device comprises a hardware part and a software part; the hardware part comprises a core processor and a 5G communication module; the software part comprises a virtual machine monitor, an operating system, a data analysis module and an edge control module; the method comprises the steps that a virtual machine monitor creates two virtual resource spaces, linux and Windows operating systems are respectively installed, a data analysis module runs in the Windows operating systems, analysis based on a protocol model is carried out on machine equipment data acquired by data acquisition equipment, and acquired data are acquired and stored; the edge control module operates in a Linux operating system, a control model of the machine equipment and corresponding composite function block codes established by the cloud layer are packaged and mapped to a Docker container, working parameters related to the working state of the machine equipment are obtained, the machine equipment is synchronized to the cloud layer, the control model parameters issued by the cloud layer are received, and a control algorithm is updated at the edge layer to realize cloud edge coordination.
Description
Technical Field
The invention belongs to the technical field of industrial automatic control, and relates to a cloud-edge cooperative 5G edge control device which is used for carrying out data acquisition and edge processing on equipment in an industrial field and realizing control on the equipment.
Background
Because the terminal equipment accessed by the Internet of things is complex and heterogeneous in network, when an intelligent factory or industrial Internet of things system is built, a data acquisition gateway is usually required to be installed in the south of a programmable controller so as to solve the problems of multi-source heterogeneous equipment protocol analysis and data access. Meanwhile, in the face of the requirements of computing pressure and privacy protection brought by mass data access, an edge computing device is generally configured in the north direction of a programmable controller, and the edge computing device mostly uses an industrial computer to monitor data configuration and upload the data configuration to a cloud layer or a corresponding local server.
Obviously, the industrial edge control and calculation functions realized based on the above method have the following problems:
(1) Through serial connection and data interaction between the gateway and the programmable controller and between the gateway and the edge computing device, the structure is complex, the deployment is inconvenient, and the cost is high.
(2) Because different gateways adapt different protocols and communication modes are different, data acquisition and protocol analysis are inflexible, access capability is limited, and communication transmission delay is large.
(3) The control system based on the traditional programmable controller has high software and hardware coupling, the interface is closed, the program adjustment or optimization is relatively difficult, and once the production of the controller of the equipment is stopped or replaced after long-time operation, the operation and maintenance of the equipment are difficult to carry out.
(4) The traditional configuration monitoring function has the defects of less edge data analysis capability and insufficient data mining depth, can not directly guide actual production work, and is seriously dependent on personal experience of field engineers.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a cloud-edge cooperative 5G edge control device so as to improve the flexibility and expansibility of deployment and reduce the delay of data transmission; and further expands the data acquisition capacity and the data calculation and analysis capacity at the edge layer, and ensures the convenience and economy of later operation and maintenance.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a cloud-edge cooperative 5G edge control device comprises a hardware part and a software part; the hardware part comprises a core processor and at least one 5G communication module; the software part comprises a virtual machine monitor, an operating system, a data analysis module and an edge control module;
the virtual machine monitor is deployed on the hardware part, and two virtual resource spaces are created;
the operating systems are a Linux operating system and a Windows operating system, and are respectively installed in the two virtual resource spaces; the data analysis module operates in the Windows operating system, and the edge control module operates in the Linux operating system;
The data analysis module is in communication connection with the data acquisition equipment deployed at the end layer, analyzes the machine equipment data acquired by the data acquisition equipment based on a protocol model, acquires the acquired data and stores the acquired data;
the edge control module packages and maps a control model of the machine equipment established by the cloud layer and a corresponding composite function block code to a Docker container; the edge control module is in communication connection with the machine equipment deployed at the end layer, acquires working parameters related to the working state of the machine equipment, synchronizes to the cloud layer, receives control model parameters issued by the cloud layer to the edge layer, updates a control algorithm at the edge layer, and realizes cloud edge coordination.
In one embodiment, the data acquisition device comprises a sensor, a camera and a meter, wherein the sensor acquires working parameter data when the machine device is in operation, the camera acquires operation field video data of the machine device, and the meter acquires parameter data related to control and optimization of the operation process of the machine device.
In one embodiment, the data analysis module includes a data storage unit and a protocol model-based analysis unit, where the protocol model-based analysis unit performs protocol model-based analysis on the machine device data collected by the data collection device, and the method includes:
Step 1, defining a protocol model f capable of characterizing a plurality of different protocols, and establishing an initial protocol model f by using a known protocol 0 The structure of the protocol model f is characterized in sequence as follows:
f=(L,N,L 1 ,L 2 ,…,L N ,T 1 ,T 2 ,…,T N ,D,P)；
wherein L represents the length of a data frame of a certain protocol except for data bits, N represents the number of protocol features of the protocol, L 1 Representing a first protocolLength of the features, L 2 Representing the second protocol characteristic length, L N Representing the nth protocol feature length, l=l 1 +L 2 +…+L N ，T 1 Representing the physical attribute corresponding to the first protocol feature, T 2 Representing the physical attribute corresponding to the second protocol feature, T N Representing a physical attribute corresponding to the Nth protocol feature, D represents data content of data bits in a data frame of the protocol, and P represents a protocol type of the protocol;
step 2, performing model training to obtain a trained protocol model f x Will initiate the protocol model f 0 And trained protocol model f x Combining and de-duplicating the same characteristics to obtain a multi-protocol analysis model fA; wherein the initial protocol model f 0 And a trained protocol model f x The multi-protocol analysis model fA is stored in the cloud layer server and the data storage unit;
step 3, inputting the protocol type P of the current data packet x Search matching with the protocol type in the multi-protocol parsing model fA to determine the protocol type P x Whether the analysis model is in the characterization range of the current multi-protocol analysis model fA or not; if the search matching is successful, entering a step 4, otherwise, entering a step 2 to perform model training and updating a multi-protocol analysis model fA;
step 4, determining the protocol type P of which the search matching is successful x Protocol characteristics of (2);
step 5, generating an XML format analysis file according to the data to be acquired and the parameter configuration thereof, thereby obtaining a trained protocol model f x Data content D of data bits in data frames of medium protocol x ；
And step 6, downloading the analysis file into the memory of the cloud-edge cooperative 5G edge control device, and sending data to a cloud layer.
In one embodiment, the initial protocol model f 0 The selected protocol type is Modbus TCP protocol, and the structure is characterized as follows in sequence:
f 0 =(L 0 ,N 0 ,L 0 1 ,L 0 2 ,L 0 3 ,T 0 1 ,T 0 2 ,T 0 3 ,D 0 ,P 0 )；
wherein P is 0 Representing the initial protocol type, i.e. Modbus TCP protocol, L 0 Representing the length, N, of the data frame of the initial protocol, excluding the data bits 0 Number of protocol characteristics, N, representing initial protocol 0 =3, first protocol characteristic length L 0 1 =7, second protocol characteristic length L 0 2 =16, third protocol characteristic length L 0 3 According to the change of the length of the transmitted data, the physical attribute T corresponding to the first protocol characteristic 0 1 Representing the physical attribute T corresponding to the message header and the second protocol feature 0 2 Representing the feature code, and the physical attribute T corresponding to the third protocol feature 0 3 Representing data bits, D 0 Representing the data content of the data bits in the data frame of the initial protocol.
