Standard management method and system for steel industry data resource classification

Abstract

The invention belongs to the technical field of data management, and discloses a standard management method and system for data resource classification in the iron and steel industry. The method comprises the following steps: a cloud data center deploys a block chain network and constructs a steel industry knowledge graph and a data resource classification model; performing semantic enhancement on the plurality of real-time steel industry data resources; performing data resource classification on the plurality of real-time steel industry data resources after semantic enhancement; and performing table entry management on the plurality of real-time iron and steel industry data resources after semantic enhancement and the real-time data resource classification results to obtain a real-time iron and steel industry data resource form, performing visual display on the real-time iron and steel industry data resource form by using a data resource map, and performing distributed storage. According to the method, the problems of obstacles in data resource management, low processing accuracy, poor model generalization ability, low management intelligence degree and low data storage security in the prior art are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data management, and in particular relates to a standard management method and system for data resource classification in the steel industry. Background Art

[0002] As an important pillar of the national economy, the steel industry generates a large amount of data during its production process. These data are characterized by large volume, diverse types, and complex semantics, which poses a huge challenge to the data resource management of the steel industry. How to effectively manage these data resources in the steel industry and improve the utilization value of data resources in the steel industry has become a challenge facing the industry.

[0003] The existing data resource management technology in the steel industry has the following defects: 1) The steel industry is in the stage of digital transformation and intelligent manufacturing, but there is a lack of unified standards for identifying and entering data resources in the industry, making data collection, governance and integration difficult, leading to obstacles in data resource management; 2) Low processing accuracy: Existing technologies often rely on traditional data processing methods, which are difficult to guarantee the accuracy of data processing when faced with massive, complex and changeable data resources in the steel industry; 3) Poor model generalization ability: The models used in existing technologies have limited generalization ability to new data or unfamiliar environments, which means that the model performs well in the training environment, but may encounter performance degradation in actual applications; 4) Low level of intelligent management: Data management in existing technologies relies too much on manual operations, which is inefficient and error-prone, and is unable to discover the potential characteristics of data resources in the steel industry, resulting in poor management results; 5) Low data storage security: Existing technologies often use centralized storage methods to store data resources in the steel industry. This method is prone to data loss due to hardware crashes, is vulnerable to hacker attacks, and has a high risk of data leakage. Summary of the invention

[0004] In order to solve the problems of data resource management obstacles, low processing accuracy, poor model generalization ability, low management intelligence and low data storage security in the prior art, the present invention aims to provide a standard management method and system for data resource classification in the steel industry.

[0005] The technical solution adopted by the present invention is: A standard management method for data resource classification in the steel industry includes the following steps: Cloud data center, deploying blockchain network, and using artificial intelligence algorithms to build steel industry knowledge graph and data resource classification model based on several historical steel industry data resources and several steel industry knowledge; The cloud data center uses the steel industry knowledge graph to semantically enhance a number of real-time steel industry data resources, and obtains a number of semantically enhanced real-time steel industry data resources; The cloud data center uses the data resource classification model to classify a number of semantically enhanced real-time steel industry data resources, and obtains a number of real-time data resource classification results and an updated data resource classification model; The cloud data center manages the entry of several semantically enhanced real-time steel industry data resources and the corresponding real-time data resource classification results into tables, obtains real-time steel industry data resource forms, uses data resource maps to visualize the real-time steel industry data resource forms, and uses blockchain networks to distribute and store real-time steel industry data resource forms.

[0006] Furthermore, the cloud data center deploys a blockchain network and uses artificial intelligence algorithms to build a steel industry knowledge graph and data resource classification model based on several historical steel industry data resources, including the following steps: In the cloud data center, several data servers are connected as data nodes in a distributed manner, and smart contracts and IPFS systems are set up to obtain a blockchain network; Collecting a number of historical steel industry data resources and a number of steel industry knowledge, and preprocessing them to obtain a number of preprocessed historical steel industry data resources and a number of preprocessed steel industry knowledge; Using a pre-trained named entity and entity relationship extraction model, named entity and entity relationship extraction is performed on a number of pre-processed steel industry knowledge to obtain a number of knowledge named entities and knowledge entity relationships between each knowledge named entity and other knowledge named entities; Construct a knowledge graph for the steel industry based on several knowledge named entities and corresponding knowledge entity relationships; Using the steel industry knowledge graph, semantic enhancement is performed on several pre-processed historical steel industry data resources to obtain several semantically enhanced historical steel industry data resources; Based on several semantically enhanced historical steel industry data resources, a data resource classification model is constructed using a fusion algorithm of continuous learning and deep learning.

