Uniform configuration method and system of industrial heterogeneous network, server and medium

Abstract

The embodiment of the invention relates to the field of industrial internet, and provides a unified configuration method and system for an industrial heterogeneous network, a server and a medium, and the method comprises the steps: receiving network configuration demand information inputted by a user in an application program through a northbound interface; performing model generation processing on the network configuration demand information to obtain a unified information model; performing configuration information generation processing on the unified information model by adopting a converter to obtain configuration information corresponding to a configuration protocol; and sending the configuration information to the target device, so that various heterogeneous network devices can be configured in a unified manner, and the configuration efficiency of the heterogeneous network devices is improved.

Description

Technical Field

[0001] The present application relates to the field of industrial Internet technology, and specifically to a unified configuration method, system, server and medium for industrial heterogeneous networks. Background Art

[0002] With the continuous development of information technology and manufacturing technology, traditional production and manufacturing methods can no longer meet the growing demand, and traditional industries are developing towards automation and intelligence. Industrial automation and intelligence are inseparable from various data on the industrial site. To achieve data interconnection between industrial equipment, it is necessary to break through the communication barriers between various types of equipment. However, there are many types of equipment in the industrial field environment, and the network protocols are different. How to achieve two-way communication between heterogeneous networks is the basic prerequisite for realizing production automation and intelligence.

[0003] Communication between various industrial heterogeneous network devices is a key implementation technology for intelligent manufacturing. To achieve the interconnection of industrial heterogeneous networks, it is necessary to first configure and manage various industrial networks. Different networks in industrial heterogeneous networks have different configuration requirements and various configuration tools. Traditional network configuration methods often use a one-to-one manual configuration mode, which is inefficient and highly repetitive, and it is difficult to meet the many business needs of the Industrial Internet of Things.

[0004] At present, most of the research on industrial heterogeneous network configuration at home and abroad only focuses on the research of configuration methods and the development of configuration systems for a single type or category of industrial networks, and most of them only support one network configuration protocol, resulting in low efficiency in heterogeneous network device configuration. Summary of the invention

[0005] The embodiments of the present application provide a unified configuration method, system, server and medium for an industrial heterogeneous network, which can uniformly configure a variety of heterogeneous network devices and improve the efficiency of heterogeneous network device configuration.

[0006] A first aspect of an embodiment of the present application provides a unified configuration method for an industrial heterogeneous network, which is applied to a unified configuration system. The method includes:

[0007] Receive network configuration requirement information input by the user in the application through the northbound interface;

[0008] Performing model generation processing on the network configuration requirement information to obtain a unified information model;

[0009] Using a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol;

[0010] The configuration information is sent to the target device.

[0011] In this example, the network configuration requirement information input by the user in the application is received through the northbound interface, the network configuration requirement information is modeled to obtain a unified information model, a converter is used to generate configuration information for the unified information model to obtain configuration information corresponding to the configuration protocol, and the configuration information is sent to the target device. Therefore, the configuration information corresponding to the configuration protocol can be automatically generated based on the network configuration requirement information input by the user, and sent to the target device. The target device performs configuration after receiving the configuration information, thereby conveniently configuring the target device and improving the efficiency of device configuration.

[0012] A second aspect of an embodiment of the present application provides a unified configuration system for an industrial heterogeneous network, the system comprising:

[0013] A receiving unit, used for receiving network configuration requirement information input by a user in an application through a northbound interface;

[0014] A processing unit, configured to perform model generation processing on the network configuration requirement information to obtain a unified information model; and use a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to a configuration protocol;

[0015] A sending unit is used to send the configuration information to a target device.

[0016] A third aspect of an embodiment of the present application provides a server, comprising a processor, an input device, an output device and a memory, wherein the processor, input device, output device and memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the step instructions in the first aspect of the embodiment of the present application.

[0017] The fourth aspect of the embodiments of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in the first aspect of the embodiments of the present application.

[0018] A fifth aspect of the embodiments of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application. The computer program product may be a software installation package. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of the architecture of a unified configuration system for an industrial heterogeneous network is provided for an embodiment of the present application;

[0021] Figure 2 A flowchart of a unified configuration method for an industrial heterogeneous network is provided for an embodiment of the present application;

[0022] Figure 3 A schematic diagram of unified information model information is provided for an embodiment of the present application;

[0023] Figure 4 A schematic diagram of a key information extraction model is provided for an embodiment of the present application;

[0024] Figure 5 A schematic diagram of a BERT model structure is provided for an embodiment of the present application;

[0025] Figure 6 A schematic diagram of a flow chart of an embedding layer is provided for an embodiment of the present application;

[0026] Figure 7 A schematic diagram of the structure of a Transformer encoder provided in an embodiment of the present application;

[0027] Figure 8 A structural diagram of a TextCNN model is provided for an embodiment of the present application;

[0028] Fig. 9 A schematic diagram of the structure of a converter is provided for an embodiment of the present application;

[0029] Fig.10 A schematic diagram of an SNMP message format is provided for an embodiment of the present application;

[0030] Fig.11 A schematic diagram of a unified information model framework is provided for the embodiment of the present application;

[0031] Fig.12 A schematic diagram of the structure of a server provided in an embodiment of the present application;

[0032] Fig.13 A structural schematic diagram of a unified configuration system for an industrial heterogeneous network is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0035] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0036] In order to better understand the unified configuration method of an industrial heterogeneous network provided in the embodiment of the present application, the unified configuration method of an industrial heterogeneous network in the existing scheme is briefly introduced below. In the existing scheme, in terms of the industrial network configuration method, Chen Bing et al. proposed a method for automatically collecting network topology information and traffic requirements for TSN automatic configuration based on the NETCONF network configuration protocol to address the shortcomings of the manual configuration method of the time-sensitive network (see the document: Chen Bing, Zhou Haotian, Liu Tian, ​​etc. A TSN network configuration method based on NETCONF [J]. Information Technology and Standardization, 2023, (05): 46-52), which can automatically adapt to changes in network topology and traffic requirements. The centralized network configuration architecture and automatic configuration method based on the centralized network configuration entity CNC and the centralized user configuration entity CUC are described, and the effectiveness of the method is verified through prototype system implementation and testing. Chen Chunlin proposed a general management solution for network device configuration based on NETCONF (see the literature: Chen Chunlin. Design and implementation of a general device configuration management system based on NETCONF [D]. Southeast University, 2018), designed a template-based configuration method, implemented unified encapsulation of configuration messages through configuration templates, and implemented dynamic rendering of configuration interfaces through interface templates. In combination with the various requirements of the device configuration management system, a detailed system design and implementation plan was given.