In one embodiment, the method for obtaining the multi-protocol parsing model fA in the step 2 is as follows:
step 2.1, classifying each data frame by using cloud layer computing resources and using a K-means clustering algorithm, thereby obtaining a trained protocol model f x ；
Step 2.2, using the initial protocol model f 0 And trained protocol model f x Updating the multi-protocol parsing model fA.
In one embodiment, the training method in step 2.1 is as follows:
step 2.1.1, preparing a data set and dividing a training set and a verification set, wherein the data content D of data bits in each data frame of the data set is known;
step 2.1.2, setting the initial clustering center number of the K-means clustering algorithm to be from 3 to 20 for each data frame of the training set, wherein each cluster is a protocol characteristic; finding out the cluster containing the most data content D of the data bits in the data frame, and calculating the percentage R in the cluster, wherein q is the data length of the data content D of the data bits in the data frame, and p is the total data length of the data content D of the data bits in the data frame; sequentially calculating each percentage R in the training set, and averaging Ra;
Step 2.1.3, sequentially calculating average value Ra corresponding to the cluster center number from 3 to 20, and determining the corresponding cluster number n when the average value Ra is maximum as the optimal cluster center number;
step 2.1.4: comparing the values corresponding to the n clustering centers to give physical attributes represented by each cluster, and respectively marking the physical attributes as T 1 ~T n ；
Step 2.1.5, comparing the corresponding clustering results, and determining a protocol model, namely a trained protocol model f, of the unknown protocol for access x The expression is as follows:
f x =(L x ,N x ,L x 1 ,L x 2 ,…,L x n ,T x 1 ,T x 2 ,…,T x n ,D x ,P x )；
wherein the number of protocol features N of the unknown protocol x =n，L x 1 ,L x 2 ,…,L x n For trained protocol model f x The first to nth protocol characteristic length of the unknown protocol, the length L of the data frame excluding the data bit x =L x 1 +L x 2 +…+L x n ，T x 1 ,T x 2 ,…,T x n Representing a trained protocol model f x Physical attributes corresponding to the first through nth protocol characteristics, D x Representing the data content of the data bits in the data frame of the unknown protocol, P x Representing the protocol type of the unknown protocol.
In one embodiment, in step 2.2, the method for updating the multi-protocol parsing model fA is as follows:
step 2.2.1, merging the trained protocol models f x And the multi-protocol analysis model fA is combined and updated, and the expression is as follows:
fA=(L 0 ,L x ,N 0 ,N x ,L 0 1 ,L 0 2 ,L 0 3 ,L x 1 ,L x 2 ,…,L x n ,T 0 1 ,T 0 2 ,T 0 3 ,T x 1 ,T x 2 ,…,T x n ,D 0 ,D x ,P 0 ,P x )；
step 2.2.2: removing duplication of elements which have the same data length and represent the same physical attribute in the multi-protocol analysis model fA;
Step 2.2.3: updating a protocol matrix a (P x ) Wherein the initial protocol matrix a (P 0 )=[1,1,1]According to the updated protocol matrix A (P x ) The following are provided:
A(P x )=[0,1,0,1,0,0,0,1,1,1,…,1,0,0,0, ,…,1,0,1,0,1]；
A(P x ) Corresponding to fA in length, updated multi-protocol analysis model fA and each protocol matrix A (P x ) Synchronously downloading to the cloud-edge cooperative 5G edge control device.
In one embodiment, the step 4 is directed to retrieving the successfully matched protocol type P x Corresponding protocol matrix a (P x ) Comparing the corresponding bit of the multi-protocol analysis model fA with the corresponding bit of 1, and deleting the corresponding bit of 0, thereby determining the protocol characteristics, namely L x 1 ,L x 2 ,…,L x n ,T x 1 ,T x 2 ,…,T x n 。
In one embodiment, the data analysis module comprises a custom billboard unit which is embedded in the form of sub-controls, wherein the sub-controls comprise variable controls, bit buttons, word buttons, bar charts, real-time curves, history curves and dashboards, and the sub-controls are arranged on a canvas in a dragging mode to realize a custom monitoring picture.
In one embodiment, the data analysis module and the cloud layer are communicated by adopting an MQTT protocol, and the edge control module and the cloud layer are communicated by adopting an OPC UA protocol; the communication interfaces are 4, and respectively support Ethernet, wiFi, 5G and LoRa communication.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the advantages of cloud computing and edge computing are utilized, a cloud-edge cooperative 5G edge control device is designed, a general protocol analysis algorithm is integrated in a data analysis module deployed at an edge layer, machine equipment data acquired by data acquisition equipment can be analyzed based on a protocol model, large-scale heterogeneous machine equipment data can be accessed conveniently, the expansion can be carried out according to field application, and the portability and reusability of an edge control function are high. The device can operate in cloud edge cooperative mode to realize cooperation with cloud layer servers, and can also operate autonomously and independently.
Drawings
Fig. 1 is a schematic structural diagram of a cloud-edge cooperative 5G edge control device according to the present invention.
Fig. 2 is a schematic diagram of a hardware portion of the cloud edge cooperative 5G edge control device according to the present invention.
Fig. 3 is a schematic diagram of a software part of the cloud edge cooperative 5G edge control device according to the present invention.
Fig. 4 is a flow chart of a protocol model-based parsing method of the present invention.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples.
The functions of the 5G edge control device are specifically described through the overall structure, the hardware part, the software part and the operation mode of the cloud edge cooperative 5G edge control device.
1. The overall structure of the device.
The cloud-edge cooperative 5G edge control device comprises a hardware part and a software part, wherein the main composition structure is shown in figure 1, the hardware part is a bottom layer part of the device and is provided with various input and output interfaces, communication ports, a core processor, a power module and the like, and in order to realize the 5G function of the invention, the hardware part comprises at least one 5G communication module. The software part mainly comprises a virtual machine monitor, an operating system, a data analysis module and an edge control module.
The virtual machine monitor is deployed on the hardware portion for implementing virtualization on the hardware portion, creating two independent virtual resource spaces for installing the operating system. In the present invention, a real-time virtual machine monitor, typically Hyper-V virtualization software, is employed.
The operating system is a Linux operating system and a Windows operating system, and two virtual resource spaces which are built in a virtual machine monitor are respectively deployed and installed. Through the virtual function of the virtual machine monitor, the two operating systems run independently in parallel, and the stability and reliability of the running of the software functions on the two operating systems are ensured.
The data analysis module runs in a Windows operating system in the form of an application program, is in communication connection with the data acquisition equipment deployed at the end layer, analyzes the machine equipment data acquired by the data acquisition equipment based on a protocol model, acquires the acquired data and stores the acquired data.
