A workshop data integration control model optimization method based on graph mining algorithm

Abstract

The application discloses a kind of workshop data integration control model optimization method based on graph mining algorithm, first, obtain workshop equipment data, and carry out pre-processing, define a node for each equipment, and assign unique identifier for each node, and determine the attribute information of each node, and based on this, build workshop equipment relationship network model, finally, based on PageRank algorithm, the importance of node in workshop equipment relationship network model is evaluated, and the constructed workshop equipment relationship network model is optimized using social network analysis algorithm.The scheme of the present application generates the ship intelligent cabin section workshop relationship social network model using graph mining algorithm for the equipment of the entire ship intelligent cabin section workshop, and optimizes it, by deeply mining the relationship between equipment, evaluating the importance of equipment, discovering group relationship, achieving a more in-depth understanding of production process, with real-time and dynamic adaptability, can be updated at any time to adapt to the changes of workshop, at the same time, it can make workshop staff can conveniently analyze the importance of workshop equipment and the dynamic relationship between workshop equipment.

Description

Technical Field

[0001] This invention belongs to the field of real-time data integration, specifically relating to an optimization method for workshop data integration control model based on graph mining algorithm. Background Technology

[0002] The intelligent shipboard workshop involves a large number of devices with complex relationships between them, making clear data integration and understanding difficult using traditional methods. Furthermore, traditional data integration algorithms cannot meet the real-time requirements of workshop production. The operation of some devices may depend on the status and data of other devices; such complex dependencies are difficult to analyze and control effectively using traditional dataset integration methods. The data in the intelligent shipboard workshop may have complex structures, such as graph structures and hierarchical structures, making it difficult for traditional methods to effectively integrate the complex business logic of the workshop and achieve comprehensive intelligent control of the entire production process.

[0003] A review of previous research and findings reveals that the application of graph databases, real-time data integration and processing technologies, and the development of IoT technology have all provided new ideas and methods for solving this problem. Graph databases, such as Neo4j and TigerGraph, can efficiently store and query graph-structured data with complex relationships, providing a more intuitive and efficient solution for the management and analysis of workshop equipment relationships. Meanwhile, real-time data processing technologies, such as Apache Kafka and Apache Flink, can handle high-throughput data streams, meeting the real-time and efficiency requirements of workshop production. The development of IoT technology enables workshop equipment to become more intelligent and interconnected, achieving real-time monitoring and management of equipment status and data through IoT platforms.

[0004] However, although existing research and technologies have solved the problems of data integration and control of workshop equipment to some extent, some challenges still exist. Traditional methods are difficult to meet the requirements of real-time performance and complex dependencies in workshop production, and existing graph databases and real-time processing technologies need to be further integrated to better achieve intelligent control of the workshop production process. Summary of the Invention

[0005] To address the aforementioned problems, the present invention aims to provide an optimization method for workshop data integration control model based on graph mining algorithm. We use graph mining algorithm to improve the traditional workshop data integration algorithm, generating a social network model of the relationship between equipment and personnel in the entire ship intelligent section workshop using graph mining algorithm, and then optimizing it.

[0006] The specific technical solution for achieving the objective of this invention is as follows:

[0007] A method for optimizing a workshop data integration control model based on graph mining algorithms includes the following steps:

[0008] Step 1: Obtain workshop equipment data and perform preprocessing;

[0009] Step 2: Define a node for each device, assign a unique identifier to each node, and determine the attribute information of each node;

[0010] Step 3: Construct a workshop equipment relationship network model;

[0011] Step 4: Evaluate the importance of nodes in the workshop equipment relationship network model based on the PageRank algorithm, and optimize the constructed workshop equipment relationship network model using social network analysis algorithms.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0013] (1) The present invention improves upon the traditional workshop data integration algorithm based on graph mining. It uses graph mining to generate a social network model of relationships between all equipment in the intelligent ship section workshop and optimizes it, resulting in significant progress compared to traditional methods. This model deeply mines the relationships between equipment, assesses the importance of equipment, discovers group relationships, and achieves a more in-depth understanding of the production process. Compared with traditional methods, the graph mining model in this invention has real-time and dynamic adaptability, and can be updated at any time to adapt to changes in the workshop. The graph mining-based decision support system provides operators with more intelligent and accurate decision support. Furthermore, this model can discover hidden relationships, reveal potential equipment dependencies, and enhance the comprehensive understanding of the production process. Its suitability for complex structures allows the graph mining model to handle large-scale, highly connected equipment networks, further improving production efficiency. Therefore, the graph mining-based intelligent ship section workshop data integration model provides deeper and more intelligent support for production control and decision-making, and is expected to achieve significant results in improving efficiency and adapting to complex environments.