[0007] Furthermore, based on several semantically enhanced historical steel industry data resources, a data resource classification model is constructed using a continuous learning and deep learning fusion algorithm, including the following steps: Set corresponding real data resource classification labels for several semantically enhanced historical steel industry data resources, obtain several data resource classification samples, and use deep learning algorithms to build an initial data resource classification model; Taking minimizing the model error as the optimization goal, the swarm intelligence optimization algorithm is used to optimize the initial model parameters of the initial data resource classification model to obtain an optimized data resource classification model; Using several data resource classification samples, the optimized data resource classification model is optimized and trained to obtain the optimal data resource classification model, and several historical data resource classification experiences are generated; Using the experience replay mechanism of the continuous learning algorithm, an experience replay pool is set up in the optimal data resource classification model, and several historical data compression experiences are stored in the experience replay pool; Using the elastic weight connection mechanism of the continuous learning algorithm, the initial loss function of the optimal data resource classification model is adjusted to obtain the adjusted loss function and the final data resource classification model.

[0008] Furthermore, the named entity and entity relationship extraction model is built based on the BERT-Double CRF algorithm.

[0009] Furthermore, the data resource classification model is constructed based on the IFWA-GCN-Elman-CLA algorithm.

[0010] Further, using a named entity and entity relationship extraction model, extracting a number of data named entities of each preprocessed real-time steel industry data resource and data entity relationships between each data named entity and other data named entities, including the following steps: Input the preprocessed real-time steel industry data resources into the named entity and entity relationship extraction model, and extract the real-time text features of the preprocessed real-time steel industry data resources; According to the real-time text features, several data named entities of the pre-processed real-time steel industry data resources are extracted; Extracting data entity relationships between each data named entity and other data named entities based on real-time text features and a number of data named entities; All pre-processed real-time steel industry data resources are traversed to obtain a number of data named entities and a number of data entity relationships of each pre-processed real-time steel industry data resource.

[0011] Furthermore, the cloud data center uses the steel industry knowledge graph to semantically enhance a number of real-time steel industry data resources to obtain a number of semantically enhanced real-time steel industry data resources, including the following steps: The cloud data center collects a number of real-time steel industry data resources and performs preprocessing to obtain a number of preprocessed real-time steel industry data resources; Using a named entity and entity relationship extraction model, extracting a number of data named entities of each pre-processed real-time steel industry data resource and data entity relationships between each data named entity and other data named entities; Obtain the Euclidean distance between each data named entity and all knowledge named entities in the steel industry knowledge graph, use the knowledge named entity with the closest Euclidean distance as the mapping named entity of the data named entity, and use several knowledge entity relationships of the mapping named entity as the mapping entity relationships of the data named entity; Mapping all mapped named entities to corresponding data named entities in the preprocessed real-time steel industry data resources, and mapping all mapped entity relationships to data entity relationships corresponding to the data named entities, to obtain semantically enhanced real-time steel industry data resources; Traverse all preprocessed real-time steel industry data resources to obtain several semantically enhanced real-time steel industry data resources.

[0012] Further, the cloud data center uses the data resource classification model to classify a number of semantically enhanced real-time steel industry data resources, and obtains a number of real-time data resource classification results and an updated data resource classification model, including the following steps: The cloud data center inputs the semantically enhanced real-time steel industry data resources into the data resource classification model to extract the corresponding real-time graph structure features; According to the real-time graph structure characteristics, data resources are classified to obtain the real-time data resource classification results and generate the corresponding real-time model error value; According to the real-time graph structure characteristics, the real-time model error value and the real-time model parameters of the data resource classification model, the corresponding real-time data resource classification experience is obtained; Traverse all semantically enhanced real-time steel industry data resources to obtain several real-time data resource classification results and some real-time data resource classification experiences; Randomly extract a number of different historical data resource classification experiences from the experience replay pool, and mix the number of historical data resource classification experiences with a number of real-time data resource classification experiences to obtain a number of mixed data resource classification experiences; Based on the experience of classifying several mixed data resources, the data resource classification model is continuously trained, and the loss value of the continuous learning training is obtained using the adjusted loss function; If the loss value is always less than the loss value threshold, the updated data resource classification model is obtained; otherwise, some historical data resource classification experiences are re-extracted and continuous learning and training is continued.