[0037] However, most of the above methods for research on industrial heterogeneous network configuration only focus on the research of configuration methods and the development of configuration systems for a single type or category of industrial networks, and most of them only support one network configuration protocol, resulting in low efficiency in heterogeneous network device configuration.

[0038] In order to solve the above technical problems, the embodiment of the present application provides a unified configuration method for industrial heterogeneous networks, which can automatically generate configuration information corresponding to the configuration protocol based on the network configuration requirement information input by the user, and send it to the target device. The target device performs configuration after receiving the configuration information, thereby conveniently configuring the target device and improving the efficiency of device configuration.

[0039] Figure 1 FIG. 1 shows a schematic diagram of the architecture of a unified configuration system for industrial heterogeneous networks. Figure 1 As shown in the figure, the unified configuration system of the industrial heterogeneous network receives the network configuration requirement information (configuration requirement in the figure) input by the user plane through the application (APP). On the control plane, each corresponding module in the unified configuration system generates a model to obtain a unified information model. On the data plane, the unified information model is used to generate the final configuration information corresponding to the configuration protocol, and the configuration information is sent to the target device. The target devices include network devices such as 5G / TSN Industrial Ethernet, and field devices such as TSN / WIA-PA / Modbus.

[0040] Specifically, the architecture is divided into the user plane, control plane, and data plane from top to bottom. With the control plane as the center, the configuration architecture can be seen as a north-south direction. The control plane interacts with the user plane and the data plane through the northbound interface and the configuration interface respectively. The functions of each layer of the unified configuration architecture are as follows:

[0041] The user plane includes various software-defined network-based applications (APPs), which can obtain business configuration requirements and implement corresponding network management functions. The user plane APP can interact and manage information with the control plane and data plane by calling the northbound interface.

[0042] The control plane is mainly composed of a unified configuration system. The unified configuration system performs unified modeling for the configuration requirements input by the user, generates configuration information through the coordination of the converter and various configuration modules, and sends it to the data plane through the configuration tool. The control plane communicates with the data plane through the configuration interface and communicates with the user plane through the northbound interface. The unified configuration system manages and configures the devices in the data plane through the configuration interface.

[0043] The data plane is the part of the network responsible for forwarding data packets. It is separated from the control plane and only needs to perform matching and actions according to the instructions of the control plane. Figure 1 It consists of industrial network devices such as 5G gateways and TSN switches, as well as industrial field devices such as PLCs and TSN terminals.

[0044] The basic tasks of the two types of interfaces are:

[0045] Northbound interface: provides an interface for user-side apps to access and manage the network, facilitating unified configuration of resources, i.e., a service interface for the user side.

[0046] Configuration interface: Provides an interface that supports management and configuration functions for field devices and network devices, and an interface for interaction of configuration information between the control plane and the data plane.

[0047] See also Figure 2 , Figure 2 The present invention provides a flowchart of a unified configuration method for an industrial heterogeneous network. Figure 2 As shown, the method is applied to a unified configuration system of an industrial heterogeneous network, and the method includes:

[0048] S20. Receive network configuration requirement information input by the user in the application through the northbound interface.

[0049] Among them, the user plane APP establishes a connection with the unified configuration of the industrial heterogeneous network in the control plane through the northbound interface, thereby realizing the acquisition of heterogeneous network configuration information and the issuance of network configuration requirement information. The northbound interface designed in this application is based on the unified configuration of heterogeneous networks, realizing the functional interaction and information transmission between the user plane and the control plane in the unified configuration architecture of heterogeneous networks.

[0050] The unified configuration system northbound interface should meet the following requirements:

[0051] (1) It has configuration information query interface, configuration information delivery interface, configuration template management interface, device management interface, etc., which can meet the needs of unified configuration of heterogeneous networks.

[0052] (2) To improve the flexibility and reusability of the system, the northbound interface provided should be convenient for other software systems to call, so that other systems can flexibly and conveniently integrate and uniformly configure system functions.

[0053] The northbound interface of the unified configuration system of this application is designed using the HTTP standard interface method, and is designed using standard methods such as GET and POST in the HTTP protocol method. According to the functions defined by the interface, the corresponding logical functions inside the unified configuration system are called by accessing the interface and passing request parameters, and the response data is returned according to the interface definition. According to the interface function requirement definition, the design description of the northbound interface of the unified configuration system is shown in Table 1:

[0054] Table 1 Northbound interface design description

[0055] Interface Name Resource Identifier Request method describe Configuration information query interface / api / getInfo GET Query configuration information Configuration information delivery interface / api / config POST Configuration information delivery Configuration template management interface / api / template POST Managing Configuration Templates Device management interface / api / device POST Manage devices Algorithm result acquisition interface / api / algoResult GET Get algorithm results

[0056] Among them, the configuration information query interface is mainly used to query the existing configuration information in the device, which can be used as a reference for modifying the configuration. The configuration information delivery interface is the main interface of the system, which is responsible for the delivery of each configuration information. The configuration template management interface is for the management of configuration templates, providing users with a way to edit configuration templates by themselves. The device management interface is for the management of device information, which can store general device information for subsequent use. The algorithm result acquisition interface is the interaction with the scheduling system. The scheduling system is an independent system of the unified configuration system but works together with each other. The scheduling system is responsible for running the scheduling algorithm and providing the algorithm results. The unified configuration system obtains the algorithm calculation results of the scheduling system and generates configuration information and returns it through this interface.

