Industrial internet service knowledge graph analysis platform
By building a knowledge graph in the industrial Internet and adopting an analysis model based on graph neural networks, the problem of not being able to fully utilize multi-source heterogeneous data in the existing technology is solved, efficient processing and in-depth analysis of industrial Internet data is achieved, and the accuracy and production efficiency of analysis results are improved.
Patent Information
- Application Number
- CN202510206709.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-25
- Publication Date
- 2025-06-10
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology cannot fully utilize multi-source heterogeneous data in the industrial Internet, resulting in inaccurate and comprehensive analysis results. In addition, traditional machine learning algorithms have limited ability to extract and represent data features, making it difficult to mine deep information in the data.
It provides an industrial Internet service knowledge graph analysis platform, through data collection, preprocessing, knowledge graph construction and management, and analysis model based on graph neural network, it makes full use of the topological structure information in the knowledge graph to explore deeper knowledge correlation and potential laws.
It realizes efficient processing and in-depth analysis of industrial Internet data, improves the accuracy of fault diagnosis, optimizes production processes, improves equipment maintenance efficiency, and provides more accurate analysis results and decision-making suggestions.
Smart Images
Figure CN120124731A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of industrial Internet, and specifically relates to an industrial Internet service knowledge graph analysis platform. Background Art
[0002] In the field of industrial Internet, with the increase in the number of devices and the complexity of the production process, higher requirements are put forward for equipment fault diagnosis, production optimization, equipment maintenance, and process improvement. Traditional analysis methods often rely on a single data source and simple statistical analysis, and cannot make full use of the multi-source heterogeneous data in the industrial Internet, resulting in inaccurate and incomplete analysis results. Moreover, most of the existing analysis models adopt traditional machine learning algorithms, with limited ability to extract and represent data features, and it is difficult to mine the deep information in the data.
[0003] Therefore, those skilled in the art have proposed an industrial Internet service knowledge graph analysis platform to solve the problems raised in the background art. Summary of the Invention
[0004] In order to solve the above technical problems, the present invention provides an industrial Internet service knowledge graph analysis platform to solve the problems in the prior art that traditional analysis methods often rely on a single data source and simple statistical analysis, cannot make full use of the multi-source heterogeneous data in the industrial Internet, resulting in inaccurate and incomplete analysis results, and most of the existing analysis models adopt traditional machine learning algorithms, with limited ability to extract and represent data features, and it is difficult to mine the deep information in the data.
[0005] An industrial Internet service knowledge graph analysis platform includes:
[0006] A data acquisition module for collecting data from various devices, systems, and sensors in the industrial Internet through various methods, where the data includes structured data, semi-structured data, and unstructured data;
[0007] A data preprocessing module for performing preprocessing operations of cleaning, transforming, and integrating the collected data, removing noise data and duplicate data, and unifying the data format;
[0008] A knowledge graph construction module for extracting entities, relationships, and attributes from the preprocessed data by using natural language processing and information extraction technologies, and constructing an industrial Internet service knowledge graph;
[0009] A knowledge graph management module for performing management operations of storing, querying, updating, and maintaining the constructed knowledge graph to ensure the integrity and consistency of the knowledge graph;
[0010] The analysis application module provides various analysis application functions based on the knowledge graph, and the analysis application module adopts an analysis model based on a graph neural network.
[0011] Preferably, the structured data collected by the data collection module includes equipment operation parameters and production order data, the semi-structured data includes log files, and the unstructured data includes equipment maintenance manuals and process documents.
[0012] Preferably, the data preprocessing module uses a data cleaning algorithm to remove noise data and duplicate data, uses a data conversion algorithm to convert the data into a unified format, and uses a data integration algorithm to integrate data from different data sources.
[0013] Preferably, the knowledge graph construction module uses a named entity recognition algorithm to extract entities from text data, uses a relation extraction algorithm to extract relations from text data, and uses an attribute extraction algorithm to extract attributes from text data. The entities include equipment, processes, products, and personnel, and the relations include the associations between equipment, the sequence of process steps, and the corresponding relationships between products and equipment.