[0014] (2) Based on the graph mining algorithm to generate a workshop relationship social network, in order to enable workshop personnel to gain a deeper understanding of the operation of equipment in the workshop, find key equipment, and understand the dependencies between equipment, we use the PageRank algorithm to evaluate the importance of equipment on the basis of the original graph mining algorithm to generate a workshop relationship social network. This allows workshop staff to easily analyze the importance of workshop equipment. At the same time, we add a social network analysis algorithm to analyze the relationships, dependencies, and interactions between equipment in the production process, so that workshop staff can more intelligently and conveniently analyze the dynamic relationships between workshop equipment. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the process of the workshop data integration control model optimization method based on graph mining algorithm of the present invention.

[0016] Figure 2 This is a schematic diagram of the workshop equipment relationship network in an embodiment of the present invention.

[0017] Figure 3 This is a schematic diagram illustrating the implementation process of the PageRank algorithm of this invention.

[0018] Figure 4 This is a schematic diagram illustrating the implementation process of the social network analysis algorithm of the present invention. Detailed Implementation

[0019] Example

[0020] To clearly describe the technical solution and effects achieved by the present invention, the technical solution of the present invention will be clearly explained below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can realize the invention without creative effort. The structure shown in the drawings is not the entirety of the actual structure but only a part of the actual structure. It should be noted that all other embodiments made by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the protection scope of the present invention.

[0021] The following examples are merely illustrative of the invention, and the scope of the invention is not limited to the examples provided. Therefore, any non-essential modifications made by those skilled in the art to the embodiments described above, applied to other embodiments, are still within the scope of protection of this invention. Furthermore, experimental methods not specified in the following examples should be performed according to conventional or manufacturer-recommended conditions. Unless otherwise specified, expressions in the text are used for distinguishing purposes only and have no other meaning.

[0022] Combination Figure 1 A method for optimizing a workshop data integration control model based on graph mining algorithms includes the following steps:

[0023] Step 1: Obtain workshop equipment data and perform preprocessing:

[0024] Collecting workshop equipment data, including diverse sources such as equipment operation logs, sensor data, control commands, and equipment alarm records, ensures a comprehensive understanding of the interactions between workshop equipment. The integrated use of various data sources helps construct a more complete equipment relationship diagram.

[0025] # Define data source

[0026] Data source = ["Device operation log", "Sensor data", "Control commands", "Device alarm records"]

[0027] The collected workshop equipment data is preprocessed to ensure data consistency and usability, including removing redundant information, handling missing values, and correcting outliers to ensure the accuracy of subsequent analysis.

[0028] This embodiment provides a code example for data preprocessing:

[0029] # Code Example - Data Preprocessing

[0030] For each data point in the data source:

[0031] Data = Redundant information removed (data)

[0032] Data = Handling Missing Values ​​(Data)

[0033] Data = Correcting Abnormal Data (Data)

[0034] Step 2: Define a node for each device, assign a unique identifier to each node, and determine the attribute information of each node:

[0035] Based on the preprocessed data, each piece of equipment in the workshop is identified and extracted as a node, and a unique identifier is assigned to each node. The uniqueness of the node can be ensured by using the equipment ID, name or other unique identifier.

[0036] # Pseudocode Example - Device Node Identification

[0037] Device node = {}

[0038] For each record in the data source:

[0039] Device ID = Extract Device ID (Record)

[0040] Device node [Device ID] = Create device node (record);

[0041] For example, the rules for creating IDs are clear and straightforward. We will use three types of equipment—sensors, production machines, and control systems—as examples to illustrate the rules for creating equipment.

[0042] - Device attribute extraction: Extract key attributes for each type of device.

[0043] Sensor: Sensor type and serial number.

[0044] Production machinery: Equipment type, model, and serial number.

[0045] Control system: Equipment type, system type, serial number.

[0046] - Unique encoding generation:

[0047] Sensor ID: The first letter of the sensor type (e.g., T for a temperature sensor) and the serial number.

[0048] Production Machine ID: First letter of equipment type, first letter of model, serial number.