The edge control module runs in a Linux operating system, and a control model of the machine equipment established by the cloud layer and corresponding composite function block codes are packed and mapped to a Docker container; the edge control module is in communication connection with the machine equipment deployed at the end layer, acquires working parameters related to the working state of the machine equipment, synchronizes to the cloud layer, receives control model parameters issued by the cloud layer to the edge layer, updates a control algorithm at the edge layer, and realizes cloud edge coordination.
The Docker container is an open-source application container engine, so that a developer can pack the application of the application in a unified mode, rely on the package to a portable container and then issue the package to any server provided with the Docker engine, and virtualization can be realized.
The control model of the machine equipment can be specifically built in a cloud layer server.
In the embodiment of the invention, the data acquisition equipment mainly comprises a sensor, a camera and an instrument, wherein the sensor is mainly used for acquiring working parameter data when the machine equipment operates, the camera is mainly used for acquiring operation field video data of the machine equipment, and the instrument is mainly used for acquiring parameter data related to control and optimization of the operation process of the machine equipment, such as flow, speed, rotating speed and the like.
2. A hardware part.
The hardware part of the invention is shown with reference to fig. 2, and mainly comprises a power module, a CPU module, a storage module, an input interface, an output interface and a communication interface.
The power supply module is connected with an external power supply and is used for realizing power supply of the cloud-edge cooperative 5G edge control device, and a direct-current 6-24V wide voltage module can be selected.
The CPU module is used for executing calculation and processing, specifically, analyzing and processing data of the machine equipment data acquired by the data acquisition equipment and the working parameters related to the working states of the machine equipment, and processing control model parameters, control models and programs issued by the cloud layer. The CPU module, namely the core processor, specifically selects a dual-core 64-bit Cortex-A55 processing module and selects a domestic RK3568J chip.
The storage module is used for storing machine equipment data acquired from all data acquisition equipment of the terminal layer and working parameters related to working states of the machine equipment, and in the embodiment of the invention, an embedded memory is selected to support 8G/16G and 64G/128G, and the expansion is supported.
The input interface is used for realizing parameter access of the data acquisition equipment and the machine equipment; the output interface is used for outputting a switching signal to control the start and stop of the machine equipment or outputting an analog signal to change the corresponding running state of the machine equipment. The input interface and the output interface both support relays and optocoupler isolation type IO, and the IO can be digital and analog. Typical input modules are 4 AI and 4 DI inputs and output modules are 2 AO and 4 DO outputs. Here AI, AO, DI, DO represents analog input, analog output, digital input, and digital output, respectively.
The communication interface is used for communication connection between the machine equipment and the data acquisition equipment. In the embodiment of the invention, 4 communication interfaces are arranged, the networking mode is flexible, wherein the first communication interface is an Ethernet communication interface, supports TCP/IP protocol and is provided with double RJ45 network ports. And the second communication interface is a WiFi module and supports 2.4GHz WiFi. The communication interface III is a 5G module, and a 4G module can be selected according to the requirement. And the fourth communication interface is a LoRa module. Through the 4 communication interfaces, the invention can support Ethernet, wiFi, 5G and LoRa communication. The communication modes of different data acquisition devices and machine devices can be different, and can be freely selected according to actual conditions.
In an embodiment of the present invention, the data analysis module analyzes the data collected by the data collection device to obtain parameters of instruments and machine devices with different communication protocols, and the present invention provides a general protocol analysis method, and further provides a function of customizing a billboard.
Wherein Modbus is a serial communication protocol, and Modbus TCP and Modbus RTU respectively represent Modbus Ethernet protocol and Modbus serial port protocol. OPC is OLE (OLE for Process Control) for process control, OLE refers to the object connect and embed (Object Linking and Embedding) technique. In order to facilitate the mutual exchange of data between equipment and application programs of different manufacturers in the automation industry, a unified interface function, namely OPC protocol specification, is defined, and is a common protocol in the industrial automation field.
3. Software part.
The software part of the cloud-edge cooperative 5G edge control device mainly comprises a virtual machine monitor, an operating system, a data analysis module and an edge control module. The data analysis module runs in a Windows operating system, and the edge control module runs in a Linux operating system.
The main architecture of the software part is shown in fig. 3, and the data analysis module is connected with data acquisition equipment such as sensors, cameras, meters and the like of the end layer and main machine equipment, acquires data of various machine equipment and carries out corresponding processing. The edge control module is in communication connection with the machine equipment of the terminal layer, acquires working parameters related to the working state of the machine equipment, and can realize data interaction between the module and other modules or third party equipment in a database mode. The data analysis module is communicated with the cloud layer by adopting an MQTT protocol, and the data after the data analysis is interacted with the cloud layer server through the MQTT protocol; the edge control module communicates with the cloud layer by adopting an OPC UA protocol, and communicates with the cloud layer server by the OPC UA protocol.
The data analysis module mainly realizes the functions of analysis based on a protocol model, data storage, customized signboard, abnormal alarm, multidimensional and multi-scale analysis and the like in sequence. Correspondingly, the system mainly comprises an analysis unit based on a protocol model, a data storage unit, a customized billboard unit, an abnormality alarm unit and a multidimensional multi-scale analysis unit.
The invention provides a general analysis method based on a protocol model, which can realize the analysis of communication protocols such as Modbus, profibus, CC-link, etherCAT and the like, has a plurality of protocol models of known protocols, and can realize the analysis of unknown protocols. Specifically, the method shields a hardware bottom layer protocol, and the bottom layer only relates to data forwarding, so that the method is an XML-based heterogeneous network protocol analysis method. The invention uses the analysis unit based on the protocol model to analyze the data collected by the data collection equipment based on the protocol model, as shown in figure 4, the specific flow is as follows:
step 1, defining a protocol model f capable of representing a plurality of different protocols based on data frames, and establishing an initial protocol model f by using a known protocol 0 The structure of the protocol model f is characterized in order as follows:
f=(L,N,L 1 ,L 2 ,…,L N ,T 1 ,T 2 ,…,T N ,D,P)；
wherein L represents the length of a data frame of a certain protocol except for data bits, N represents the number of protocol features of the protocol, L 1 Representing the first protocol feature length, L 2 Representing the second protocol characteristic length, L N Representing the nth protocol feature length, l=l 1 +L 2 +…+L N ，T 1 Representing the correspondence of the first protocol featurePhysical properties, T 2 Representing the physical attribute corresponding to the second protocol feature, T N And the physical attribute corresponding to the Nth protocol characteristic is represented, D represents the data content of data bits in a data frame of the protocol, and P represents the protocol type of the protocol.