[0014] Preferably, the knowledge graph management module uses a graph database for storage and management, uses the storage and query functions of the graph database to manage the knowledge graph, and uses the update and maintenance functions of the graph database to update and maintain the knowledge graph.
[0015] Preferably, the propagation and aggregation process of the analysis model based on the graph neural network is represented by the following formula: $h_i^{(k + 1)}=\sigma\left(\sum_{j\in N(i)}\frac{1}{\sqrt{|\mathcal{N}
[0016] (i)|\sqrt{|\mathcal{N}(j)|}}W^{(k)}h_j^{(k)}\right)$
[0017] where $h_i^{(k)}$ represents the representation vector of node $i$ in the $k$-th module, $N(i)$ represents the set of neighbor nodes of node $i$, $W^{(k)}$ represents the weight matrix of the $k$-th module, $\sigma$ represents the activation function, and $|\mathcal{N}(i)|$ represents the number of neighbor nodes of node $i$.
[0018] Preferably, the analysis model based on the graph neural network is trained using a supervised learning method, and the model is trained using labeled data, and the labeled data includes fault diagnosis results, production optimization suggestions, and equipment maintenance plans.
[0019] Preferably, when the graph neural network-based analysis model is applied, it is used for fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendation.
[0020] Preferably, the analysis application module provides analysis application functions such as fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendation, and displays the analysis results in a visual manner.
[0021] Preferably, the platform can run in real time in the industrial Internet environment, and monitor and analyze the equipment status and production process in real time.
[0022] Compared with the prior art, the present invention has the following beneficial effects:
[0023] 1. The present invention adopts a graph neural network-based analysis model, which can make full use of the topological structure information in the knowledge graph, excavate deeper knowledge associations and potential laws, provide more accurate analysis results and decision-making suggestions for industrial Internet services. By constructing a comprehensive industrial Internet service knowledge graph and adopting a graph neural network-based analysis model, it can more accurately diagnose the causes of equipment failures, predict the probability of failures, take maintenance measures in advance, reduce equipment downtime, and improve production efficiency.
[0024] 2. The present invention analyzes the data in the production process based on the knowledge graph, can excavate the bottleneck links and potential optimization points in the production process, provide a decision-making basis for the optimization of the production process, and improve production quality and efficiency.
[0025] 3. The present invention formulates a personalized equipment maintenance plan according to the equipment information and operation data in the knowledge graph, and then can reasonably arrange the maintenance time and resources, improve the equipment maintenance efficiency, and reduce the maintenance cost. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 It is a system diagram of an industrial Internet service knowledge graph analysis platform. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0027] The following further describes in detail the embodiments of the present invention in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present invention, but cannot be used to limit the scope of the present invention.
[0028] Embodiment 1
[0029] As shown in the Figure 1 accompanying drawings, this embodiment provides an industrial Internet service knowledge graph analysis platform, including
[0030] The data acquisition module is used to collect data from various devices, systems, and sensors in the industrial Internet through multiple methods. The data includes structured data, semi-structured data, and unstructured data, providing sufficient data support for subsequent data preprocessing and knowledge graph construction, and improving the integrity and accuracy of the data;
[0031] The data preprocessing module is used to perform preprocessing operations of cleaning, transforming, and integrating the collected data, removing noise data and duplicate data, unifying the data format, improving the quality and usability of the data, and providing a high-quality data foundation for knowledge graph construction;
[0032] The knowledge graph construction module is used to extract entities, relationships, and attributes from the preprocessed data by using natural language processing and information extraction technologies, construct an industrial Internet service knowledge graph, and integrate the scattered data into a knowledge graph with semantic relationships, providing rich knowledge support for subsequent analysis and applications;
[0033] The knowledge graph management module is used to perform management operations of storing, querying, updating, and maintaining the constructed knowledge graph, ensuring the integrity and consistency of the knowledge graph, improving the usability and reliability of the knowledge graph, and providing stable knowledge support for analysis and applications;
[0034] The analysis and application module provides multiple analysis and application functions based on the knowledge graph. The analysis and application functions include fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendation. The analysis and application module adopts an analysis model based on graph neural networks, which can provide more accurate analysis results and decision-making suggestions for industrial Internet services, improve the production efficiency and quality of enterprises, reduce the production costs of enterprises, and enhance the market competitiveness of enterprises.