[0013] Furthermore, the cloud data center manages the entry of a number of semantically enhanced real-time steel industry data resources and corresponding real-time data resource classification results into a table, obtains a real-time steel industry data resource form, uses a data resource map to visualize the real-time steel industry data resource form, and uses a blockchain network to perform distributed storage of the real-time steel industry data resource form, including the following steps: The cloud data center associates the semantically enhanced real-time steel industry data resources with the corresponding real-time data resource classification results to obtain standard real-time steel industry data resources; Manage the entry of several standard real-time steel industry data resources into tables, obtain the real-time steel industry data resource form, and use the pre-built data resource map to visualize the real-time steel industry data resource form; Store the real-time steel industry data resource form in the IPFS system of the blockchain network, obtain the data hash value, and generate the corresponding real-time data storage request; Use the PBFT consensus algorithm to reach consensus on real-time data storage requests among several distributed connected data nodes. If the consensus succeeds, proceed to the next step; otherwise, end the data storage. Call the smart contract of the blockchain network, generate corresponding real-time transaction data according to the consensus success message and data hash value, and convert the real-time transaction data into real-time data blocks; The real-time data block is sent to several distributed connected data nodes, and the real-time data block is linked to the chain using the data nodes that received the real-time data block.

[0014] A standard management system for data resource classification in the steel industry is used to implement a standard management method. The system is arranged in a cloud data center, and the system includes a model building unit, a semantic enhancement unit, a data resource classification unit and a distributed storage unit which are connected in sequence.

[0015] The beneficial effects of the present invention are: The present invention provides a standard management method and system for data resource classification in the steel industry, which covers the full life cycle standard management of data resource collection, classification, governance, table entry and storage in the steel industry, formulates a data resource management process, ensures the standardized processing of data resources in the full life cycle, so as to improve the data resource management level and resource utilization efficiency of steel enterprises, and eliminates the obstacles to data resource management; by constructing a knowledge graph and a data resource classification model for the steel industry, the semantic enhancement of real-time data resources in the steel industry is achieved, thereby improving the accuracy and quality of data processing and analysis; by using a continuous learning algorithm to improve the data resource classification model, it is ensured that the model is continuously trained for real-time data, and the model effectively reduces the forgetting of historical experience when learning new data, thereby improving the generalization ability of the model; an automated data resource classification and storage system is adopted to improve the intelligence level and management efficiency of management, and avoid excessive manual dependence, wherein the deep learning data resource classification model can discover the potential characteristics of data resources in the steel industry and improve the management effect; a decentralized blockchain network is adopted to perform distributed storage on data resources and data resource classification results in the steel industry, and the distributed backup of the blockchain network is utilized to improve the reliability of data storage, and data can be protected from tampering and leakage, thereby ensuring the security and integrity of data.

[0016] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the standard management method for data resource classification in the steel industry in the present invention.

[0018] Figure 2 It is a framework diagram of the data resource map in the present invention.

[0019] Figure 3 It is a structural diagram of the standard management system used for data resource classification in the steel industry in the present invention. DETAILED DESCRIPTION