[0057] S21. Perform model generation processing on the network configuration requirement information to obtain a unified information model.

[0058] Among them, there are different network types and different device types in industrial heterogeneous network scenarios, and the network configuration methods are also diverse. Therefore, for industrial heterogeneous networks, there must be network configuration information in different formats, so it is necessary to uniformly model the input of industrial heterogeneous network configuration requirements. In view of the different network types and different configuration methods in industrial heterogeneous networks, the basic device information, configuration protocol information, and device configuration information are abstracted from them and combined to form a unified information model, so as to standardize the input of different network configuration requirements and achieve the purpose of unifying the input format of multiple heterogeneous network configuration requirements.

[0059] The unified information model is used to uniformly model different configuration requirements, including basic device information, configuration protocol information, and device configuration information, such as Figure 3 As shown. The basic device information mainly includes basic device parameters such as device type, IP address, and communication port, which are used to identify and describe some basic information of the device. The configuration protocol information includes information such as protocol type, protocol communication address, and possible protocol parameters. The configuration protocol information is used by the subsequent converter to identify the device configuration protocol and perform corresponding conversion. The device configuration information is the specific configuration requirements for the device.

[0060] This application uses the BERT+TextCNN model (key information extraction model) to extract and model key information in user configuration requirements. The overall model is as follows: Figure 4 As shown in the figure. The overall architecture of the key information extraction model can be seen as two parts: BERT and TextCNN. The BERT model is a powerful pre-training model proposed by Google in 2018. It uses a multi-layer Transformer encoder to learn the information before and after each word, and model the text in combination with context-related information, thereby more comprehensively reflecting the semantics of the sentence, and is used to perform deep bidirectional language representation of the input configuration information. The output after BERT model training contains rich semantic knowledge. Therefore, this solution targets the task of text classification and extraction of configuration information, and adds a TextCNN model downstream of the BERT model to perform classification and extraction of configuration information.

[0061] Specifically, performing model generation processing on the network configuration requirement information to obtain a unified information model may specifically include the following steps:

[0062] S211, preprocessing the network configuration requirement information to obtain preprocessed network configuration requirement information;

[0063] Preprocessing can specifically include the following steps:

[0064] S2111, text format information of network configuration requirement information;

[0065] S2112, extracting the network configuration requirement information according to the file extraction method corresponding to the text format information to obtain the network configuration requirement information text;

[0066] S2113: Perform text preprocessing on the network configuration requirement information text to obtain preprocessed network configuration requirement information.

[0067] Among them, in view of the various formats of the network configuration requirement information input by the user, when extracting the network configuration requirement information, this solution will first determine the format in which the network configuration requirement information text is saved, and then use different methods to extract the network configuration requirement information in the file according to the text saving format. After successfully extracting the network configuration requirement information input by the user, it is necessary to perform text preprocessing on the network configuration requirement information to remove noise, stop words, word segmentation, etc., and clean, normalize and convert the network configuration requirement information text for subsequent feature extraction, model training and other tasks.

[0068] S212, using a BERT model to extract features from the preprocessed network configuration requirement information to obtain a configuration information feature vector;

[0069] Among them, the network configuration requirement information text after preprocessing is still not recognizable by the computer, so it is necessary to encode the preprocessed network configuration requirement information text and convert it into a series of vectors that can be recognized by the computer. This application uses the BERT model to encode the network configuration requirement information text. The BERT model structure is as follows Figure 5 shown.

[0070] The BERT model is divided into an embedding layer and an encoding layer. The preprocessed network configuration requirement information text is first input into the embedding layer to generate an embedding vector. Then, the multi-layer Transformer encoders of the encoding layer of the BERT model operate in parallel to extract the feature information of the embedding vector E and obtain the output feature vector.

[0071] Embedding layer

[0072] Embedding vector E i The three embedding vectors of the current configuration information text, the word vector (token), the segment vector (segment) of the sentence where the word is located, and the position vector (position) of the word in the text are concatenated and summed, and CLS and SEP are added as markers for the beginning and end of a text, respectively. Figure 6 As shown in the figure, [CLS] indicates the beginning of a sentence, and [SEP] indicates the end of a sentence.

[0073] The BERT model performs word embedding, segment embedding, and position embedding on the network configuration requirement information text to convert it into a vector representation, and then sums the three embedded vectors to generate a single vector representation as the output of the embedding layer.

[0074] Input the configuration information text X to the BERT model, and the calculation formula is as follows:

[0075] X=(x 1 ,x 2 ,…,x n ) T (Formula 1)

[0076] T i =TokenEmbedding(x i ) (Formula 2)

[0077] S i =SegmentEmbedding(x i ) (Formula 3)

[0078] P i =PositionEmbedding(x i ) (Formula 4)

[0079] E i =T i +S i +P i (Formula 5)

[0080] E=(E 1 ,E 2 ,…,E n ) T (Formula 5)

[0081] Among them, x i is a word or character that makes up the configuration information text X. TokenEmbedding(·) means embedding a single word in the input sequence, SegmentEmbedding(·) means embedding the fragment to which the word belongs, PositionEmbedding(·) means embedding the position of the word, T i represents the word embedding vector, S i represents the sentence token embedding vector, P i represents the position embedding vector, E i is the final embedding vector of the i-th word in the input sequence, and E is the output of the embedding layer, which is an n×d model Dimensional matrix, different BERT models have different embedding vector dimensions. In BERT base, d model It is 768 dimensions.