[0035] From the above, we can see that the data collected by the data acquisition module is directly passed to the data preprocessing module for processing, and the data processed by the data preprocessing module is then passed to the knowledge graph construction module to ensure the integrity and accuracy of the data and provide high-quality data support for the subsequent knowledge graph construction. The knowledge graph constructed by the knowledge graph construction module is directly passed to the knowledge graph management module for storage and management. The knowledge graph management module stores, queries, updates, maintains and other operations on the knowledge graph to ensure the integrity and consistency of the knowledge graph and provide stable knowledge support for the analysis application module. The knowledge graph stored and managed by the knowledge graph management module is directly provided to the analysis application module for analysis. The analysis application module performs analysis based on the knowledge graph, generates analysis results and decision-making recommendations, ensures the accuracy and reliability of the analysis results, and provides strong decision-making support for the company's production operations. This platform can process structured, semi-structured and unstructured data. Compared with the existing technology that can only process a single data type, it has more comprehensive data processing capabilities and can better meet the processing needs of complex data in the industrial Internet. By adopting an analysis model based on graph neural networks, it can make full use of the topological structure information in the knowledge graph and explore deeper knowledge associations and potential laws. Compared with the existing technology, the analysis accuracy is higher and it can provide more accurate analysis results and decision-making recommendations for industrial Internet services.
[0036] Embodiment 2
[0037] As attached Figure 1 As shown, this embodiment is basically the same as the previous embodiment, with the difference that the structured data collected by the data acquisition module include equipment operating parameters and production order data, the semi-structured data include log files, and the unstructured data include equipment maintenance manuals and process documents. It can collect various types of data in the industrial Internet in a variety of ways, provide sufficient data support for subsequent data preprocessing and knowledge graph construction, and improve the integrity and accuracy of the data.
[0038] Specifically, the data preprocessing module uses a data cleaning algorithm to remove noise data and duplicate data, uses a data conversion algorithm to convert data into a unified format, and uses a data integration algorithm to integrate data from different data sources to improve data quality and availability, providing a high-quality data foundation for knowledge graph construction.
[0039] Furthermore, the knowledge graph construction module uses a named entity recognition algorithm to extract entities from text data, a relationship extraction algorithm to extract relationships from text data, and an attribute extraction algorithm to extract attributes from text data. Entities include equipment, processes, products, and personnel. Relationships include the association between equipment, the sequence of process steps, and the correspondence between products and equipment. Natural language processing, information extraction and other technologies are used to extract entities, relationships, and attributes from preprocessed data to construct an industrial Internet service knowledge graph, integrating scattered data into a knowledge graph with semantic relationships to provide rich knowledge support for subsequent analysis applications.
[0040] Furthermore, the knowledge graph management module adopts a graph database for storage and management, uses the storage and query functions of the graph database to manage the knowledge graph, uses the update and maintenance functions of the graph database to update and maintain the knowledge graph, and performs storage, query, update, maintenance and other management operations on the constructed knowledge graph to ensure the integrity and consistency of the knowledge graph, improve the availability and reliability of the knowledge graph, and provide stable knowledge support for analytical applications.
[0041] From the above, it can be seen that the data collected by the data acquisition module is directly passed to the data preprocessing module for processing, and the data processed by the data preprocessing module is then passed to the knowledge graph construction module. The knowledge graph constructed by the knowledge graph construction module is directly passed to the knowledge graph management module for storage and management. The knowledge graph management module stores, queries, updates, maintains and other operations on the knowledge graph. This platform can process structured, semi-structured and unstructured data. Compared with the existing technology that can only process a single data type, it has more comprehensive data processing capabilities and can better meet the processing needs of complex data in the industrial Internet. The analysis model based on graph neural network can make full use of the topological structure information in the knowledge graph and explore deeper knowledge associations and potential laws. Compared with the existing technology, the analysis accuracy is higher and it can provide more accurate analysis results and decision-making recommendations for industrial Internet services.