[0020] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0021] Embodiment 1: like Figure 1 As shown, this embodiment provides a standard management method for data resource classification in the steel industry, comprising the following steps: S1: Cloud data center, deploying blockchain network, and using artificial intelligence algorithms to build a steel industry knowledge graph and data resource classification model based on several historical steel industry data resources and some steel industry knowledge, including the following steps: S1-1: Cloud data center, connects several data servers as data nodes in a distributed manner, sets up smart contracts and IPFS system, and obtains a blockchain network; S1-2: Collect a number of historical steel industry data resources and a number of steel industry knowledge, and perform preprocessing to obtain a number of preprocessed historical steel industry data resources and a number of preprocessed steel industry knowledge; S1-3: Using a pre-trained named entity and entity relationship extraction model, extract named entities and entity relationships from a number of pre-processed steel industry knowledge to obtain a number of knowledge named entities and knowledge entity relationships between each knowledge named entity and other knowledge named entities; The named entity and entity relationship extraction model is constructed based on the Bidirectional Encoder Representations from Transformers (BERT)-Double Conditional Random Fields (CRF) algorithm, and the named entity and entity relationship extraction model includes a text feature extraction module constructed based on the BERT algorithm, a named entity extraction module constructed based on the CRF algorithm, and an entity relationship extraction module constructed based on the CRF algorithm. The text feature extraction module is connected to the named entity extraction module and the entity relationship extraction module respectively; Using a pre-trained named entity and entity relationship extraction model, named entities and entity relationships are extracted from a number of pre-processed steel industry knowledge to obtain a number of knowledge named entities and knowledge entity relationships between each knowledge named entity and other knowledge named entities, including the following steps: S1-3-1: input the preprocessed real-time steel industry data resources into the named entity and entity relationship extraction model, and use the text feature extraction module to extract the real-time text features of the preprocessed real-time steel industry data resources; S1-3-2: Based on the real-time text features, a named entity extraction module is used to extract several data named entities of the pre-processed real-time steel industry data resources; S1-3-3: extracting the data entity relationship between each data named entity and other data named entities using an entity relationship extraction module based on the real-time text features and a number of data named entities; S1-3-4: traverse all pre-processed real-time steel industry data resources to obtain a number of data named entities and a number of data entity relationships of each pre-processed real-time steel industry data resource; S1-4: Construct a knowledge graph for the steel industry based on several knowledge named entities and corresponding knowledge entity relationships; S1-5: Use the steel industry knowledge graph to semantically enhance several pre-processed historical steel industry data resources to obtain several semantically enhanced historical steel industry data resources; Through mapping, the original information in the steel industry data resources is endowed with additional semantic information, which comes from the structured knowledge in the steel industry knowledge graph. The originally simple steel industry data resources have become semantic entities and relationships rich in contextual information. In addition, the steel industry data resources have changed from original low-dimensional data to high-dimensional data with graph structure information, which improves the representation ability of steel industry data resources for complex relationships and provides high-dimensional information expression for subsequent model analysis. The data resources of the steel industry include the entire life cycle of the steel industry, including data generated during the process of R&D, design, procurement, construction, production, sales, and operation; the knowledge graph of the steel industry needs to cover the data, models, drawings, documents, and value of the asset chain of steel production, the data, models, documents, and value of the product chain, and the liquidity, security, profitability, and risk management information of the capital chain; S1-6: Based on several semantically enhanced historical steel industry data resources, a data resource classification model is constructed using a continuous learning and deep learning fusion algorithm, including the following steps: S1-6-1: Set corresponding real data resource classification labels for several semantically enhanced historical steel industry data resources, obtain several data resource classification samples, and use deep learning algorithms to build an initial data resource classification model; the initial data resource classification model includes an initial model parameter optimization module, an initial graph structure feature extraction module, an initial data resource classification module, and a continuous learning module; The data resource classification model is constructed based on the Improved Fireworks Optimization Algorithm (IFWA)-Graph Convolutional Network (GCN)-Elman-Continuous Learning Algorithm (CLA) algorithm, and the data resource classification model includes an initial model parameter optimization module constructed based on the IFWA algorithm, a graph structure feature extraction module constructed based on the GCN algorithm, a data resource classification module constructed based on the Elman algorithm, and a continuous learning module constructed based on the CLA algorithm, which are sequentially connected; The initial model parameter optimization module optimizes the initial model parameters of the data resource classification model by using the IFWA algorithm to avoid the defects of premature model, overfitting and sensitivity to the initial values ​​of the model, and improve the model accuracy and training efficiency. The initial model parameters include the initial number of neurons, initial weights and thresholds of neurons, and initial learning rate of the GCN network and Elman network. The GCN network propagates features on the graph structure data through operations similar to convolution, extracts node features (position features, image features, etc.) of several segmented images in the graph structure data, and edge features of the positional relationship between nodes, and forms complex, high-dimensional global graph structure features. S1-6-2: Based on the initial model parameter optimization module, with minimizing the model error as the optimization goal, the IFWA algorithm is used to optimize the initial model parameters of the initial data resource classification model to obtain an optimized data resource classification model, including the following steps: S1-6-2-1: Based on the initial model parameter optimization module, the optimization goal is to minimize the model error, and according to the initial model parameters, the individual coding vector format and IFWA population parameters are set; S1-6-2-2: Set the fitness function of the IFWA algorithm according to the optimization goal; The formula is: Where Fit(x) is the fitness function; MSN is the mean square error; S1-6-2-3: Initialize using the Circle chaotic mapping sequence according to the individual coding vector format and IFWA population parameters to generate several initial IFWA individuals of the initial IFWA population; The formula is: In the formula, is the initial IFWA individual of Circle chaos mapping; is the randomly generated initial IFWA individual; It is the IFWA individual indicator; S1-6-2-4: According to the fitness function, the explosion radius, number of sparks and fitness value of the initial IFWA individual are obtained; The formula is: In the formula, For the initial IFWA individuals The number of sparks; is a numerical constant; is the maximum fitness value in the initialized IFWA population; For the initial IFWA individuals The fitness value of is an infinitesimal constant; is the convergence factor; is a positive real number that is not 0; In the formula, For the initial IFWA individuals Explosion radius; Adjusted constant for explosion radius; is the minimum fitness value in the initialized IFWA population; L is the total number of IFWA individuals; In the formula, is the convergence factor; tanh(.) is the hyperbolic tangent function; is the iteration indicator; is the maximum number of iterations; a max , a min are the maximum and minimum values ​​of the convergence factor, respectively; λ is the deceleration rate parameter, is the decreasing cycle parameter, ; The number of sparks determines the number of sub-fireworks produced after each firework explodes, and the explosion radius determines the distribution range of the sparks produced after the firework explodes in the solution space. In the early stage of iteration, a The larger the value of , the smaller the number of sparks of IFWA individuals and the larger the explosion radius, which helps to reduce the computational burden and is more widely distributed, which helps to explore more solution spaces. In the later stages of iteration, a smaller explosion radius helps to perform a fine search in a local area, and a larger number of sparks helps to increase the diversity of the search. S1-6-2-5: Perform fireworks explosion according to the explosion radius, the number of sparks and the fitness value to obtain a number of updated IFWA individuals of the updated IFWA population; The formula is: In the formula, For