[0082] Encoding layer

[0083] The network configuration requirement information text is processed by the BERT embedding layer to obtain the embedding layer output E, which is input into the encoding layer to extract the initial text features. The encoding layer of the BERT model is formed by stacking a bidirectional multi-layer Transformer encoder structure. The Transformer encoder structure is as follows: Figure 7 shown.

[0084] The Transformer encoder is mainly divided into three parts: multi-head attention mechanism, residual connection and layer normalization, and feedforward neural network. After the configuration information embedding vector E is input into the encoding layer, it first passes through the multi-head self-attention mechanism to extract the feature vector, and then passes through the residual connection and layer normalization to prevent the overfitting of the model and speed up the convergence of the model. It then passes through the feedforward neural network to further extract features and enhance the model's expression ability.

[0085] 1) Multi-head self-attention mechanism

[0086] In order to better capture the contextual relationship of the input configuration information text and learn the expression of multiple meanings, the embedding layer output E is connected to the self-attention mechanism.

[0087] Each configuration information word vector E in the embedding layer output vector sequence E i Mapped into query vector (query, q), key vector (key, k) and value vector (value, v). The query vector q represents the query relationship between a certain element in the input configuration information sequence and all other positions; the key vector k represents the possible answer, which is used to match the query vector; the value vector v represents the information associated with the key vector, which is the feature that really needs to be extracted. The steps to calculate the self-attention mechanism are as follows:

[0088] First, perform a linear transformation on the embedding layer output E to obtain the Q, K, and V matrices, that is, the matrices corresponding to the q, k, and v vectors. The specific formula is as follows:

[0089] Q=EW Q (Formula 6)

[0090] K=EW K (Formula 7)

[0091] V=EW V (Formula 8)

[0092] Among them, W Q , W K , W V is the weight matrix, and its dimension is d model ×d model , with a mean of 0 and a standard deviation of The normal distribution is initialized and continuously updated during the training process.

[0093] After linear mapping to obtain matrices Q, K, and V, the output of the self-attention mechanism is obtained from these three matrices:

[0094]

[0095] Among them, Softmax(·) is a normalized exponential function. After the Softmax(·) function, the row vector elements are proportionally compressed to [0,1], and the sum of the compressed vector elements is 1. The self-attention mechanism outputs X attention The dimension is n×d model , each row represents the self-attention vector of the corresponding word in the input configuration information, which has integrated the information of words in other positions.

[0096] Get the single-head self-attention output X attention Finally, by setting the number of attention heads h, the self-attention matrix is ​​horizontally spliced, and finally an additional weight matrix is ​​multiplied with the matrix to project it to the original dimension d of the model model , thus obtaining the output of the multi-head self-attention mechanism, the calculation formula is as follows:

[0097]

[0098] MultiHead(Q,K,V)=Concat(head 1 ,head 2 ,…,head h )W O (Formula 11)

[0099] Among them, head i represents the output of the single-head self-attention mechanism, W i Q , represents the W of the i-th head Q , W K , W V Weight matrix, W O represents an additional weight matrix with dimension (h·d model )×d model , Concat(·) represents the concatenation function.

[0100] 2) Residual connection and layer normalization

[0101] The residual connection adds the input and output to pass the information of the previous layer to the next layer. Layer normalization normalizes each sub-layer before output to prevent gradient vanishing and gradient exploding problems. The specific calculation formula is as follows:

[0102] X out =LayerNorm(E+MultiHead(Q,K,V)) (Equation 12)

[0103] Where MultiHead(Q,K,V) is the output of the multi-head self-attention mechanism, and LayerNorm(·) is the layer normalization function.

[0104] 3) Feedforward Neural Network

[0105] The feedforward neural network accepts the output of the multi-head self-attention mechanism after residual connection and layer normalization as input, and maps it into a new representation vector through two linear transformations and a nonlinear activation function. The specific formula is:

[0106] X hidden =ReLU(X out W 1 +b 1 )W 2 +b 2 (Formula 13)

[0107] Where W 1 , W 2 are the weight matrices of two different linear transformations, W 1 The dimension is d model ×n,W 2 The dimension is n×d model , b 1 、b 2 are two bias vectors with dimensions n and d respectively model , ReLU is the rectified linear unit activation function.

[0108] The BERT model uses a multi-layer encoder. Each layer of the encoder uses the above structure. The output of the previous encoder is used as the input of the next encoder. After multiple layers of encoding, the configuration information text is preliminarily extracted to obtain the output feature vector T with a dimension of n×d. model ,This vector contains the text context relationship of the configuration information, contains rich semantic knowledge, and more comprehensively reflects the semantics of the sentence.

[0109] S213, using a TextCNN model to classify the configuration information feature vector to obtain a label classification result corresponding to the network configuration requirement information;

[0110] Among them, ERT as a pre-training model can learn the semantic information of word context. The output after BERT model training contains rich semantic knowledge, which cannot be fully utilized. Therefore, an additional output layer needs to be added to the BERT model after pre-training to adapt to different downstream tasks and new text data. For the task of text classification of configuration information in this solution, it is necessary to add a TextCNN model after the BERT model to perform configuration information classification and extraction.

[0111] TextCNN is a variant of the CNN model. It is an algorithm that uses convolutional neural networks for text classification. Figure 8 As shown in Figure 2, the TextCNN model consists of an embedding layer, a convolutional layer, a pooling layer, and a fully connected layer.

[0112] Embedding layer

[0113] This solution uses a model that combines BERT and TextCNN. Therefore, the BERT model embeds and encodes the input configuration information to obtain the word feature vector T = {T 1 ,T 2 ,…,T n} to replace the original embedding layer of the TextCNN model.