[0049] Control system ID: First letter of device type, first letter of system type, serial number.

[0050] -Identifier assignment:

[0051] Sensor example: Temperature sensor (T), serial number "001" with device ID "T-001".

[0052] Example of a production machine: The device ID of a machine with device type "M", model "123", and serial number "002" is "M-123-002".

[0053] Example of a control system: The device ID of a device with device type "C", system type "PLC", and serial number "003" is "C-PLC-003".

[0054] In this way, through the rules for creating device IDs, each device has a unique and highly readable identifier, ensuring accurate identification in the workshop equipment relationship network model. Based on the accurate definition of nodes, we added attribute information to each device node to further enrich its characteristics.

[0055] The attribute information of each node includes, but is not limited to:

[0056] Equipment types: Sensors, production machines, control systems, etc.

[0057] Operating status: Normal, Under maintenance, Fault, etc.

[0058] Performance indicators: such as the working efficiency of production machines and the measurement accuracy of sensors.

[0059] These attribute information endow each device node with more detailed characteristics, making the workshop equipment relationship network model not just a graph of connections, but also a complex network that fully describes the characteristics of the equipment. This will provide a more comprehensive and in-depth foundation for subsequent graph mining and analysis.

[0060] Next, using information from the data source, the relationships between nodes are identified, i.e.:

[0061] First, it's necessary to determine the relationship type. In workshop equipment data collection, a clear definition of the relationship type helps to accurately depict the connections between equipment. Relationship types can include physical connections, data transmission, energy supply, etc. Clearly defining the relationship type facilitates the subsequent establishment of relationship edges.

[0062] # Define relation types

[0063] Relationship type = {physical connection, data transfer, energy supply}

[0064] Identify the relationships between nodes by using information from the data source, which may include physical connections, data transmission channels, and control commands between devices.

[0065] Based on the identification results, relationship edges are established to represent various relationships between nodes, thereby constructing a node relationship graph:

[0066] Let r be each record in the data source, containing source device ID S, target device ID T, and relation type R. Define the set of relation edges between devices as E, then:

[0067] E={(S,T,R)||r∈DataSource}

[0068] Where (S,T,R) represents the relation edge from source device S to target device T, and the relation type is R;

[0069] The constructed set of relation edges E is used to form the device relation graph.

[0070] Finally, the attribute information of nodes and relationship edges is extracted from the data, including the device type, operating status, and performance indicators of device nodes; while the attribute information of relationship edges can include data transmission speed, connection strength, etc.

[0071] # Code Example - Attribute Information Extraction

[0072] For each device ID in the device node:

[0073] Attribute = Extract Device Attributes (Device ID)

[0074] Device node [Device ID].Update(Attributes)

[0075] For each edge in relation edge:

[0076] Attribute = Extract edge attributes (edge)

[0077] edge.update(property)

[0078] Through the above steps, we collected and processed workshop equipment data, identified equipment nodes and relationship edges, and extracted relevant attribute information, laying a solid foundation for the subsequent construction of a workshop equipment relationship network model.

[0079] Step 3: Construct a workshop equipment relationship network model:

[0080] First, an adjacency list is constructed for each device node to record its connections with other devices. This data structure represents the connections between devices. Each adjacency list entry contains the connected device ID and possible edge attribute information, such as relationship type and data transmission speed. This process is accomplished by traversing the actual connections between devices, ensuring that the graph model accurately reflects the actual relationships between devices in the workshop.

[0081] Create an adjacency list:

[0082] # Initialize nodes and their adjacency lists

[0083] node = {

[0084] "T-001": {"Type": "Temperature_Sensor", "Status": "active", "Precision": 0.5, "Adjacency List": []},

[0085] "M-123-002": {"Type": "Production_Machine", "Status": "idle", "Efficiency": 90, "Adjacency List": []},

[0086] "C-PLC-003": {"Type": "Control_System", "Status": "active", "System Type":"PLC", "Adjacency List": []}

[0087] }

[0088] Secondly, by establishing connections between devices through their collaborative relationships, each adjacency list entry represents an edge. The relationships between devices are comprehensively described through the connected device IDs and detailed edge attribute information.