updated IFWA individuals; A random number between -1 and 1; S1-6-2-6: Use the Gaussian mutation algorithm to perform Gaussian mutation on the initialized IFWA population to generate several Gaussian mutated IFWA individuals of the Gaussian mutated IFWA population; The formula is: In the formula, is the IFWA individual of Gaussian variation; is a Gaussian distributed random number with mean and variance both 1; S1-6-2-7: Use the dynamic reverse learning algorithm to perform dynamic reverse learning on the initialized IFWA population to obtain several reverse IFWA individuals of the reverse IFWA population; The formula is: In the formula, For reverse IFWA individuals; γ is the decreasing inertia coefficient; L max , L min are the maximum and minimum values ​​of the vector space respectively; S1-6-2-8: Obtain the fitness value of each updated IFWA individual, Gaussian mutated IFWA individual and reverse IFWA individual, and take the IFWA individual with the minimum fitness value as the optimal individual; S1-6-2-9: If the number of iterations reaches the maximum number of iterations or the fitness value of the optimal individual meets the requirements, the individual encoding vector of the optimal individual is decoded to obtain the optimal initial model parameters of the initial data resource classification model; S1-6-2-10: Optimize the initial data resource classification model according to the optimal initial model parameters to obtain an optimized data resource classification model; S1-6-3: Use several data resource classification samples to optimize and train the optimized data resource classification model to obtain the optimal data resource classification model and generate several historical data resource classification experiences; S1-6-4: Based on the continuous learning module, using the experience replay mechanism of the continuous learning algorithm, an experience replay pool is set in the continuous learning module of the optimal data resource classification model, and a number of historical data compression experiences are stored in the experience replay pool; S1-6-5: Based on the continuous learning module, the elastic weight connection mechanism of the continuous learning algorithm is used to adjust the initial loss function of the optimal data resource classification model, obtain the adjusted loss function, and obtain the final data resource classification model; When training on real-time data, in order to prevent the data resource classification model from forgetting previously learned experience, an additional penalty term is added to the loss function using the Elastic Weight Consolidation (EWC) method. This penalty term is proportional to the importance measure of the weight and inversely proportional to the amount of change in the weight on the new task. This penalty term ensures that when training on real-time data, the weights that are important for historical experience will not change too much, thereby reducing the risk of catastrophic forgetting. The larger the importance measure, the more restricted the corresponding weight change on real-time data. In this way, the data resource classification model can retain historical experience while learning real-time data; S2: Cloud data center uses the steel industry knowledge graph to semantically enhance a number of real-time steel industry data resources, and obtains a number of semantically enhanced real-time steel industry data resources, including the following steps: S2-1: Cloud data center, collects a number of real-time steel industry data resources, and performs preprocessing to obtain a number of preprocessed real-time steel industry data resources; S2-2: using a named entity and entity relationship extraction model, extracting a number of data named entities of each pre-processed real-time steel industry data resource and data entity relationships between each data named entity and other data named entities; S2-3: Obtain the Euclidean distance between each data named entity and all knowledge named entities in the steel industry knowledge graph, take the knowledge named entity with the closest Euclidean distance as the mapping named entity of the data named entity, and take several knowledge entity relationships of the mapping named entity as the mapping entity relationships of the data named entity; S2-4: Map all mapped named entities to corresponding data named entities in the preprocessed real-time steel industry data resources, and map all mapped entity relationships to data entity relationships corresponding to the data named entities, to obtain semantically enhanced real-time steel industry data resources; S2-5: traverse all pre-processed real-time steel industry data resources to obtain a number of semantically enhanced real-time steel industry data resources; S3: The cloud data center uses the data resource classification model to classify a number of semantically enhanced real-time steel industry data resources, and obtains a number of real-time data resource classification results and an updated data resource classification model, including the following steps: S3-1: Cloud data center, inputs semantically enhanced real-time steel industry data resources into the data resource classification model, and uses the graph structure feature extraction module to extract the corresponding real-time graph structure features; S3-2: According to the real-time graph structure characteristics, use the data resource classification module to classify the data resources, obtain the real-time data resource classification results, and generate the corresponding real-time model error value; S3-4: Obtain corresponding real-time data resource classification experience according to the real-time graph structure characteristics, the real-time model error value, and the real-time model parameters of the data resource classification model; S3-5: Traverse all semantically enhanced real-time steel industry data resources to obtain several real-time data resource classification results and some real-time data resource classification experiences; S3-6: randomly extracting a number of different historical data resource classification experiences from the experience playback pool, and mixing the number of historical data resource classification experiences with the number of real-time data resource classification experiences to obtain a number of mixed data resource classification experiences; By combining historical experience, the data resource classification model can transfer the experience learned from previous tasks to new tasks, which helps improve learning efficiency and reduce the demand for training data for new tasks. Combining historical experience can help prevent the data resource classification model from forgetting the experience learned on historical tasks when training on real-time data. This is a common continuous learning challenge called "catastrophic forgetting". Historical experience can provide a broader view of data distribution, which helps the data resource classification model generalize better on real-time data. S3-7: Based on the experience of classifying several mixed data resources, continuously learn and train the data resource classification model, and use the adjusted loss function to obtain the loss value of the continuous learning and training; S3-8: If the loss value is always less than the loss value threshold, then the updated data resource classification model is obtained; otherwise, a number of historical data resource classification experiences are re-extracted, and continuous learning and training are continued; S4: Cloud data center, which manages the entry of several semantically enhanced real-time steel industry data resources and the corresponding real-time data resource classification results into tables, obtains a real-time steel industry data resource form, uses a data resource map to visualize the real-time steel industry data resource form, and uses a blockchain network to perform distributed storage of the real-time steel industry data resource form, including the following steps: S4-1: Cloud data center, associates the semantically enhanced real-time steel industry data resources with the corresponding real-time data resource classification results to obtain standard real-time steel industry data resources; S4-2: Manage the entry of several standard real-time steel industry data resources into tables to obtain a real-time steel industry data resource form, and use a pre-built data resource map to visualize the real-time steel industry data resource form; like Figure 2 As shown in the figure, the data resource map integrates the relevant databases and knowledge bases of the full life cycle design of the steel industry, realizes the visualization of the row and column data in the real-time steel industry data resource form, and improves the intuitiveness of the real-time steel industry data resources and the classification results of real-time data resources; S4-3: Store the real-time steel industry data resource form in the IPFS system of the blockchain network, obtain the data hash value, and generate a corresponding real-time data storage request; S4-4: Use the Byzantine Fault Tolerance (PBFT) consensus algorithm to reach consensus on real-time data storage requests among several distributed connected data nodes. If the consensus succeeds, proceed to the next step; otherwise, end the data storage. S4-5: Call the smart contract of the blockchain network, generate corresponding real-time transaction data according to the consensus success message and data hash value, and convert the real-time transaction data into real-time data blocks; S4-6: Send the real-time data block to several distributed connected data nodes, and use the data nodes that receive the real-time data block to link the real-time data block to the chain; The use of blockchain networks for distributed storage of real-time steel industry data resources improves the reliability of real-time steel industry data resource management, data security, and traceability.