[0042] Embodiment 3
[0043] As attached Figure 1 As shown, this embodiment is basically the same as the previous embodiment, except that the propagation and aggregation process of the analysis model based on the graph neural network is represented by the following formula: $h_i^{(k+1)}=\sigma\left(\sum_{j\in N(i)}\frac{1}{\sqrt{|\mathcal{N}
[0044] (i)|\sqrt{|\mathcal{N}(j)|}}W^{(k)}h_j^{(k)}\right)$
[0045] Among them, $h_i^{(k)}$ represents the representation vector of node $i$ in module $k$, $N(i)$ represents the set of neighbor nodes of node $i$, $W^{(k)}$ represents the weight matrix of the $k$-th module, $\sigma$ represents the activation function, and $|\mathcal{N}(i)|$ represents the number of neighbor nodes of node $i$. This formula updates the representation vector of the node by aggregating the information of neighbor nodes and combining the features of the node itself. This mechanism enables each node to learn the feature information of its neighbor nodes, thereby better capturing the topological relationship in the graph structure.
[0046] Specifically, the analysis model based on the graph neural network is trained using the supervised learning method. The model is trained using labeled data, and the labeled data includes fault diagnosis results, production optimization suggestions, and equipment maintenance plans. During the training process, the analysis model updates the weights by minimizing the loss function, such as cross-entropy loss or mean squared error loss, thereby improving the prediction accuracy of the model.
[0047] Furthermore, when the analysis model based on the graph neural network is applied, it is used for fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendation. Specifically, it includes: using the model to analyze the equipment nodes in the knowledge graph to predict the fault probability of the equipment. By analyzing the equipment nodes in the knowledge graph, the model can learn the associations and fault patterns between equipment, thereby realizing early warning of potential faults; using the model to analyze the process nodes in the knowledge graph to optimize the process flow. By analyzing the process nodes, the model can optimize the production process, discover bottleneck links in the production process, and provide optimization suggestions, thereby improving production efficiency; using the model to analyze the equipment nodes in the knowledge graph to formulate equipment maintenance plans. According to the operating status and fault prediction results of the equipment, personalized equipment maintenance plans are generated, and the maintenance time and resources are reasonably arranged; using the model to analyze the process nodes in the knowledge graph to recommend process improvement plans. By analyzing the process nodes, improvement plans are recommended to improve the process level and product quality.
[0048] Furthermore, the analysis application module provides analysis application functions for fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendation, and displays the analysis results in a visual way. The visual display not only facilitates the user to intuitively understand the analysis results, but also can quickly locate problems and take corresponding measures.
[0049] Furthermore, the platform can run in real time in the industrial Internet environment, monitor and analyze the equipment status and production process in real time. By collecting equipment operating parameters, production order data, etc. in real time, the platform can promptly detect equipment failures and problems in the production process, and perform real-time analysis and prediction through an analysis model based on graph neural networks.
[0050] From the above, we can see that this platform realizes efficient processing and in-depth analysis of industrial Internet data by constructing an industrial Internet service knowledge graph and combining it with an analysis model based on a graph neural network. It can improve the accuracy of fault diagnosis, optimize production processes, and improve equipment maintenance efficiency. The graph neural network can make full use of the topological structure information in the knowledge graph to mine the associations and failure modes between equipment, thereby improving the accuracy of fault diagnosis. By analyzing process nodes, the model can discover bottlenecks in the production process and provide optimization suggestions to improve production efficiency. According to the operating status of the equipment and fault prediction results, it can generate personalized equipment maintenance plans, reasonably arrange maintenance time and resources, and integrate and display various types of knowledge in the industrial Internet in the form of a knowledge graph, which is convenient for knowledge sharing and inheritance among internal personnel of the enterprise.
[0051] The standard parts used in the present invention can all be purchased from the market, and the special-shaped parts can be customized according to the description and the drawings. The specific connection methods of each part adopt conventional means such as mature bolts, rivets, welding, etc. in the prior art. The machinery, parts and equipment all adopt conventional models in the prior art, and the circuit connection adopts the conventional connection method in the prior art, which will not be described in detail here. The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.
[0052] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.