In an embodiment of the invention, the initial protocol model f 0 The selected protocol type is Modbus TCP protocol, and the structure is characterized as follows in sequence:
f 0 =(L 0 ,N 0 ,L 0 1 ,L 0 2 ,L 0 3 ,T 0 1 ,T 0 2 ,T 0 3 ,D 0 ,P 0 )；
wherein P is 0 Representing the type of the initial protocol, in this embodiment, modbus TCP protocol, L 0 Representing the length, N, of the data frame of the initial protocol, excluding the data bits 0 Number of protocol characteristics representing initial protocol, N for Modbus TCP protocol 0 =3, first protocol characteristic length L 0 1 =7, second protocol characteristic length L 0 2 =16, third protocol characteristic length L 0 3 According to the change of the length of the transmitted data, the physical attribute T corresponding to the first protocol characteristic 0 1 Representing the physical attribute T corresponding to the message header and the second protocol feature 0 2 Representing the feature code, and the physical attribute T corresponding to the third protocol feature 0 3 Representing data bits, D 0 Representing the data content of the data bits in the data frame of the initial protocol. Specifically, the Modbus TCP protocol model is characterized by having 27 bytes of fixed length protocol feature information, including a 7 byte header and a 16 byte feature code.
Step 2, training the model to obtain a trained protocol model f x Will initiate the protocol model f 0 And trained protocol model f x Combining and de-duplicating the same characteristics to obtain a multi-protocol analysis model fA, wherein the multi-protocol analysis model fA is initially f 0 Wherein the initial protocol model f 0 And a trained protocol model f x Stored inThe cloud layer server, the multi-protocol analysis model fA is stored in a cloud-edge cooperative 5G edge control device, and can be specifically stored in the data storage unit, such as a hard disk.
In the embodiment of the invention, the method for obtaining the multi-protocol analysis model fA is as follows:
step 2.1, fully utilizing the calculation resources of cloud layers, and classifying each data frame by using a K-means clustering algorithm so as to obtain a trained protocol model f x The specific training method is as follows:
step 2.1.1, preparing a data set and dividing a training set and a validation set, wherein the data content D of data bits in each data frame of the data set is known.
Specifically, a data set of at least 15 minutes is prepared, wherein the first 10 minutes of data are used as a training set, the rest 5 minutes of data are used as a verification set, and each data comprises a data frame.
Step 2.1.2, setting an initial cluster center number of a K-means clustering algorithm for each data frame of the training set, wherein each cluster is a protocol characteristic; finding out the cluster containing the most data content D of the data bits in the data frame, and calculating the percentage R in the cluster, wherein q is the data length of the data content D of the data bits in the data frame, and p is the total data length of the data content D of the data bits in the data frame; and (5) sequentially calculating each percentage R in the training set, and averaging Ra. In this embodiment, the number of initialized cluster centers is set from 3 to an empirical value of 20.
And 2.1.3, sequentially calculating average value Ra corresponding to the cluster center numbers from 3 to 20, and determining the corresponding cluster number n when the average value Ra is maximum as the optimal cluster center number.
Step 2.1.4: combining physical meanings of the training set and common communication protocol features, comparing the numerical values corresponding to the n clustering centers, giving out physical attributes represented by each cluster, and respectively marking as T 1 ~T n ；
Step 2.1.5, comparing the corresponding clustering results, and determining a protocol model, namely a trained protocol model f, of the unknown protocol for access x The expression is as follows:
f 1 =(L 1 ,N 1 ,L 1 1 ,L 1 2 ,…,L 1 n ,T 1 1 ,T 1 2 ,…,T 1 n ,D 1 ,P 1 )；
f x =(L x ,N x ,L x 1 ,L x 2 ,…,L x n ,T x 1 ,T x 2 ,…,T x n ,D x ,P x )；
wherein f 1 Is a protocol model after the first training, N 1 Is f 1 Number of protocol features, L, of medium protocol 1 1 ,L 1 2 ,…,L 1 n For the first trained protocol model f 1 First through nth protocol feature lengths, f 1 Length L of data frame of medium protocol except data bit 1 =L 1 1 +L 1 2 +…+L 1 n ，T 1 1 ,T 1 2 ,…,T 1 n Representing the protocol model f after the first training 1 Physical attributes corresponding to the first through nth protocol characteristics, D 1 Represents f 1 Data content, P, of data bits in data frames of a medium protocol x Represents f 1 Protocol type of the medium protocol.
After training the protocol model f x In the number N of protocol characteristics of the unknown protocol of the access x =n，L x 1 ,L x 2 ,…,L x n For trained protocol model f x The first to nth protocol characteristic length of the unknown protocol, the length L of the data frame excluding the data bit x =L x 1 +L x 2 +…+L x n I.e. the value of the training set data length minus the data bit data length. T (T) x 1 ,T x 2 ,…,T x n Representing the first to nth protocol features of the trained protocol model fxPhysical properties, D x Representing the data content of the data bits in the data frame of the unknown protocol, P x Representing the protocol type of the unknown protocol.
Step 2.2, training the protocol model f x Stored in cloud layer and then using initial protocol model f 0 And trained protocol model f x Updating the multi-protocol analysis model fA, wherein the specific method is as follows:
step 2.2.1, merging the trained protocol models f x And the multi-protocol analysis model fA is updated through combination.
According to the foregoing, the multi-protocol analytical model fA is initially f 0 The initial protocol model f is then 0 And trained protocol model f x Merging, expressed as follows:
fA=(L 0 ,L x ,N 0 ,N x ,L 0 1 ,L 0 2 ,L 0 3 ,L x 1 ,L x 2 ,…,L x n ,T 0 1 ,T 0 2 ,T 0 3 ,T x 1 ,T x 2 ,…,T x n ,D 0 ,D x ,P 0 ,P x )；
step 2.2.2: and removing duplicate elements which have the same data length and represent the same physical attribute in the multi-protocol analytic model fA, and deleting duplicate items.
For example, e.g. L 0 2 =L x 3 And T is 0 2 And T x 3 Characterizing the same physical properties, the multi-protocol analytical model fA after deduplication of the duplicate term is expressed as follows:
fA=(L 0 ,L x ,N 0 ,N x ,L 0 1 ,L 0 2 ,L 0 3 ,L x 1 ,L x 2 ,L x 4 ,…,L x n ,T 0 1 ,T 0 2 ,T 0 3 ,T x 1 ,T x 2 ,T x 4 …,T x n ,D 0 ,D x ,P 0 ,P x )；
L x 4 for trained protocol model f x Is the fourth protocol feature length, T x 4 And representing the physical attribute corresponding to the fourth protocol characteristic of the trained protocol model fx.
Step 2.2.3: updating a protocol matrix a (P x ) Wherein the initial protocol matrix a (P 0 )=[1,1,1]According to the updated protocol matrix A (P x ) The following are provided:
A(P 0 )=[1,0,1,0,1,1,1,0,0,0,…,0,1,1,1,…,0,1,0,1,0]；
A(P x )=[0,1,0,1,0,0,0,1,1,1,…,1,0,0,0, ,…,1,0,1,0,1]；
A(P x ) Corresponding to fA in length, updated multi-protocol analysis model fA and each protocol matrix A (P x ) Synchronously downloading the cloud-edge-coordinated 5G edge control devices. For example, for initializing protocol model f 0 ，A(P 0 ) 1,3,5,6,7 bit is 1, the corresponding bit corresponding to fA is reserved; a (P) 0 ) Bit 2,4,8,9,10 is 0, the corresponding bit corresponding to fA is deleted, and the remaining bits are similar.