[0022] Embodiment 2: like Figure 3 As shown, this embodiment provides a standard management system for data resource classification in the steel industry, which is used to implement a standard management method. The system is set in a cloud data center, and the system includes a model building unit, a semantic enhancement unit, a data resource classification unit, and a distributed storage unit connected in sequence; A model building unit, which is used to deploy a blockchain network and build a steel industry knowledge graph and data resource classification model based on a number of historical steel industry data resources and a number of steel industry knowledge using artificial intelligence algorithms; A semantic enhancement unit is used to use the steel industry knowledge graph to perform semantic enhancement on a number of real-time steel industry data resources to obtain a number of semantically enhanced real-time steel industry data resources; A data resource classification unit is used to classify a number of semantically enhanced real-time steel industry data resources using a data resource classification model, and obtain a number of real-time data resource classification results and an updated data resource classification model; The distributed storage unit is used to use the blockchain network to distribute the storage of a number of semantically enhanced real-time steel industry data resources and corresponding real-time data resource classification results.

[0023] The present invention provides a standard management method and system for data resource classification in the steel industry, which covers the full life cycle standard management of data resource collection, classification, governance, table entry and storage in the steel industry, formulates a data resource management process, ensures the standardized processing of data resources in the full life cycle, so as to improve the data resource management level and resource utilization efficiency of steel enterprises, and eliminates the obstacles to data resource management; by constructing a knowledge graph and a data resource classification model for the steel industry, the semantic enhancement of real-time data resources in the steel industry is achieved, thereby improving the accuracy and quality of data processing and analysis; by using a continuous learning algorithm to improve the data resource classification model, it is ensured that the model is continuously trained for real-time data, and the model effectively reduces the forgetting of historical experience when learning new data, thereby improving the generalization ability of the model; an automated data resource classification and storage system is adopted to improve the intelligence level and management efficiency of management, and avoid excessive manual dependence, wherein the deep learning data resource classification model can discover the potential characteristics of data resources in the steel industry and improve the management effect; a decentralized blockchain network is adopted to perform distributed storage on data resources and data resource classification results in the steel industry, and the distributed backup of the blockchain network is utilized to improve the reliability of data storage, and data can be protected from tampering and leakage, thereby ensuring the security and integrity of data.