[0053] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0054] In the present invention, unless otherwise clearly defined or limited, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature may be that the first feature is directly above or obliquely above the second feature, or simply means that the horizontal height of the first feature is higher than that of the second feature. The first feature being "under", "beneath" and "underneath" the second feature may be that the first feature is directly below or obliquely below the second feature, or simply means that the horizontal height of the first feature is less than that of the second feature.
[0055] In the description of this specification, the description of reference terms such as "an embodiment", "some embodiments", "examples", "specific examples" or "some examples", etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not have to be directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.
[0056] In the drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved, and other structures can refer to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other.
[0057] Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.
Claims
1. An industrial Internet service knowledge graph analysis platform, characterized by: include, A data collection module is used to collect data from various devices, systems, and sensors in the industrial Internet in a variety of ways, and the data includes structured data, semi-structured data, and unstructured data; The data preprocessing module is used to clean, convert, and integrate the collected data, remove noise data, duplicate data, and unify the data format; The knowledge graph construction module is used to extract entities, relationships and attributes from pre-processed data using natural language processing and information extraction technologies to build an industrial Internet service knowledge graph; The knowledge graph management module is used to store, query, update, and maintain the constructed knowledge graph to ensure the integrity and consistency of the knowledge graph; The analysis application module provides a variety of analysis application functions based on the knowledge graph, and the analysis application module adopts an analysis model based on a graph neural network.
2. An industrial Internet service knowledge graph analysis platform as claimed in claim 1, characterized in that: The structured data collected by the data collection module include equipment operating parameters and production order data, the semi-structured data include log files, and the unstructured data include equipment maintenance manuals and process documents.
3. An industrial Internet service knowledge graph analysis platform as claimed in claim 2, characterized in that: The data preprocessing module uses a data cleaning algorithm to remove noise data and duplicate data, uses a data conversion algorithm to convert data into a unified format, and uses a data integration algorithm to integrate data from different data sources.
4. An industrial Internet service knowledge graph analysis platform as claimed in claim 3, characterized in that: The knowledge graph construction module uses a named entity recognition algorithm to extract entities from text data, uses a relationship extraction algorithm to extract relationships from text data, and uses an attribute extraction algorithm to extract attributes from text data. The entities include equipment, processes, products, and personnel, and the relationships include the association between equipment, the sequence of process steps, and the correspondence between products and equipment.
5. An industrial Internet service knowledge graph analysis platform as claimed in claim 4, characterized in that: The knowledge graph management module adopts a graph database for storage and management, uses the storage and query functions of the graph database to manage the knowledge graph, and uses the update and maintenance functions of the graph database to update and maintain the knowledge graph.
6. An industrial Internet service knowledge graph analysis platform as claimed in claim 5, characterized in that: The propagation and aggregation process of the graph neural network-based analysis model is represented by the following formula: $h_i^{(k+1)}=\sigma\left(\sum_{j\in N(i)}\frac{1}{\sqrt{|\mathcal{N} (i)|\sqrt{|\mathcal{N}(j)|}}W^{(k)}h_j^{(k)}\right)$ Among them, $h_i^{(k)}$ represents the representation vector of node $i$ in the $k$ module, $N(i)$ represents the set of neighbor nodes of node $i$, $W^{(k)}$ represents the weight matrix of the $k$th module, $\sigma$ represents the activation function, and $|\mathcal{N}(i)|$ represents the number of neighbor nodes of node $i$.
7. An industrial Internet service knowledge graph analysis platform as claimed in claim 6, characterized in that: The graph neural network-based analysis model is trained using a supervised learning method and uses labeled data to train the model. The labeled data includes fault diagnosis results, production optimization suggestions, and equipment maintenance plans.
8. An industrial Internet service knowledge graph analysis platform as claimed in claim 7, characterized in that: The graph neural network-based analysis model, when applied, is used for fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendations.
9. An industrial Internet service knowledge graph analysis platform as claimed in claim 8, characterized in that: The analysis application module provides analysis application functions such as fault diagnosis and prediction, production optimization suggestions, equipment maintenance plan formulation, and process improvement plan recommendation, and displays the analysis results in a visual manner.
10. An industrial Internet service knowledge graph analysis platform as claimed in claim 9, characterized in that: The platform can run in real time in an industrial Internet environment and conduct real-time monitoring and analysis of equipment status and production processes.