[0089] # Define the connection relationships between devices

[0090] Connection relationship = [

[0091] ("T-001", "M-123-002", {"Relationship": "monitors", "Data Transfer": True})

[0092] ("M-123-002", "C-PLC-003", {"Relationship": "controlled_by", "Data Transmission":True}) ]

[0094] # Update adjacency list

[0095] For source node, target node, edge attribute in connection relationship:

[0096] Node[source node]["adjacency list"].append((target node, edge attributes))

[0097] # Display the adjacency list

[0098] Print(node["T-001"]["Adjacency List"]) # Output: [("M-123-002", {"Relationships": "monitors", "Data Transfer": True})]

[0099] Print(node["M-123-002"]["Adjacency List"]) # Output: [("C-PLC-003", {"Relationship": "controlled_by", "Data Transmission": True})]

[0100] Finally, using the information from the device nodes and the adjacency list, a workshop equipment relationship network model is constructed. That is, the adjacency list reflects the relationships between nodes, and the combination of nodes and the adjacency list constitutes the workshop equipment relationship network model. The approximate topology diagram of the workshop network relationship model in this embodiment is shown below. Figure 2 As shown.

[0101] # Initialize the workshop network model

[0102] Workshop network model = {

[0103] "Node": Node,

[0104] "Adjacency list": {node: node[node]["adjacency list"] for node in node}

[0105] }

[0106] # Display workshop network model

[0107] Print (Workshop Network Model)

[0108] The generated workshop equipment relationship network model is a graph structure based on adjacency lists. In this graph structure, each node represents a piece of equipment in the workshop, and each equipment's adjacency list records the connections between that equipment and other equipment. Each entry in the adjacency list includes the connected equipment ID and possible edge attribute information. This graph-based structure clearly presents the relationships between workshop equipment, enabling us to more comprehensively understand the collaborative relationships, data transmission relationships, and control relationships between equipment.

[0109] Visualizing the generated model using graph visualization tools helps to intuitively understand the collaborative relationships between devices and further discover potential patterns and rules. We can use the open-source software Tableau to visualize the generated workshop equipment relationship network model.

[0110] # Import device relationship data into Tableau

[0111] Tableau.ImportData(Workshop Network Model)

[0112] # Select chart type and attribute mapping

[0113] Tableau.SelectChartType("Network Graph")

[0114] Tableau.MapAttributes(node="device ID", edge="relationship type")

[0115] # Visualized Device Relationship Network Model

[0116] Tableau.Visualize()

[0117] # Explore deeper through interactive features

[0118] Tableau.EnableInteraction()

[0119] When using Tableau to visualize workshop equipment relationship networks, the first step is to import equipment relationship data. Through Tableau's graphical interface, we can choose appropriate chart types and attribute mappings to represent equipment as nodes and relationships as edges, using different colors, shapes, or sizes to present equipment characteristics and relationship strength. Leveraging Tableau's interactive features, users can explore the network structure in depth by hovering and filtering, monitoring data changes in real time, quickly identifying collaborative relationships between equipment, and building a more comprehensive visual presentation through dashboards. This approach not only provides an intuitive network diagram but also allows users to dynamically adjust and analyze it as needed. Through these steps, we establish a clear and detailed workshop equipment relationship network model, laying a solid foundation for subsequent graph mining and in-depth analysis.

[0120] Step 4: Evaluate the importance of nodes in the workshop equipment relationship network model based on the PageRank algorithm, and optimize the constructed workshop equipment relationship network model using social network analysis algorithms.

[0121] Based on the workshop equipment relationship network model constructed using graph mining, in order to enable workshop personnel to have a deeper understanding of the production process, discover hidden relationships, reveal potential equipment dependencies, and improve the overall understanding of the production process, PageRank algorithm and social network analysis algorithm can be introduced to enhance the completeness and intelligence of the model.

[0122] Among them, combined Figure 3 The importance of nodes is assessed based on the PageRank algorithm. The PageRank algorithm determines the relative importance of nodes in the network by analyzing the links between them. In a workshop equipment relationship network, equipment nodes can be viewed as pages, and the PageRank algorithm can be used to identify key equipment nodes. These devices may play important roles in the production process, specifically:

[0123] The weights of the N nodes in the constructed workshop equipment relationship network model are initialized, that is, for any node i, its initial weight W(i) is:

[0124] W(i)=1 / N

[0125] Thus, each element in vector W represents the initial weight of each node and is assigned to each node in the network graph.