When new unknown protocols are continuously accessed, the prior art tends to obtain the characteristics of each protocol and the data content D of the data bits in the data frames of the unknown protocol by continuously trying to find out x The present embodiment uses cloud layer elastic resource training. However, there are two problems with cloud training models: one problem is that the number of protocol models is large, and if all edge layers exist, the data storage amount is large; another problem is that the protocol features in each protocol have a large number of identical features, not only occupying storage, but also affecting retrieval. Therefore, in this embodiment, different protocol models are formed through cloud layer elastic resource training; merging different protocol models, deleting repeated characteristics, compressing to form a multi-protocol analysis model fA, storing in an edge control device with limited computing resources, and recovering corresponding protocol analysis by a protocol matrix when analyzing different protocols And the protocol model is used for realizing analysis nuclear data acquisition.
Step 3, inputting the protocol type P of the current data packet x Search matching with the protocol type in the multi-protocol parsing model fA to determine the protocol type P x Whether the analysis model is in the characterization range of the current multi-protocol analysis model fA or not; if the search matching is successful, the step 4 is entered, otherwise, the step 2 is entered to perform model training and update the multi-protocol analysis model fA.
Step 4, determining the protocol type P of which the search matching is successful x Is a protocol feature of (a).
Protocol type P for successful search matching x Corresponding protocol matrix a (P x ) By contrast with the reservation of 1 and the deletion of 0 of the corresponding bit in the multi-protocol analysis model fA, the protocol model can be quickly obtained, thereby determining the protocol characteristics, namely L x 1 ,L x 2 ,…,L x n ,T x 1 ,T x 2 ,…,T x n 。
Step 5, generating an XML format analysis file according to the data to be acquired and the parameter configuration thereof, thereby obtaining a trained protocol model f x Data content D of data bits in data frames of medium protocol x . In the present invention, data content D x The related data parameters comprise variable ID, variable name, variable address, data type, data storage, readable and writable, and the like, and the specific description of each parameter is as follows:
variable ID: each variable is automatically numbered, defaulting from 0 to 500, preferably with an upper variable limit of 500.
Variable name: custom, supporting Chinese, numbers, case letters, etc., is a mnemonic.
Variable address: the unique identifier, which is a variable, consists of an address type and an address variant.
Data type: support Bool, int16, uint16, int32, uint32, float, long, double, bcd, bcd32, etc. alternatives.
Variable storage: the method is divided into periodic storage and non-periodic storage. Typically periodic storage, may be set to 1 minute, 5 minutes, 10 minutes, 30 minutes, or 60 minutes. Default to periodic storage, 10 minutes of storage period.
Readable and writable: the three states of read-only, write-only and read-write are read-only by default.
And 6, downloading the generated XML format analysis file into a memory of the cloud-edge collaborative 5G edge control device, and sending data to the cloud layer through the MQTT in the XML description form. Further, the data may also be stored in the local control device. Step 5 and step 6 are generally completed through the operation of the engineer station, and the analysis file in the XML format is also uploaded to the cloud layer and can be downloaded, so that other engineer stations can conveniently adjust or reconfigure the corresponding configuration file. So far, the protocol analysis flow is ended.
According to the method, a protocol model capable of representing various protocols is established through analysis, and the data storage in the process adopts remote dictionary service storage, saves time sequence data and stores the time sequence data in a cloud layer server. The data analysis module obtains data information containing semantic features after processing, the data information is processed and stored in a local database by a data storage unit, the database can be stored according to a preset storage rule, the storage rule can be dynamically adjusted, and the storage period is 60 seconds in a default state.
Custom signs refer to the user's ability to perform the presentation of personalized data analysis in a simple manner similar to "building blocks" without having to write the underlying code program of the computer. The custom billboard unit of the invention can be used as a sub-control function to be embedded into the data analysis module. The application function of the part is rich, the child controls comprise variable controls, bit buttons, word buttons, a histogram, a real-time curve, a history curve, an instrument panel and the like, and the child controls are arranged on a canvas in a dragging mode to realize customized monitoring pictures, so that the key information can be conveniently displayed.
The abnormal alarming function is to pre-alarm and prompt the abnormal condition of the collected data. The abnormal alarm unit supports instantaneous alarm and history early warning. The instantaneous alarm is to set a normal fluctuation threshold range for a certain data point, typically set to 20%, and if the fluctuation range of the acquired data exceeds the normal fluctuation threshold range in the next sampling period (usually 1 second or 10 seconds, which is adjustable), the instantaneous alarm is triggered, and the software outputs alarm information to the outside and synchronously records the alarm information into a log. Historical early warning is to analyze historical data of each parameter in a multi-time scale mode, and early warning is typically carried out according to years, months and days. Taking the day as a time scale example, taking an average value Va of a more detailed time scale (hours) of a last time unit (yesterday) as a reference, if the average value V of the more detailed time scale (hours) in the last time unit (yesterday) deviates from Va by a certain percentage (typically 20%), triggering early warning, and outputting alarm information to the outside and synchronously recording the alarm information in a log. The history early warning method for month and year is similar.
The data analysis function performs quantitative analysis of different time scales based on historical data and real-time data, and provides a quantitative basis for production decisions. The multidimensional and multiscale analysis unit supports analysis based on multiple dimensions such as time scale, alarm type, frequency, fluctuation deviation and the like. The unit comprehensively considers a plurality of factors and provides more comprehensive data analysis and decision basis in the industrial production field.
In a multi-dimensional multi-scale analysis unit, the time scale is an important dimension. By analysis of the historical data and the real-time data, time can be divided into different scales, such as minutes, hours, days, weeks, months, etc. This allows data at different time scales to be compared and analyzed to reveal time-dependent patterns and trends. When analyzing key machine equipment, time can be divided into two scales of hour level and day level. At the hour scale, the operating state of the machine device at different hours, such as the start-up time and the operating time of the machine device, can be analyzed. At the daily scale, the operating conditions of the machine equipment on different days, such as the yield of the machine equipment, the number of faults, etc., can be analyzed.
Alarm type and frequency are also one of the key dimensions in multidimensional, multi-scale analysis. Different types of alarms may have different impact on production, and analysis of alarm type and frequency may help determine which problems are serious and need to be addressed immediately. In addition, the stability and the reliability of the production system can be evaluated according to the alarm frequency. For example, analyzing different types of alarms generated by a machine device, such as excessive temperature, pressure anomalies, excessive vibration, etc., counts the frequency of each alarm type, i.e., the number of occurrences per day or hour. This allows to know which alarm types occur more often, requiring special attention.
Fluctuation deviation refers to the degree of fluctuation and deviation of the production data at different time scales. By analyzing the fluctuation deviation, the change and the abnormal situation in the production process can be known. This helps to quickly find potential problems and take appropriate action to make adjustments and improvements. For example, the operating data of the machine may be subjected to a wave analysis, such as a change in parameters of the machine, such as temperature, pressure, vibration, etc. By calculating indexes such as mean value, standard deviation, range and the like, the running stability and the change degree of the equipment can be evaluated. If fluctuations in a certain parameter are found to exceed a certain threshold value, this may mean that there is an abnormal situation in the machine equipment.