[0114] Convolutional Layer

[0115] The convolution layer uses convolution kernels of different sizes to perform convolution operations on the sentence matrix generated by the embedding layer to obtain the feature matrix c = {c 1 ,c 2 ,…,c n}. Let the convolution kernel w have dimension h×d model , this scheme sets three convolution kernels, namely 2×d model , 3×d model , 4×d model , 2, 3, and 4 are the number of words covered by the convolution kernel, and the convolution kernel is combined with the i-th window T in the configuration information word feature vector matrix T i:i+h-1 The word vectors in the convolution operation are convolved to obtain the feature c i , the convolution operation formula is as follows:

[0116] c i =ReLU(w·T i:i+h-1 +b) (14)

[0117] c′={c 1 ,c 2 ,…,c n-h+1} (15)

[0118] Among them, c iis the i-th element of the convolution result, is a scalar, w is the convolution kernel, T i:i+h-1 is a fragment of the input word feature matrix, the symbol · represents the matrix dot product, b is the bias term, is a scalar, and ReLU is the activation function. After the convolution kernel w is convolved with the word vectors in all windows in the word vector matrix T, the feature map c′ is obtained, and the dimension of c′ is 1×(n-h+1).

[0119] Pooling Layer

[0120] Since the dimensions of feature maps generated by convolution kernels of different sizes are different, it is necessary to perform pooling on the feature map c output by the convolution layer through a pooling layer to obtain a feature vector of fixed length. This solution adopts the 1-max pooling method to select the maximum value in each feature map as the representative value of the feature map, thereby achieving compression of each feature map. On this basis, the pooling results of all convolution kernels are spliced ​​to obtain a global feature map composed of the maximum eigenvalues. The calculation formula is as follows:

[0121]

[0122] Where max(·) is the maximum value function, c′ i The feature map generated by the i-th convolution kernel, is the i-th maximum eigenvalue, where m is the number of feature maps, For splicing operation, It is a global feature with a dimension of 1×m.

[0123] Fully connected layer and output layer

[0124] After the convolutional layer and the pooling layer extract and merge the features, the softmax activation function is used in the fully connected layer to calculate the prediction probability to obtain the final configuration information classification result. The specific calculation formula is as follows:

[0125]

[0126] Among them, y is the final output of the model, which represents the probability distribution of each category, with a dimension of 1×k, where k is the number of categories, and W o is the fully connected layer weight matrix, with dimension m×k, is the global feature after pooling, · is the vector dot product, b o is a bias term with a dimension of 1×k. The softmax function converts the output value into a relative probability to obtain the classification result of the network configuration requirement information.

[0127] S214, extracting key configuration information from the network configuration requirement information according to the label classification result;

[0128] S215: configure the key configuration information to a corresponding position in a preset unified information template to obtain the unified information model.

[0129] After being processed by the above-mentioned BERT model and TextCNN model, the categories of network configuration requirement information are obtained. According to the predefined unified information model, the corresponding categories and parameters are filled into the unified information model respectively, thereby realizing the establishment from user input configuration information to the unified information model.

[0130] S22: Use a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol.

[0131] The main task of the converter is to convert the unified information model into the configuration mode supported by the user equipment, which is the core of realizing unified configuration. Fig. 9 As shown in the figure, the input of the converter is the unified information model established above, and the output is the configuration information in the configuration protocol format supported by the device. The converter contains a model parsing module and a configuration information conversion module. The model parsing is responsible for parsing the unified information model and extracting key information such as configuration protocol information and device configuration information. The configuration information conversion module contains conversion modules for configuration protocols such as CLI, SNMP, NETCONF, and OPC UA, which are responsible for converting the information extracted from the model into configuration information in the corresponding format. After the unified information model is converted by the converter, configuration information in the corresponding format is generated, and then sent to the device through the configuration tool integrated in the system.

[0132] The specific method of using a converter to generate configuration information for the unified information model to obtain configuration information corresponding to the configuration protocol may include the following steps:

[0133] S221, extracting information from the unified information model using the model parsing module to obtain configuration protocol information and device configuration information;

[0134] S222. Determine a configuration protocol template according to the configuration protocol information;

[0135] S223: Perform configuration information generation processing using the configuration protocol template and the device configuration information to obtain configuration information corresponding to the configuration protocol.

[0136] Configuration protocol templates include CLI configuration protocol template, SNMP configuration protocol template, Netconf configuration protocol template, OPC UA configuration protocol template, private protocol configuration protocol template, and default configuration protocol template.

[0137] In a specific implementation, when the configuration protocol template is a CLI configuration protocol template, the following method may be used to generate configuration information corresponding to the configuration protocol.

[0138] The Command Line Interface (CLI) is not a configuration protocol in a strict sense, but CLI is the most commonly used configuration method for traditional devices. Some devices even only support CLI. Therefore, the converter of this solution provides conversion of CLI commands. The conversion steps of the CLI module are as follows:

[0139] A1. Extract key information such as protocol type and device IP address from the unified information model. If the protocol type is CLI, use this module to convert the configuration information.

[0140] A2. Select different CLI configuration information generation templates based on your needs. The configuration commands of different devices and different manufacturers are not the same. To enable the converter to convert the unified information model into the CLI configuration format, you first need to unify the configuration system to integrate the commonly used configuration commands of different devices, and provide an open editing interface for uncommon configuration commands for users to edit and store as command templates for next use. In addition, you can also abstract formatted configuration templates from configuration commands, abstract the required corresponding configuration commands according to different configuration functions, and leave wildcards where configuration data is needed. When the converter determines which configuration commands need to be called, it replaces the wildcards in the configuration template with the data to be configured, thereby completing the construction of the configuration command.