[0126] In a specific workshop equipment relationship network, the initial weight of the equipment can be appropriately adjusted to increase the importance of the equipment from the beginning. Factors affecting the initial weight of workshop equipment include equipment type, equipment status, equipment performance indicators, equipment historical data, business requirements, etc.

[0127] A damping factor is introduced, and the PageRank algorithm is used to iterate the PageRank value PR(i) of the node:

[0128]

[0129] Where d is the damping factor, which is usually 0.85; M(i) represents the set of nodes pointing to node i, that is, the set of devices that are directly related to device i; if device j is related to device i, then j belongs to M(i); L(j) represents the out-degree of node j, that is, the number of devices that are related to device j, reflecting the influence of device j; the initial PageRank value PR(i) of each node is the initial weight W(i);

[0130] In the workshop equipment relationship network, the damping factor in the PageRank algorithm mainly simulates behavior in the actual production environment, taking into account the randomness and non-linked navigation that may exist in the workflow. This means that the collaborative relationship between equipment is not only determined by direct physical connections and control relationships, but may also be affected by random factors, such as changes in equipment status and temporary task changes.

[0131] In each iteration, a new PageRank value is calculated for each node based on the current PageRank value, and the stored PageRank vector is updated. The iteration continues until the change in the PageRank value is less than a set threshold ϵ, i.e., the condition is met.

[0132] ϵ

[0133] in and Let represent the PageRank values ​​of node i in the t-th and t+1-th iterations, respectively.

[0134] The commonly used convergence threshold ϵ is set as follows: Or even smaller, to ensure the accuracy of the results. This condition indicates that the algorithm has converged and the PageRank value has stabilized.

[0135] Once the algorithm converges, the final PageRank value PR(i) of node i represents the importance of device i in the workshop equipment relationship network, reflecting the device's connection relationships and influence in the network.

[0136] During the iteration process, the damping factor d adjusts the PageRank calculation in each iteration so that the PageRank value of a node not only depends on the contribution of its directly linked nodes, but also includes a certain global randomness factor, thus more accurately reflecting the actual importance of the device in the network.

[0137] By calculating the importance of equipment in the workshop, we can identify critical equipment, optimize performance, predict failures, allocate resources rationally, and optimize production scheduling. This helps improve the availability and intelligence of the entire workshop equipment relationship network model, enabling more efficient and reliable production operations.

[0138] Combination Figure 4 This study optimizes the constructed workshop equipment relationship network model based on Social Network Analysis (SNA) algorithms, using SNA to delve deeper into the social relationships between equipment nodes. Social Network Analysis can reveal the close connections and mutual influences between equipment, helping to understand the details of equipment dependencies. By identifying key indicators such as social groups and degree centrality among equipment nodes, a more comprehensive understanding of the interactions between equipment can be achieved.

[0139] Overall modularity (Q) is an important indicator for evaluating the rationality of community partitioning in a network. It assesses the quality of the partition by calculating the density of connections within a community and the sparsity between communities. Specifically, each device node is first initialized as an independent community, forming an initial community partition, and the rationality of the partition is then determined using overall modularity (Q).

[0140]

[0141] Where: Q represents modularity, This indicates whether there is a connection between node i and node j. and Let be the degree of nodes i and j (i.e., the total number of connections), and m be the total number of edges in the workshop equipment relationship network. This is an indicator function; if nodes i and j belong to the same community, the value is 1, otherwise it is 0.

[0142] During the iteration, to maximize the overall modularity Q value, nodes with higher degrees (i.e., more connections) are selected for movement; that is, nodes are moved to communities that can improve the overall modularity Q value.

[0143] Specifically, in each iteration, the initial value of the overall modularity Q is first calculated based on the current community division; then, a node i with a high degree is selected and attempted to be moved to a different community in order to calculate the modularity increment after joining the community. If moving node i improves modularity, then move it to the new community and update its Q-value; otherwise, keep it in the original community. Modularity increment. The formula is:

[0144]

[0145] Where, ∑ in This represents the sum of the number of edges within the community. ∑ represents the number of connections between node i and other nodes in the community, while ∑ tot It is the total degree of the community;

[0146] The iteration continues until no node movement can further improve the overall modularity Q-value. Specifically, in each iteration, for each node i, the modularity increment after moving it to a different community is calculated. If it is found that movement can improve modularity (i.e. If the value is greater than 0, then node i is retained in the new community, and modularity Q is updated. If the move no longer improves modularity (i.e., ...), then the node i is retained in the new community, and modularity Q is updated. If the value is less than or equal to 0, then node i remains in the original community.