Through the multidimensional and multi-scale analysis, quantitative indexes such as the starting time of machine equipment per hour, the output per day, the frequency of each alarm type, the fluctuation condition of each parameter and the like can be obtained. These metrics can be used in production decisions such as adjusting the run time of the equipment, optimizing production plans, handling alarms in time, etc., to improve production efficiency and quality, reduce equipment failure and downtime.
The edge control module comprises a control runtime unit, a fault management unit and a library management unit.
The control runtime unit can integrate the running environment which accords with IEC61499 and IEC61131-3 standards, the fault management unit carries out hierarchical management on fault information generated by the machine equipment, and the library management unit realizes the management on the control function module of the control object and realizes the quick calling and multiplexing of the program function.
The control runtime unit can install a runtime environment which accords with the IEC61499 and IEC61131-3 standards, and the runtime environment which accords with the IEC61499 and IEC61131-3 standards supports control source programs written by five programming languages of a ladder diagram, a function block diagram, a structured text, a sequence function diagram and an instruction list.
In the invention, an integrated development environment is deployed in a cloud layer, an overall control program for data acquisition equipment and machine equipment is established, and the control program takes a single main machine equipment as a controlled object. The modeling process uses a top-down object-oriented programming method conforming to IEC61499 to form a control program based on a function block network. And mapping the established control program to the operation implementation in the edge control device. Similarly, if IDE based on traditional IEC61131-3 is adopted, the control program is downloaded to the edge control device to run.
The fault management unit carries out hierarchical management on fault information generated by various machine equipment at the terminal layer in the following manner, and sequentially carries out fault information collection, fault classification, fault grade definition and priority determination, fault processing resource allocation, processing result feedback and the like. After the machine equipment at the end layer receives the fault information, the fault information can be classified according to preset rules and strategies. The basis for classification may be the nature of the fault, the extent of impact, the degree of urgency, etc. For different levels of faults, different processing priorities may be determined and reported to the edge control device. Processing resources are allocated reasonably according to the failure level and the processing priority. In the fault processing process, the progress and the result of the processing can be monitored in real time through the management system.
The library management unit is an IEC61499 based control function library that provides a set of models using function block concepts and describes the behavior and structure of distributed process measurements and controls using function block concepts. Through the library, some commonly used functions can be packaged, then the functions are multiplexed in a plurality of programs, if a certain function in the library needs to be updated or repaired, only the library file needs to be updated, and all programs using the function do not need to be modified, so that the library can be more easily used for modularized development. Some of the relevant functions may be packaged into a library, which may then be developed and tested as a stand-alone module.
The edge control module has the advantages of comprising functions of controlling operation, fault management, library management and the like, along with high portability and reusability, universality and flexibility.
4. Operation mode.
The operation modes of the edge control device are divided into two modes, namely an edge autonomous mode and a cloud edge cooperative mode. The operation of the edge autonomous mode is as follows:
1. the operation mode is as follows: in the edge autonomous mode, the edge control device is only connected with the machine equipment of the terminal layer and the data acquisition equipment, and is not connected with the cloud layer server. The edge control device has the capability of self-managing corresponding connection equipment, and can independently control and manage various equipment of the terminal layer.
2. And (3) data processing: in the edge autonomous mode, the data collected by the edge control device is only stored in the local area of the edge control device, and is not uploaded, interacted and stored in the cloud layer server. The data transmission problems due to network delays and bandwidth limitations can be reduced.
3. Code execution: likewise, the code executed by the edge control device in the edge autonomous mode is limited to be local, and has no interaction with the code in the cloud server.
The cloud edge cooperative mode operates as follows:
1. the operation mode is as follows: in the cloud edge cooperative mode, the edge control device establishes connection with the cloud layer server and interacts and cooperates with the cloud layer server. In this mode, the edge control device performs cooperative control with the cloud layer.
2. And (3) data processing: and in the cloud edge cooperative mode, the edge control device uploads the acquired data to the cloud layer server, and performs data interaction and storage with the cloud layer server. This allows for remote monitoring, analysis and processing of the data, as well as facilitating data sharing and collaboration between multiple edge control devices.
3. Code execution: in cloud edge cooperative mode, the code executed by the edge control device can interact with the code in the cloud layer server. This allows for dynamic algorithm updating and functional expansion by transmitting code from the cloud layer to the edge device.
It should be noted that, the data acquired by the data analysis module is synchronized to the cloud layer for storage and calculation according to a preset rule, and the code executed by the edge control module is downloaded to the edge control device according to the preset rule. Under initialization and default conditions, the edge control device will typically select an edge autonomous mode as its operational mode. The edge autonomous mode is independent of the cloud layer server, has high stability and reliability, and can meet basic control requirements.
However, according to specific scenes and requirements, the user can switch the edge control device to the cloud-edge cooperative mode as required so as to realize more comprehensive functions and applications. The default rule is updated according to the time threshold value, and can be defined according to the requirement. In the cloud edge cooperative mode, the edge control device can fully utilize the elastic and telescopic computing and storage resources of the cloud layer, and make up the defect brought by the limitation of computing force of the edge layer.
The English abbreviations related to the invention are explained as follows:
5G: fifth generation mobile communication technology.
Linux: an operating system.
Windows: an operating system based on a graphical user interface.
Dock: an open-source application container engine allows the developer to package their applications and rely on packages into a lightweight, portable container, which is then published to any popular Linux machine, and also allows virtualization.
XML: extensible markup language for describing, transmitting, and storing data.
ID: an identification number.
MQTT: the message queue telemetry transport protocol is a lightweight communication protocol based on a publish/subscribe mode.
Modbus TCP: the advanced Modbus protocol is adapted, a standardized TCP interface is provided, and Modbus devices are allowed to perform seamless communication through Ethernet, so that efficient and reliable data exchange is realized.
NumPy: an open source numeric computation extension toolkit.
Concate: and a function for connecting the arrays according to the designated axes.
OPC UA: a data communication standard allows industrial devices running on different operating system platforms and using different protocols to communicate with each other.
WiFi: a wireless local area network transmission technique.
LoRa: a low power consumption local area network wireless standard.
IEC61499: criteria for a distributed industrial process measurement and control system function block.
4diac Forte: an open source project based on IEC61499 standard is composed of development environment 4diac-ide, runtime Forte, function block library and system instance.
IEC61131-3: international standards for programmable logic controller programming systems, as formulated by the International Electrotechnical Commission (IEC).
Beremiz: and the free software is used for writing a programmable logic controller software program.