[0141] A3. Select the generated template and then perform model conversion. The following is a brief introduction to the model conversion of the CLI configuration information generation template. The configuration command generation template of a manufacturer's TSN terminal VLAN and the converted configuration commands are as follows:

[0142] Generate template: vlan bridge vlan add dev*vid*pvid

[0143] Configuration command: vlan bridge vlan add dev swp1 vid 100pvid

[0144] For this template, the CLI module fills the specific configuration information extracted from the unified information model, such as {swp1:100}, into the configuration template to replace the wildcard *, thereby generating a configuration command.

[0145] A4. Output the generated configuration command. After the configuration command is output, configuration information generation processing is performed according to the configuration command to obtain configuration information corresponding to the configuration protocol.

[0146] In a specific implementation, when the configuration protocol template is an SNMP configuration protocol template, the following method may be used to generate configuration information corresponding to the configuration protocol.

[0147] SNMP is based on the SGMP (Simple Gateway Monitoring Protocol) protocol and can remotely manage network devices that support this protocol, including device discovery, operating status monitoring, device configuration modification, etc. The conversion steps of the SNMP module are as follows:

[0148] B1. Extract key information such as protocol type and device IP address from the unified information model. If the protocol type is SNMP, use this module to convert the configuration information.

[0149] B2.SNMP has a standard protocol message format, so there is no need to select a template, just encapsulate the configuration information according to the message format. SNMP specifies 5 types of protocol data units (PDUs) for exchange between management processes and agents. The unified configuration system mainly uses SNMP to configure devices, so only the get and set operations in the 5 specified protocol data units are used.

[0150] The SNMP message format is as follows: Fig.10 As shown in the figure, the get / set variable part in the SNMP message consists of the variable name and value. When performing a get operation to obtain the variable value, only the variable name is filled in, and the variable value is ignored. The variable value is filled in when performing a set operation. SNMP uses object identifiers (OIDs) to uniquely identify devices. Therefore, when setting the get / set variable part, the variable name should be set to the object identifier of the parameter to be configured. For public MIBs supported by SNMP, the OIDs of commonly used public variables can be preset in the unified configuration system for users to choose. For private MIBs, an open editing interface can be provided to allow users to fill in the OIDs of private MIBs by themselves. In the unified configuration system, the variable name entered by the user and the corresponding OID can be bound together, so that the variable name can be easily accessed when the OID is used later.

[0151] B3. Perform model conversion for SNMP message format. The SNMP module fills the specific configuration information extracted from the unified information model, such as {1.3.6.1.2.1.1.5: "test"}, into the get / set variable part of the SNMP message, and fills in parameters such as community and PDU type according to the message format, thereby generating a configuration message.

[0152] B4. Output the generated configuration message.

[0153] In a specific implementation, when the configuration protocol template is a Netconf configuration protocol template, the following method may be used to generate configuration information corresponding to the configuration protocol.

[0154] Netconf (Network Configuration Protocol) is a network configuration protocol used to configure network devices and monitor their status. It is designed to replace the traditional CLI method to manage network devices and provides a more structured and automated configuration management method. The conversion steps of the Netconf module are as follows:

[0155] C1. Extract key information such as protocol type and device IP address from the unified information model. If the protocol type is Netconf, use this module to convert configuration information.

[0156] C2. Netconf has a standard protocol message format, so there is no need to select a template, just encapsulate the configuration information according to the message format. The Netconf protocol can be divided into 4 layers, and the Netconf module mainly converts the configuration information in the operation layer. Netconf defines three operation objects in the operation layer, namely the running configuration library (running), the candidate configuration library (candidate), and the startup configuration library (startup). Different configuration libraries have different corresponding states and can be flexibly selected by users. In addition, the operation layer defines a set of basic protocol operations based on XML encoding parameters and RPC method calls, which mainly include four aspects in terms of functions: value operation, configuration operation, lock operation and session operation, and supports user-defined RPC operations. This solution mainly uses Netconf to configure the device, so the Netconf module will use common value and configuration operations.

[0157] Table 2 Netconf message example

[0158]

[0159]

[0160] As shown in Table 2, Netconf encodes the request message based on XML format, where <rpc>The element is used to encapsulate the RPC request, and the RPC request contains the editing operation of the operation layer. <edit-config>The operation layer contains operation objects and some common operation options such as default operation and error options. <config>Inside the tag is the Netconf content layer, which contains the configuration changes to be made, such as adding or modifying a vlan.

[0161] C3. Model conversion is performed for the Netconf message format. The Netconf module uses the open source XML library to construct the Netconf message. First, the outer rpc layer has nothing to do with user configuration requirements and is automatically generated by the program. Then the internal operation layer encapsulates the configuration operation. <edit-config>Operation, encapsulation when performing configuration query <get-config>The specific configuration content of the operation, configuration object and content layer is entered by the user in the configuration requirements, and then the specific configuration information is extracted from the unified information model, converted into XML tag format by the Netconf module, and finally encapsulated into a Netconf configuration message.

[0162] C4. Output the generated configuration message.

[0163] In a specific implementation, when the configuration protocol template is an OPC UA configuration protocol template, the following method can be used to generate configuration information corresponding to the configuration protocol:

[0164] OPC UA (Open Platform Communications Unified Architecture) is an open platform communications unified architecture for interoperability and integration in industrial automation and control systems. OPC UA is based on standardized communication protocols and data models, and is designed to solve communication and integration problems between different manufacturers, devices, and systems. The conversion steps of OPC UA modules are as follows:

[0165] D1. Extract key information such as protocol type and device IP address from the unified information model. If the protocol type is OPC UA, use this module to convert the configuration information.

[0166] D2.OPC UA has a standard protocol message format, so there is no need to select a template, just encapsulate the configuration information according to the message format. The encoding format of OPC UA messages usually adopts XML text format or binary format. Since the XML encoding format is convenient for different applications and platforms to use XML parsers to interpret OPC UA messages, this solution mainly describes the conversion of configuration information in XML encoding format.