[0147] This process is repeated for all nodes until each node can no longer improve modularity, at which point the iteration stops. At this point, the community partitioning reaches a relatively stable state, and the resulting community structure clearly reveals the social relationships between devices, helping to deepen the understanding of dependencies and interactions between them. Through this algorithm, the final output community partitioning structure can reveal a set of closely related devices, thereby helping to identify social relationships between devices and providing structured results for more in-depth network analysis.

[0148] The application of social network analysis algorithms in workshop equipment relationship network models can deeply explore the social relationships between equipment nodes, revealing the connections and influences between devices. By identifying key indicators, a comprehensive understanding of equipment interactions can be achieved, and equipment nodes can be divided into communities, which helps to optimize layout and configuration, improve production efficiency and collaboration. This further optimizes workshop equipment management and production processes, enhancing overall operational efficiency.

[0149] Based on the PageRank algorithm and social network analysis, significant improvements have been made to the entire system. Firstly, these two algorithms allow us to accurately identify key nodes in the workshop equipment relationship network. These nodes may play a crucial role in the production process, and their performance and status directly affect the operation of the entire production system. By identifying these key nodes, workshop managers can gain a deeper understanding of the critical aspects of the production process, enabling targeted resource allocation and equipment maintenance, thereby improving production efficiency and product quality.

[0150] Secondly, graph visualization tools are used to display the results of the PageRank algorithm and social network analysis, providing workshop personnel with an intuitive way to present data. Through visualization, workshop personnel can clearly see key nodes, social groups, and important relationships between equipment. This intuitive data presentation method enables workshop personnel to gain a deeper understanding of the dependencies and mutual influences between equipment, thereby more effectively planning production processes, optimizing resource allocation, and making timely decisions to address various challenges in production.

[0151] Finally, the results of key node identification and visualization were applied to optimize the decision support system, further enhancing its functionality and performance. Through node importance analysis and key equipment identification, the system can provide more accurate decision support for workshop managers. The decision support system not only monitors equipment status in real time and identifies anomalies, but also provides targeted production optimization suggestions, helping workshop managers respond quickly to changes in the production process and optimize production plans to maximize production efficiency.

[0152] In summary, the identification and visualization of key nodes based on the PageRank algorithm and social network analysis, along with the application of optimized decision support systems, provide powerful tools and support for workshop production management. This helps to improve production efficiency, reduce costs, optimize resource utilization, and ultimately achieve intelligent and automated production processes.

[0153] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0154] Step 1: Obtain workshop equipment data and perform preprocessing;

[0155] Step 2: Define a node for each device, assign a unique identifier to each node, and determine the attribute information of each node;

[0156] Step 3: Construct a workshop equipment relationship network model;

[0157] Step 4: Evaluate the importance of nodes in the workshop equipment relationship network model based on the PageRank algorithm, and optimize the constructed workshop equipment relationship network model using social network analysis algorithms.

[0158] The present invention also provides a computer-storable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the following steps:

[0159] Step 1: Obtain workshop equipment data and perform preprocessing;

[0160] Step 2: Define a node for each device, assign a unique identifier to each node, and determine the attribute information of each node;

[0161] Step 3: Construct a workshop equipment relationship network model;

[0162] Step 4: Evaluate the importance of nodes in the workshop equipment relationship network model based on the PageRank algorithm, and optimize the constructed workshop equipment relationship network model using social network analysis algorithms.

[0163] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for optimizing a workshop data integration control model based on graph mining algorithms, characterized in that, Includes the following steps: Step 1: Obtain workshop equipment data and perform preprocessing; Step 2: Define a node for each device, assign a unique identifier to each node, and determine the attribute information of each node; Step 3: Construct a workshop equipment relationship network model: First, an adjacency table is built for each device node to record the connection relationship between the device node and other devices. Each adjacency table entry contains the connected device ID and possible edge attribute information, such as relationship type and data transmission speed. Secondly, by establishing connections between devices through their collaborative relationships, each adjacency list entry represents an edge. The relationships between devices are comprehensively described through the connected device IDs and detailed edge attribute information. Finally, using the information of equipment nodes and adjacency lists, a workshop equipment relationship network model is constructed. That is, the adjacency list reflects the association between nodes, and the combination of nodes and adjacency lists constitutes the workshop equipment relationship network model. Step 4: Evaluate the importance of nodes in the workshop equipment relationship network model based on the PageRank algorithm, and optimize the constructed workshop equipment relationship network model using social network analysis algorithms.