Claims (7)
1. The cloud edge cooperative 5G edge control device is characterized by comprising a hardware part and a software part; the hardware part comprises a core processor and at least one 5G communication module; the software part comprises a virtual machine monitor, an operating system, a data analysis module and an edge control module;
the virtual machine monitor is deployed on the hardware part, and two virtual resource spaces are created;
the operating systems are a Linux operating system and a Windows operating system, and are respectively installed in the two virtual resource spaces; the data analysis module operates in the Windows operating system, and the edge control module operates in the Linux operating system;
the data analysis module is in communication connection with the data acquisition equipment deployed at the end layer, analyzes the machine equipment data acquired by the data acquisition equipment based on a protocol model, acquires the acquired data and stores the acquired data;
the edge control module packages and maps a control model of the machine equipment established by the cloud layer and a corresponding composite function block code to a Docker container; the edge control module is in communication connection with the machine equipment deployed at the end layer, acquires working parameters related to the working state of the machine equipment, synchronizes to the cloud layer, receives control model parameters issued by the cloud layer to the edge layer, updates a control algorithm at the edge layer, and realizes cloud edge coordination;
The data analysis module comprises a data storage unit and a protocol model-based analysis unit, wherein the protocol model-based analysis unit is used for analyzing machine equipment data acquired by data acquisition equipment based on a protocol model, and the method comprises the following steps of:
step 1, defining a protocol model f capable of characterizing a plurality of different protocols, and establishing an initial protocol model f by using a known protocol 0 The structure of the protocol model f is characterized in sequence as follows:
f=(L,N,L 1 ,L 2 ,…,L N ,T 1 ,T 2 ,…,T N ,D,P)；
wherein L represents the length of a data frame of a certain protocol except for data bits, N represents the number of protocol features of the protocol, L 1 Representing the first protocol feature length, L 2 Representing the second protocol characteristic length, L N Representing the nth protocol feature length, l=l 1 +L 2 +…+L N ，T 1 Representing the physical attribute corresponding to the first protocol feature, T 2 Representing the physical attribute corresponding to the second protocol feature, T N Representing a physical attribute corresponding to the Nth protocol feature, D represents data content of data bits in a data frame of the protocol, and P represents a protocol type of the protocol;
step 2, performing model training to obtain a trained protocol model f x Will initiate the protocol model f 0 And trained protocol model f x Combining and de-duplicating the same characteristics to obtain a multi-protocol analysis model fA; wherein the initial protocol model f 0 And a trained protocol model f x Stored in cloud layer server, multi-protocol analysis model fA stored in cloud layer serverThe data storage unit;
step 3, inputting the protocol type P of the current data packet x Search matching with the protocol type in the multi-protocol parsing model fA to determine the protocol type P x Whether the analysis model is in the characterization range of the current multi-protocol analysis model fA or not; if the search matching is successful, entering a step 4, otherwise, entering a step 2 to perform model training and updating a multi-protocol analysis model fA;
step 4, determining the protocol type P of which the search matching is successful x Protocol characteristics of (2);
step 5, generating an XML format analysis file according to the data to be acquired and the parameter configuration thereof, thereby obtaining a trained protocol model f x Data content D of data bits in data frames of medium protocol x ；
And step 6, downloading the analysis file into the memory of the cloud-edge cooperative 5G edge control device, and sending data to a cloud layer.
2. The cloud-edge collaborative 5G edge control apparatus of claim 1, wherein the data collection device comprises a sensor that collects operational parameter data of the machine device during operation, a camera that collects on-site video data of the machine device, and a meter that collects parameter data related to control and optimization of the machine device operation.
3. The cloud-edge collaborative 5G edge control apparatus according to claim 1, wherein the initial protocol model f 0 The selected protocol type is Modbus TCP protocol, and the structure is characterized as follows in sequence:
f 0 =(L 0 ,N 0 ,L 0 1 ,L 0 2 ,L 0 3 ,T 0 1 ,T 0 2 ,T 0 3 ,D 0 ,P 0 )；
wherein P is 0 Representing the initial protocol type, i.e. Modbus TCP protocol, L 0 Number representing initial protocolLength other than data bit in data frame, N 0 Number of protocol characteristics, N, representing initial protocol 0 =3, first protocol characteristic length L 0 1 =7, second protocol characteristic length L 0 2 =16, third protocol characteristic length L 0 3 According to the change of the length of the transmitted data, the physical attribute T corresponding to the first protocol characteristic 0 1 Representing the physical attribute T corresponding to the message header and the second protocol feature 0 2 Representing the feature code, and the physical attribute T corresponding to the third protocol feature 0 3 Representing data bits, D 0 Representing the data content of the data bits in the data frame of the initial protocol.
4. The cloud edge collaborative 5G edge control apparatus according to claim 3, wherein the method for obtaining the multi-protocol analysis model fA in step 2 is as follows:
step 2.1, classifying each data frame by using cloud layer computing resources and using a K-means clustering algorithm, thereby obtaining a trained protocol model f x ；
Step 2.2, using the initial protocol model f 0 And trained protocol model f x Updating the multi-protocol parsing model fA.
5. The cloud edge cooperative 5G edge control apparatus according to claim 4, wherein the training method in step 2.1 is as follows:
step 2.1.1, preparing a data set and dividing a training set and a verification set, wherein the data content D of data bits in each data frame of the data set is known;
step 2.1.2, setting the initial clustering center number of the K-means clustering algorithm to be from 3 to 20 for each data frame of the training set, wherein each cluster is a protocol characteristic; finding out the cluster containing the most data content D of the data bits in the data frame, and calculating the percentage R in the cluster, wherein q is the data length of the data content D of the data bits in the data frame, and p is the total data length of the data content D of the data bits in the data frame; sequentially calculating each percentage R in the training set, and averaging Ra;
step 2.1.3, sequentially calculating average value Ra corresponding to the cluster center number from 3 to 20, and determining the corresponding cluster number n when the average value Ra is maximum as the optimal cluster center number;
step 2.1.4: comparing the values corresponding to the n clustering centers to give physical attributes represented by each cluster, and respectively marking the physical attributes as T 1 ~T n ；
Step 2.1.5, comparing the corresponding clustering results, and determining a protocol model, namely a trained protocol model f, of the unknown protocol for access x The expression is as follows:
f x =(L x ,N x ,L x 1 ,L x 2 ,…,L x n ,T x 1 ,T x 2 ,…,T x n ,D x ,P x )；
wherein the number of protocol features N of the unknown protocol x =n，L x 1 ,L x 2 ,…,L x n For trained protocol model f x The first to nth protocol characteristic length of the unknown protocol, the length L of the data frame excluding the data bit x =L x 1 +L x 2 +…+L x n ，T x 1 ,T x 2 ,…,T x n Representing a trained protocol model f x Physical attributes corresponding to the first through nth protocol characteristics, D x Representing the data content of the data bits in the data frame of the unknown protocol, P x Representing the protocol type of the unknown protocol.