[0024] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.

Claims

1. A standard management method for data resource classification in the steel industry, characterized by: The steps include: Cloud data center, deploying blockchain network, and using artificial intelligence algorithms to build steel industry knowledge graph and data resource classification model based on several historical steel industry data resources and several steel industry knowledge; The cloud data center uses the steel industry knowledge graph to semantically enhance a number of real-time steel industry data resources, and obtains a number of semantically enhanced real-time steel industry data resources; The cloud data center uses the data resource classification model to classify a number of semantically enhanced real-time steel industry data resources, and obtains a number of real-time data resource classification results and an updated data resource classification model; The cloud data center manages the entry of several semantically enhanced real-time steel industry data resources and the corresponding real-time data resource classification results into tables, obtains real-time steel industry data resource forms, uses data resource maps to visualize the real-time steel industry data resource forms, and uses blockchain networks to distribute and store real-time steel industry data resource forms.

2. A standard management method for data resource classification in the steel industry according to claim 1, characterized in that: The cloud data center deploys a blockchain network and uses artificial intelligence algorithms to build a steel industry knowledge graph and data resource classification model based on several historical steel industry data resources, including the following steps: In the cloud data center, several data servers are connected as data nodes in a distributed manner, and smart contracts and IPFS systems are set up to obtain a blockchain network; Collecting a number of historical steel industry data resources and a number of steel industry knowledge, and preprocessing them to obtain a number of preprocessed historical steel industry data resources and a number of preprocessed steel industry knowledge; Using a pre-trained named entity and entity relationship extraction model, named entity and entity relationship extraction is performed on a number of pre-processed steel industry knowledge to obtain a number of knowledge named entities and knowledge entity relationships between each knowledge named entity and other knowledge named entities; Construct a knowledge graph for the steel industry based on several knowledge named entities and corresponding knowledge entity relationships; Using the steel industry knowledge graph, semantic enhancement is performed on several pre-processed historical steel industry data resources to obtain several semantically enhanced historical steel industry data resources; Based on several semantically enhanced historical steel industry data resources, a data resource classification model is constructed using a fusion algorithm of continuous learning and deep learning.

3. A standard management method for data resource classification in the steel industry according to claim 2, characterized in that: Based on several semantically enhanced historical steel industry data resources, a data resource classification model is constructed using a continuous learning and deep learning fusion algorithm, including the following steps: Set corresponding real data resource classification labels for several semantically enhanced historical steel industry data resources, obtain several data resource classification samples, and use deep learning algorithms to build an initial data resource classification model; Taking minimizing the model error as the optimization goal, the swarm intelligence optimization algorithm is used to optimize the initial model parameters of the initial data resource classification model to obtain an optimized data resource classification model; Using several data resource classification samples, the optimized data resource classification model is optimized and trained to obtain the optimal data resource classification model, and several historical data resource classification experiences are generated; Using the experience replay mechanism of the continuous learning algorithm, an experience replay pool is set up in the optimal data resource classification model, and several historical data compression experiences are stored in the experience replay pool; Using the elastic weight connection mechanism of the continuous learning algorithm, the initial loss function of the optimal data resource classification model is adjusted to obtain the adjusted loss function and the final data resource classification model.

4. A standard management method for data resource classification in the steel industry according to claim 2, characterized in that: The named entity and entity relationship extraction model is built based on the BERT-Double CRF algorithm.

5. A standard management method for data resource classification in the steel industry according to claim 3, characterized in that: The data resource classification model is constructed based on the IFWA-GCN-Elman-CLA algorithm.