[0167] According to the definition of OPC UAService in Part 4 of the OPC UA specification (OPC 10000-4: UA Part 4: Services) and the description of the data encoding mechanism in Part 6 of the OPC UA specification (OPC 10000-6: UA Part 6: Mappings), the request message body in the XML encoding format of OPC UA is shown in Table 3.

[0168] Table 3 OPC UA request message example

[0169]

[0170]

[0171] According to the OPC UA specification, in the above OPC UA request message body, ReadRequest is a fixed request tag, the RequestHeader tag contains the request header parameters, the MaxAge tag is the maximum interval time, the TimestampsToReturn tag is the return timestamp option, and the NodesToWrite tag contains the nodes to be written. If the node is to be read, it can be replaced with the NodesToRead tag accordingly.

[0172] D3. Model conversion is performed for the OPC UA message format. The OPC UA module uses an open source XML library to construct OPC UA messages. Tags such as RequestHeader and MaxAge are all optional parameters, and usually the default values ​​can be set. The NodesToRead and NodesToWrite tags perform read / write operations on nodes in OPC UA. Extract specific configuration parameters such as {NodeId:”ns=2;i=1001”,Value=123.45} from the unified information model, and fill them into the NodesToWrite tag to complete the construction of the OPC UA request message.

[0173] D4. Output the generated configuration message.

[0174] In a specific implementation, when the configuration protocol template is a private protocol configuration protocol template, the following method can be used to generate configuration information corresponding to the configuration protocol:

[0175] In order to improve the universality, this solution provides an editing interface for private protocols. For simple private protocols, private protocol information can be edited through the configuration template management interface of the unified configuration system. The conversion steps of the private protocol module are as follows:

[0176] E1. Extract key information such as protocol type and device IP address from the unified information model. If the protocol type is PRIVATE, use this module to convert the configuration information.

[0177] E2. Select different private protocol configuration information generation templates according to needs. Here, the private protocol configuration information generation template is edited by the user through the configuration template management interface of the unified configuration system. Similar to the CLI template, the formatted configuration template is also abstracted from the private protocol configuration information, leaving wildcards where configuration data is required.

[0178] E3. After selecting the generated template, perform model conversion. Use the specific configuration information extracted from the unified information model to replace the wildcard, thereby completing the construction of the private protocol configuration information.

[0179] E4. Output the generated configuration information.

[0180] In a specific implementation, when the configuration protocol template is a default configuration protocol template, the following method can be used to generate configuration information corresponding to the configuration protocol:

[0181] If the protocol type is not identified in the unified information model, that is, the protocol type is UNRECOG, or some necessary parameters are missing, two methods are provided to solve the problem.

[0182] F1. If the configuration requirements input by the user meet the protocol type in the converter and the necessary parameters are complete, consider letting the unified information model re-identify. If the necessary parameters are missing, the user needs to complete them in the input configuration requirements before re-identification. If the necessary parameters are still missing, the unified configuration system will prompt the user that the items are missing and ask the user to enter them manually in the unified configuration system.

[0183] F2. If the configuration requirements entered by the user do not conform to the protocol type in the converter, the private protocol module will be used, and the user can edit the configuration information to generate the template.

[0184] S23. Send the configuration information to the target device.

[0185] The unified information model generates configuration messages in the corresponding protocol format after conversion by the converter for different configuration protocol formats. The next step is to send them to specific devices through the configuration tool. The unified configuration system integrates configuration tools such as Netconf client, OPC UA client, SNMP Manager, etc., and is used in conjunction with the Netconf server and OPC UA server in the device to implement configuration operations on the device, such as Fig.11 shown.

[0186] For devices that support different configuration protocols, call the corresponding configuration protocol tool to implement configuration delivery. For example, TSN switches use the Netconf protocol for configuration, which can configure the switch's gating list configuration, port enable status and other functions. The Netconf configuration information in XML format generated by the converter is delivered to the switch through the Netconf client integrated in the unified configuration system. The Netconf server deployed in the switch can verify and parse the configuration information, and perform functional configuration based on the configuration results obtained by the analysis. Industrial wireless networks mainly include three types of devices: industrial wireless gateway devices, industrial wireless routing devices, and field wireless nodes. For industrial wireless gateways, the unified configuration system establishes a connection with the OPC UA server inside the industrial wireless gateway through the OPC UA client, and sends the industrial wireless network configuration information to the industrial wireless gateway device. After the industrial wireless gateway device obtains the configuration information, it generates a routing table, a superframe table, and a link table, sends configuration information frames to the industrial wireless routing devices in the network, and the industrial wireless routing devices send beacon frames and superframe configuration information to the field wireless nodes. The field wireless nodes complete the superframe configuration task based on the configuration information.

[0187] For the above embodiments, please refer to Fig.12 , Fig.12 A structural diagram of a server provided in an embodiment of the present application, as shown in the figure, includes a processor, an input device, an output device and a memory, the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, the processor is configured to call the program instructions, and the program includes instructions for executing the following steps;

[0188] Receive network configuration requirement information input by the user in the application through the northbound interface;

[0189] Performing model generation processing on the network configuration requirement information to obtain a unified information model;

[0190] Using a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol;

[0191] The configuration information is sent to the target device.

[0192] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that in order to realize the above functions, the terminal includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0193] The embodiment of the present application can divide the terminal into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0194] In line with the above, see Fig.13 , Fig.13 The present application provides a schematic diagram of a unified configuration system for an industrial heterogeneous network. Fig.13 As shown, the system comprises:

[0195] The receiving unit 501 is used to receive the network configuration requirement information input by the user in the application through the northbound interface;

[0196] The processing unit 502 is used to perform model generation processing on the network configuration requirement information to obtain a unified information model; and use a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol;

[0197] The sending unit 503 is used to send the configuration information to the target device.

[0198] In a specific implementation, in the aspect of performing model generation processing on the network configuration requirement information to obtain a unified information model, the processing unit 502 is specifically configured to:

[0199] Preprocessing the network configuration requirement information to obtain preprocessed network configuration requirement information;

[0200] Using the BERT model to extract features from the preprocessed network configuration requirement information to obtain a configuration information feature vector;

[0201] The TextCNN model is used to classify the configuration information feature vector to obtain a label classification result corresponding to the network configuration requirement information;

[0202] Extracting key configuration information from the network configuration requirement information according to the label classification result;

[0203] The key configuration information is configured to a corresponding position in a preset unified information template to obtain the unified information model.

[0204] In a specific implementation, in the aspect of preprocessing the network configuration requirement information to obtain the preprocessed network configuration requirement information, the processing unit 502 is specifically used to:

[0205] Network configuration requirement information in text format;

[0206] Extract the network configuration requirement information according to the file extraction method corresponding to the text format information to obtain the network configuration requirement information text;

[0207] The network configuration requirement information text is preprocessed to obtain preprocessed network configuration requirement information.

[0208] In a specific implementation, the converter includes a model parsing module and a configuration information conversion module. In the aspect of using the converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol, the processing unit 502 is specifically used to:

[0209] The model parsing module is used to extract information from the unified information model to obtain configuration protocol information and device configuration information;

[0210] Determine a configuration protocol template according to the configuration protocol information;

[0211] The configuration information generation process is performed using the configuration protocol template and the device configuration information to obtain configuration information corresponding to the configuration protocol.

[0212] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any one of the unified configuration methods for industrial heterogeneous networks recorded in the above method embodiments.

[0213] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program enables a computer to execute part or all of the steps of any one of the unified configuration methods for industrial heterogeneous networks recorded in the above method embodiments.

[0214] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0215] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0216] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0217] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0218] In addition, the functional units in the various embodiments of the application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software program modules.

[0219] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, disk or optical disk and other media that can store program codes.

[0220] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which can include: a flash drive, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.

[0221] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application. < / config> < / rpc>

Claims

1. A unified configuration method for industrial heterogeneous networks, characterized in that: Applied to a unified configuration system, the method comprises: Receive network configuration requirement information input by the user in the application through the northbound interface; Performing model generation processing on the network configuration requirement information to obtain a unified information model; Using a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol; The configuration information is sent to the target device.

2. The unified configuration method for industrial heterogeneous networks according to claim 1 is characterized in that: The performing model generation processing on the network configuration requirement information to obtain a unified information model includes: Preprocessing the network configuration requirement information to obtain preprocessed network configuration requirement information; Using the BERT model to extract features from the preprocessed network configuration requirement information to obtain a configuration information feature vector; The TextCNN model is used to classify the configuration information feature vector to obtain a label classification result corresponding to the network configuration requirement information; Extracting key configuration information from the network configuration requirement information according to the label classification result; The key configuration information is configured to a corresponding position in a preset unified information template to obtain the unified information model.

3. The unified configuration method for industrial heterogeneous networks according to claim 2 is characterized in that: The preprocessing of the network configuration requirement information to obtain the preprocessed network configuration requirement information includes: Network configuration requirement information in text format; Extract the network configuration requirement information according to the file extraction method corresponding to the text format information to obtain the network configuration requirement information text; The network configuration requirement information text is preprocessed to obtain preprocessed network configuration requirement information.

4. The unified configuration method for industrial heterogeneous networks according to any one of claims 1 to 3, characterized in that: The converter includes a model parsing module and a configuration information conversion module. The converter is used to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to the configuration protocol, including: The model parsing module is used to extract information from the unified information model to obtain configuration protocol information and device configuration information; Determine a configuration protocol template according to the configuration protocol information; The configuration information generation process is performed using the configuration protocol template and the device configuration information to obtain configuration information corresponding to the configuration protocol.

5. The unified configuration method for industrial heterogeneous networks according to claim 4 is characterized in that: The configuration protocol templates include a CLI configuration protocol template, an SNMP configuration protocol template, a Netconf configuration protocol template, an OPC UA configuration protocol template, a private protocol configuration protocol template and a default configuration protocol template.

6. A unified configuration system for industrial heterogeneous networks, characterized in that: The system comprises: A receiving unit, used for receiving network configuration requirement information input by a user in an application through a northbound interface; A processing unit, configured to perform model generation processing on the network configuration requirement information to obtain a unified information model; and use a converter to perform configuration information generation processing on the unified information model to obtain configuration information corresponding to a configuration protocol; A sending unit is used to send the configuration information to a target device.

7. The unified configuration method for industrial heterogeneous networks according to claim 6, characterized in that: In the aspect of performing model generation processing on the network configuration requirement information to obtain a unified information model, the processing unit is specifically used to: Preprocessing the network configuration requirement information to obtain preprocessed network configuration requirement information; Using the BERT model to extract features from the preprocessed network configuration requirement information to obtain a configuration information feature vector; The TextCNN model is used to classify the configuration information feature vector to obtain a label classification result corresponding to the network configuration requirement information; Extracting key configuration information from the network configuration requirement information according to the label classification result; The key configuration information is configured to a corresponding position in a preset unified information template to obtain the unified information model.

8. The unified configuration method for industrial heterogeneous networks according to claim 7, characterized in that: In the aspect of preprocessing the network configuration requirement information to obtain the preprocessed network configuration requirement information, the processing unit is specifically used to: Network configuration requirement information in text format; Extract the network configuration requirement information according to the file extraction method corresponding to the text format information to obtain the network configuration requirement information text; The network configuration requirement information text is preprocessed to obtain preprocessed network configuration requirement information.

9. A server, characterized in that: The method comprises a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 5.

Images