2. The method for optimizing the workshop data integration control model based on graph mining algorithm according to claim 1, characterized in that, The acquisition of workshop equipment data in step 1 specifically involves: Collect workshop equipment data, including equipment operation logs, sensor data, control commands, and equipment alarm records; The collected workshop equipment data is preprocessed to ensure data consistency and usability, including removing redundant information, handling missing values, and correcting abnormal data.

3. The method for optimizing the workshop data integration control model based on graph mining algorithm according to claim 1, characterized in that, The determination of node attribute information in step 2 specifically includes: Based on the preprocessed data, each piece of equipment in the workshop is identified and extracted as a node, and a unique identifier is assigned to each node; Identify the relationships between nodes using information from the data source; Extract attribute information of nodes and relation edges from the data.

4. The workshop data integration control model optimization method based on graph mining algorithm according to claim 3, characterized in that, The relationship edges between the identified nodes are specifically as follows: Identify the relationships between nodes by using information from the data source, including physical connections, data transmission channels, and control commands between devices; Based on the identification results, relationship edges are established to represent various relationships between nodes, thereby constructing a node relationship graph: Let r be each record in the data source, containing source device ID S, target device ID T, and relation type R. Define the set of relation edges between devices as E, then: E={(S,T,R)||r∈DataSource} Where (S,T,R) represents the relation edge from source device S to target device T, and the relation type is R; The constructed set of relation edges E is used to form the device relation graph.

5. The workshop data integration control model optimization method based on graph mining algorithm according to claim 3, characterized in that, The attribute information includes the device type, operating status, and performance indicators of the device node.

6. The method for optimizing the workshop data integration control model based on graph mining algorithm according to claim 1, characterized in that, The node importance assessment based on the PageRank algorithm in step 4 is specifically as follows: The weights of the N nodes in the constructed workshop equipment relationship network model are initialized, that is, for any node i, its initial weight W(i) is: W(i)=1 / N A damping factor is introduced, and the PageRank algorithm is used to iterate the PageRank value PR(i) of the node: ； Where d is the damping factor, M(i) represents the set of nodes pointing to node i, that is, the set of devices that are directly related to device i. If device j is related to device i, then j belongs to M(i); L(j) represents the out-degree of node j, that is, the number of devices that are related to device j, reflecting the influence of device j. The initial PageRank value PR(i) of each node is the initial weight W(i). In each iteration, a new PageRank value is calculated for each node based on the current PageRank value, and the stored PageRank vector is updated. The iteration continues until the change in the PageRank value is less than a set threshold ϵ, i.e., the condition is met. ϵ; in and Let represent the PageRank values ​​of node i in the t-th and t+1-th iterations, respectively.

7. The method for optimizing the workshop data integration control model based on graph mining algorithm according to claim 1, characterized in that, The optimization of the constructed workshop equipment relationship network model based on the social network analysis algorithm in step 4 is specifically as follows: First, each device node is divided into independent communities to form the initial community, and the rationality of the division is judged by the overall modularity Q: ； Where: Q represents modularity, This indicates whether there is a connection between node i and node j. and Let be the degree of nodes i and j, i.e., the total number of connections, and m be the total number of edges in the workshop equipment relationship network. This is an indicator function; if nodes i and j belong to the same community, the value is 1, otherwise it is 0. During the iteration, nodes with higher degrees are selected for movement, that is, nodes are moved to communities that can improve the overall modularity Q-value. The specific optimization criterion is to calculate the modularity increment ΔQ in each iteration and select the community that maximizes the modularity Q-value. The increment formula is as follows: ； Where, ∑ in This represents the sum of the number of edges within the community. ∑ represents the number of connections between node i and other nodes in the community, while ∑ tot It is the total degree of the community; During the iteration process, the algorithm continuously tries to move nodes to different communities to optimize the overall community division. If it finds that the connection strength between some nodes is high and the modularity increases, the algorithm will merge these nodes into the same community. This process is repeated until the modularity Q can no longer be improved, indicating that the algorithm has reached a stable state and the optimization of the workshop equipment relationship network model is completed.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

9. A computer-storable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images