6. A standard management method for data resource classification in the steel industry according to claim 4, characterized in that: Using a named entity and entity relationship extraction model, extracting a number of data named entities of each preprocessed real-time steel industry data resource and data entity relationships between each data named entity and other data named entities, including the following steps: Input the preprocessed real-time steel industry data resources into the named entity and entity relationship extraction model, and extract the real-time text features of the preprocessed real-time steel industry data resources; According to the real-time text features, several data named entities of the pre-processed real-time steel industry data resources are extracted; Extracting data entity relationships between each data named entity and other data named entities based on real-time text features and a number of data named entities; All pre-processed real-time steel industry data resources are traversed to obtain a number of data named entities and a number of data entity relationships of each pre-processed real-time steel industry data resource.

7. A standard management method for data resource classification in the steel industry according to claim 6, characterized in that: The cloud data center uses the steel industry knowledge graph to semantically enhance a number of real-time steel industry data resources, and obtains a number of semantically enhanced real-time steel industry data resources, including the following steps: The cloud data center collects a number of real-time steel industry data resources and performs preprocessing to obtain a number of preprocessed real-time steel industry data resources; Using a named entity and entity relationship extraction model, extracting a number of data named entities of each pre-processed real-time steel industry data resource and data entity relationships between each data named entity and other data named entities; Obtain the Euclidean distance between each data named entity and all knowledge named entities in the steel industry knowledge graph, use the knowledge named entity with the closest Euclidean distance as the mapping named entity of the data named entity, and use several knowledge entity relationships of the mapping named entity as the mapping entity relationships of the data named entity; Mapping all mapped named entities to corresponding data named entities in the preprocessed real-time steel industry data resources, and mapping all mapped entity relationships to data entity relationships corresponding to the data named entities, to obtain semantically enhanced real-time steel industry data resources; Traverse all preprocessed real-time steel industry data resources to obtain several semantically enhanced real-time steel industry data resources.

8. A standard management method for data resource classification in the steel industry according to claim 5, characterized in that: The cloud data center uses the data resource classification model to classify a number of semantically enhanced real-time steel industry data resources, and obtains a number of real-time data resource classification results and an updated data resource classification model, including the following steps: The cloud data center inputs the semantically enhanced real-time steel industry data resources into the data resource classification model to extract the corresponding real-time graph structure features; According to the real-time graph structure characteristics, data resources are classified to obtain the real-time data resource classification results and generate the corresponding real-time model error value; According to the real-time graph structure characteristics, the real-time model error value and the real-time model parameters of the data resource classification model, the corresponding real-time data resource classification experience is obtained; Traverse all semantically enhanced real-time steel industry data resources to obtain several real-time data resource classification results and some real-time data resource classification experiences; Randomly extract a number of different historical data resource classification experiences from the experience replay pool, and mix the number of historical data resource classification experiences with a number of real-time data resource classification experiences to obtain a number of mixed data resource classification experiences; Based on the experience of classifying several mixed data resources, the data resource classification model is continuously trained, and the loss value of the continuous learning training is obtained using the adjusted loss function; If the loss value is always less than the loss value threshold, the updated data resource classification model is obtained; otherwise, some historical data resource classification experiences are re-extracted and continuous learning and training is continued.

9. A standard management method for data resource classification in the steel industry according to claim 2, characterized in that: The cloud data center manages the entry of several semantically enhanced real-time steel industry data resources and the corresponding real-time data resource classification results into tables, obtains a real-time steel industry data resource form, uses a data resource map to visualize the real-time steel industry data resource form, and uses a blockchain network to perform distributed storage of the real-time steel industry data resource form, including the following steps: The cloud data center associates the semantically enhanced real-time steel industry data resources with the corresponding real-time data resource classification results to obtain standard real-time steel industry data resources; Manage the entry of several standard real-time steel industry data resources into tables, obtain the real-time steel industry data resource form, and use the pre-built data resource map to visualize the real-time steel industry data resource form; Store the real-time steel industry data resource form in the IPFS system of the blockchain network, obtain the data hash value, and generate the corresponding real-time data storage request; Use the PBFT consensus algorithm to reach consensus on real-time data storage requests among several distributed connected data nodes. If the consensus succeeds, proceed to the next step; otherwise, end the data storage. Call the smart contract of the blockchain network, generate corresponding real-time transaction data according to the consensus success message and data hash value, and convert the real-time transaction data into real-time data blocks; The real-time data block is sent to several distributed connected data nodes, and the real-time data block is linked to the chain using the data nodes that received the real-time data block.

10. A standard management system for data resource classification in the steel industry, used to implement the standard management method as claimed in any one of claims 1 to 9, characterized in that: The system is arranged in a cloud data center, and the system comprises a model building unit, a semantic enhancement unit, a data resource classification unit and a distributed storage unit which are connected in sequence.

Images