6. The cloud edge cooperative 5G edge control apparatus according to claim 5, wherein the method for updating the multi-protocol parsing model fA in step 2.2 is as follows:
step 2.2.1, merging the trained protocol models f x And the multi-protocol analysis model fA is combined and updated, and the expression is as follows:
fA=(L 0 ,L x ,N 0 ,N x ,L 0 1 ,L 0 2 ,L 0 3 ,L x 1 ,L x 2 ,…,L x n ,T 0 1 ,T 0 2 ,T 0 3 ,T x 1 ,T x 2 ,…,T x n ,D 0 ,D x ,P 0 ,P x )；
step 2.2.2: removing duplication of elements which have the same data length and represent the same physical attribute in the multi-protocol analysis model fA;
step 2.2.3: updating a protocol matrix a (P x ) Wherein the initial protocol matrix a (P 0 )=[1,1,1]According to the updated protocol matrix A (P x ) The following are provided:
A(P x )=[0,1,0,1,0,0,0,1,1,1,…,1,0,0,0, ,…,1,0,1,0,1]；
A(P x ) Corresponding to fA in length, updated multi-protocol analysis model fA and each protocol matrix A (P x ) Synchronously downloading to the cloud-edge cooperative 5G edge control device.
7. The cloud-edge cooperative 5G edge control apparatus according to claim 6, wherein the step 4 is specific to a protocol type P for which the search matching is successful x Corresponding protocol matrix a (P x ) Comparing the corresponding bit of the multi-protocol analysis model fA with the corresponding bit of 1, and deleting the corresponding bit of 0, thereby determining the protocol characteristics, namely L x 1 ,L x 2 ,…,L x n ,T x 1 ,T x 2 ,…,T x n 。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410080383.1A CN117590801B (en) | 2024-01-19 | 2024-01-19 | Cloud edge cooperative 5G edge control device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410080383.1A CN117590801B (en) | 2024-01-19 | 2024-01-19 | Cloud edge cooperative 5G edge control device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117590801A CN117590801A (en) | 2024-02-23 |
| CN117590801B true CN117590801B (en) | 2024-04-02 |
Family
ID=89920593
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410080383.1A Active CN117590801B (en) | 2024-01-19 | 2024-01-19 | Cloud edge cooperative 5G edge control device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117590801B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110572448A (en) * | 2019-08-30 | 2019-12-13 | 烽火通信科技股份有限公司 | distributed edge cloud system architecture |
| CN112073461A (en) * | 2020-08-05 | 2020-12-11 | 烽火通信科技股份有限公司 | Industrial Internet system based on cloud edge cooperation |
| CN112286691A (en) * | 2020-11-12 | 2021-01-29 | 济南浪潮高新科技投资发展有限公司 | Cloud edge-side cooperation method based on heterogeneous decision model generation technology |
| CN113467771A (en) * | 2020-03-30 | 2021-10-01 | 中国科学院沈阳自动化研究所 | Model-based industrial edge cloud cooperation system and method |
| JP2022081952A (en) * | 2020-11-20 | 2022-06-01 | 株式会社日立製作所 | Distributed system and data transmission method |
| CN115473947A (en) * | 2022-07-21 | 2022-12-13 | 浪潮通信技术有限公司 | Data acquisition method and device based on cloud edge cooperation and server |
| CN116010114A (en) * | 2023-03-28 | 2023-04-25 | 西安交通大学 | Equipment energy efficiency management and control system based on edge calculation |
-
2024
- 2024-01-19 CN CN202410080383.1A patent/CN117590801B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110572448A (en) * | 2019-08-30 | 2019-12-13 | 烽火通信科技股份有限公司 | distributed edge cloud system architecture |
| CN113467771A (en) * | 2020-03-30 | 2021-10-01 | 中国科学院沈阳自动化研究所 | Model-based industrial edge cloud cooperation system and method |
| CN112073461A (en) * | 2020-08-05 | 2020-12-11 | 烽火通信科技股份有限公司 | Industrial Internet system based on cloud edge cooperation |
| CN112286691A (en) * | 2020-11-12 | 2021-01-29 | 济南浪潮高新科技投资发展有限公司 | Cloud edge-side cooperation method based on heterogeneous decision model generation technology |
| JP2022081952A (en) * | 2020-11-20 | 2022-06-01 | 株式会社日立製作所 | Distributed system and data transmission method |
| CN115473947A (en) * | 2022-07-21 | 2022-12-13 | 浪潮通信技术有限公司 | Data acquisition method and device based on cloud edge cooperation and server |
| CN116010114A (en) * | 2023-03-28 | 2023-04-25 | 西安交通大学 | Equipment energy efficiency management and control system based on edge calculation |
Non-Patent Citations (3)
| Title |
|---|
| 云协同中作战资源两阶段虚拟化方法;孙海洋;张安;高飞;;系统工程与电子技术;20171228(05);第81-87页 * |
| 汽车智能制造中的工业物联网技术;林锦州, 王俊程, 胥博;《自动化博览》;20230331;40(3);第20-24页 * |
| 物联网异构系统综合管理平台的设计开发;林凡;张秋镇;杨峰;;物联网技术;20200720(07);第78-81页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117590801A (en) | 2024-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111562769B (en) | AI extension and intelligent model validation for industrial digital twinning | |
| EP2924561B1 (en) | Component factory for human-machine interface migration to a cloud platform | |
| US10740293B2 (en) | Modular control manifest generator for cloud automation | |
| CN110609512B (en) | Internet of things platform and Internet of things equipment monitoring method | |
| US10095202B2 (en) | Multiple controllers configuration management interface for system connectivity | |
| EP3037901B1 (en) | Cloud-based emulation and modeling for automation systems | |
| CN112579653B (en) | Gradual contextualization and analysis of industrial data | |
| JP2021121954A (en) | Data modeling studio, computer system, and computer-implemented method | |
| CN112580813B (en) | Contextualization of industrial data at device level | |
| US9723060B2 (en) | Integrated nodes, computer readable media and program products, and computer-implemented methods for providing an integrated node for data acquisition, verification and conditioning, and for remote subsystem control | |
| JP6784595B2 (en) | Distributed big data, process control device, distributed big data support method, process control device in process control system | |
| CN108681288A (en) | A kind of novel maintenance system based on cloud platform | |
| US20130179387A1 (en) | Analytical Generator of Key Performance Indicators For Pivoting On Metrics For Comprehensive Visualizations | |
| US20140032169A1 (en) | Systems and methods for improving control system reliability | |
| CN115102827A (en) | Digital product real-time monitoring general Internet platform for small and medium-sized manufacturing industry | |
| CN114281903A (en) | Digital twin construction method, device, equipment and storage medium | |
| Jiang et al. | Design of software-defined gateway for industrial interconnection | |
| CN113176948A (en) | Edge gateway, edge computing system and configuration method thereof | |
| CN117590801B (en) | Cloud edge cooperative 5G edge control device | |
| CN117331614A (en) | On-line detection configuration file management method for aviation manufacturing production site | |
| US20230141305A1 (en) | Symbolic access of industrial device systems and methods based on an on-premise gateway device | |
| Wollschlaeger et al. | Integration of fieldbus systems into on-line asset management solutions based on fieldbus profile descriptions | |
| EP3582034A1 (en) | Method and apparatus, in the field of automation technology, of updating processing data | |
| Sølvsberg et al. | Analysis-oriented structure for runtime data in Industry 4.0 asset administration shells | |
| EP4304157A1 (en) | Edge controller apparatus and corresponding systems, method, and computer program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |