Electromechanical equipment fault prediction method based on dynamic graph structure embedded model

The timing knowledge graph of electromechanical equipment is constructed through dynamic graph structure embedding model, which solves the problem of data type diversity and time information screening in the prior art, and improves the accuracy and efficiency of fault prediction.

CN119003706AActive Publication Date: 2024-11-22CHINA JILIANG UNIV

Patent Information

Application Number
CN202411480204.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-23
Publication Date
2024-11-22
Estimated Expiration
2044-10-23

AI Technical Summary

Technical Problem

The prior art is difficult to effectively process multiple data types and quickly filter out time information in the fault prediction of electromechanical equipment, resulting in insufficient accuracy of fault prediction.

Method used

Using a method based on a dynamic graph structure embedding model, by constructing a timing knowledge graph of electromechanical equipment, learning node classification information representation, capturing time evolution information, and outputting hidden state vectors for fault prediction. The method includes building a parallel structural information model, a dynamic graph convolutional embedding model, and fractional function calculation to achieve fault prediction.

Benefits of technology

It realizes a deep understanding and feature extraction of heterogeneous data of electromechanical equipment, improves the accuracy and efficiency of fault prediction, and can effectively process the time information in structured and unstructured data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119003706A_ABST
    Figure CN119003706A_ABST
Patent Text Reader

Abstract

The invention discloses an electromechanical equipment fault prediction method based on a dynamic graph structure embedding model, and belongs to the field of electromechanical equipment fault prediction and knowledge graph fault prediction. The method comprises the following steps: firstly, realizing parallel structure information modeling of electromechanical equipment through an ontology construction method, and extracting a tetrad from structured data and unstructured data to construct an electromechanical equipment time sequence knowledge graph; secondly, learning information representation used for node classification, and generating a feature vector for each related node; then capturing time evolution information of electromechanical equipment operation from the vector sequence; and finally, outputting the hidden state vector of the required time period to realize the prediction of the electromechanical equipment time sequence knowledge graph. According to the method, the time sequence knowledge graph is built specially for heterogeneous data of the electromechanical equipment for reasoning, and fault prediction is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of electromechanical equipment fault prediction and knowledge graph fault prediction, and specifically relates to a method for fault prediction of electromechanical equipment by using temporal knowledge graph ontology modeling and dynamic graph structure embedding model. Background Art

[0002] The innovation of Industry 4.0 has led to the rapid development of monitoring technology and the generation of a large amount of data, which has promoted the development of complex artificial intelligence algorithms for analyzing big data. A large amount of heterogeneous data has made it possible to predict the faults of electromechanical equipment. At present, how to deal with a large amount of heterogeneous data is a difficult problem that needs to be solved urgently. The concept of knowledge graph was first proposed by Google in 2012. The core idea of ​​knowledge graph is to represent information as entities and their relationships, thereby forming a structured data network. This concept has since been widely used in various fields, including but not limited to search engines, question-answering systems, recommendation systems, fault prediction, etc.

[0003] The application of knowledge graphs on electromechanical equipment can explore the potential relationships between heterogeneous data, thereby realizing fault prediction, which can not only ensure the normal operation of the system, but also improve the reliability of the system. Someone proposed a method for constructing domain knowledge graphs for complex mechanical equipment and combined it with graph neural networks to realize fault prediction. Someone proposed a neural network-based named entity recognition model for coal mine equipment maintenance and constructed a knowledge graph for fault prediction of coal mine equipment.

[0004] Neural ODEs are a method that uses a neural network structure to parameterize solutions to ordinary differential equations. Since it was proposed in 2018, this method has received widespread attention in the field of dynamic system modeling. Some people have applied Neural ODEs to supervised learning and time series prediction of image datasets. Although Neural ODEs are used in this method, the embedding of knowledge graphs is not considered for prediction. Some people have embedded the nodes of the knowledge graph into the trajectory of the ordinary differential equation and constructed a neural network to realize the application of ordinary differential equations in knowledge graph embedding.

[0005] Although the above methods adjust the graph embedding structure for specific problems or directly obtain fault knowledge in the target domain, their research has made certain contributions in the field of fault prediction. However, in the field of electromechanical equipment fault prediction, the current challenges include: (1) Most neural network models only process one data type. Since the data of electromechanical equipment includes text data, sequence data, etc., it is difficult to achieve accurate fault prediction by relying on only a single type of data. (2) When processing the knowledge graph of electromechanical equipment containing time information, traditional graph structure embedding models often find it difficult to quickly filter out key information from huge data. Summary of the invention

[0006] In order to overcome the shortcomings of the existing technology, the present invention proposes a method for predicting electromechanical equipment faults based on a dynamic graph structure embedding model. First, the parallel structural information modeling of electromechanical equipment is realized through the ontology construction method, and quadruples are extracted from structured data and unstructured data to construct a time series knowledge graph of electromechanical equipment. Secondly, the information representation for node classification is learned, and a feature vector is generated for each related node. Then, the time evolution information of the operation of the electromechanical equipment is captured from the vector sequence; finally, the hidden state vector of the required time period is output to realize the prediction of the time series knowledge graph of the electromechanical equipment; this method is specifically for heterogeneous data of electromechanical equipment to construct a time series knowledge graph for reasoning and realize fault prediction.

[0007] In order to solve the above technical problems, the present invention proposes a method for predicting mechanical and electrical equipment faults based on a dynamic graph structure embedding model, the method comprising the following steps: The first step is to build a time series knowledge graph of electromechanical equipment, that is, to achieve parallel structural information modeling of electromechanical equipment through ontology construction method, and to extract quadruples from structured data and unstructured data to build a time series knowledge graph of electromechanical equipment; The second step is to use the dynamic graph convolution embedding model to realize the dynamic embedding of the electromechanical equipment operation and maintenance time series knowledge graph. The process is as follows: Step (2.1) uses a dynamic graph convolutional model to obtain the tail entity embedding vector of the knowledge graph at query time; Step (2.2) uses the dynamic graph convolutional model in step (2.1) to obtain the head entity embedding vector of the knowledge graph at query time; The third step is to obtain the optimal fault prediction result through the score function. The process is as follows: Step (3.1) uses the score function to calculate the scores of all quadruplets accompanying the candidate entity; Step (3.2) calculates the ranking of the prediction results as a basis for comparing the model effects; Step (3.3) selects the object with the highest score as the fault prediction result of the model.

[0008] Further, the process of the first step is as follows: Step (1.1) Ontology structure construction, the electromechanical equipment timing knowledge graph ETKG ontology is defined as follows: ; Among them, E represents the set of entities; R represents the set of relationships between these entities; T represents the set of time segments corresponding to entities and relationships. E includes three major categories: C represents the system and components of the equipment; F represents the failure of the equipment; M represents the maintenance of the equipment; Step (1.2) data collection; Step (1.3) Data preprocessing: Use ETL technology to preprocess the data from the previous step, remove invalid data, convert all raw data into data schemas, and store them uniformly in a relational database. ETL technology is used to extract data from one or more sources and load it into the target database after conversion. Step (1.4) entity and relationship extraction, automatically extracting entities and relationships from large amounts of structured and unstructured data.

[0009] Preferably, the process of step (1.2) is: Step (1.2.1) collects structured operating data in the electromechanical equipment control cabinet, including vibration signals, fault codes, operating data, electrical parameters, temperature and humidity; uses TCP / IP communication protocol to remotely obtain real-time operating data of electromechanical equipment in sensors; all sensors are connected to devices and gateways via serial ports, and data communication is carried out through Modbus and RS485 protocols; Step (1.2.2) collects unstructured operation and maintenance data in the electromechanical equipment IoT system, including repair records, maintenance records, fault causes and fault phenomena; and uses crawler technology to obtain unstructured data on fault information in the electromechanical equipment field from the electromechanical equipment IoT system.

[0010] More preferably, the process of step (1.4) is: Step (1.4.1) for structured vibration data Continuous wavelet transform is used to extract entities and relations; Step (1.4.2) uses Word2vec, Bi-LSTM, and CRF technologies to extract entities and relationships for unstructured text data; Step (1.4.3) classifies entities according to the defined ontology, draws the electromechanical equipment timing knowledge graph in the graph database based on the acquired entities and relationships, and effectively queries and updates the entities and relationships of the graph database in combination with the graph database language.

[0011] More preferably, the process of step (1.4.1) is as follows: Step (1.4.1.1) Vibration data After continuous wavelet transform, ; ; in, is a wavelet basis function with two parameters a and b, Represents the coefficients of wavelet transform, which are obtained by using wavelet basis functions Vibration data Perform integral transformation; represents the conjugate complex root of the wavelet basis function, is the scale parameter, is the translation parameter; ∫ is the integral operation, used in mathematics and statistics; Step (1.4.1.2) calculates the energy spectrum, ; Among them, the energy spectrum The energy distribution at different scales is described to identify the main frequency components; Step (1.4.1.3) extreme point detection realizes fault entity extraction; Through detection The local extreme point of the signal is used to determine the sudden events or abnormal vibrations in the signal, and the entity is extracted according to the ontology classification in the first step; Step (1.4.1.4) calculates the co-occurrence matrix to achieve relationship extraction, ; Construct the co-occurrence matrix of wavelet coefficients at different scales , analyze the relationship between different frequency components.

[0012] More preferably, the process of step (1.4.2) is as follows: Step (1.4.2.1) Text data embedding: Text data needs to be vectorized, and the corresponding character sequence is mapped to a word vector sequence. Through the word embedding method Word2vec, the word embedding lookup table is updated through continuous training and learning, and finally the input text is matched through the lookup table and converted into the corresponding word vector sequence; Step (1.4.2.2) The text embedding vector is processed by the Bi-LSTM layer; First, the multiple initial parameters used by the Bi-LSTM layer are randomly initialized, and the text embedding vector sequence output by the embedding layer is input into the forward LSTM layer and the reverse LSTM layer model to learn the context information of the sequence vector. Finally, the output vectors in the two directions are concatenated as the output sequence of the Bi-LSTM layer, and the dropout technology is introduced to prevent overfitting. Step (1.4.2.3) The output vector of the Bi-LSTM layer is processed by the CRF layer; The CRF layer considers the relationship between adjacent labels to obtain the global optimal sequence, models the dependencies between entities based on the output of the Bi-LSTM layer, and selects the sequence with the highest probability from the label space; Step (1.4.2.4) calculates the loss function of the model output to obtain the optimal result; The output of the CRF layer is used as the final label sequence of the model output, and compared with the actual labeled sequence to calculate the Loss function. ; in, It represents the probability of outputting sequence Y when given input sequence X. Represents a given input sequence and tag sequence The score function when , log is the logarithmic operation, exp is the natural exponential operation; By continuously adjusting the parameters of each layer to minimize the Loss value, the optimal recognition model is obtained.

[0013] Even more preferably, the process of step (1.4.3) is as follows: Step (1.4.3.1) structured data format conversion; Export the structured data processed in step (1.4.1) from the relational database into csv format. To express the temporal characteristics of process state data, convert the traditional triple form into Expanded to a quadruple with timing information , where t represents the time information when the triple is formed. However, the current graph database only supports triple storage, so the timing information is converted and embedded in the relationship between entities to achieve the establishment of the electromechanical equipment timing knowledge graph based on the graph database;

[0014] Step (1.4.3.2) unstructured data format conversion; Convert the entities and entity relationships extracted in step (1.4.2) into csv file format. Since the extracted data are all text, convert the csv file encoding format to UTF-8 encoding; Step (1.4.3.3) imports the csv file data into the graph database.

[0015] Furthermore, in step (2.1), a dynamic graph convolution model is used to obtain The tail entity embedding vector of the knowledge graph , the process is: Step (2.1.1) The tail entity is embedded based on the traditional graph convolution model. ; in, is the adjacency matrix, is the tail entity at time t The hidden state vector at layer k+1 is is the entity in the subgraph at time t that is related to the tail entity Head entity connected by relationship r The collection of , is the normalization constant, Represents the head entity at level k The hidden state vector of is the diagonal relationship weight matrix when it is k layers, Represents the hidden state vector embedded in the initial state, and the activation function uses ReLU, is a trainable parameter matrix; Step (2.1.2) improves the traditional graph convolution model based on the Neural Ordinary Differential Equation; In step (2.2), the dynamic graph convolution model in step (2.1) is used to obtain The embedding vector of the head entity of the knowledge graph .

[0016] Preferably, the process of step (2.1.2) is as follows: Step (2.1.2.1) decomposes the propagation depth k of the traditional graph convolution into the integration time and step length The combination of , the Neural ODE representation of graph convolution is realized by adding a variable integration step size dt, ; in, is the hidden state vector of the tail entity at time t, is the hidden state vector embedded in the initial state of the tail entity, For the head entity, the range is 0 to The hidden state vector at time t, is the diagonal relationship weight matrix; Step (2.1.2.2) uses Taylor expansion to convert the ordinary differential equation into a differential equation form. ; in, is the hidden state vector of the head entity at time t. Taylor expansion approximates the function with a polynomial near a certain point, which is used to approximate the solution of complex functions in differential equations. is the identity matrix; Step (2.1.2.3) , a set of parameters Θ is used to characterize the entire continuous time aggregation process, and the complete tail entity embedding vector is obtained by summing up, ; in, It is a black-box ODE solver used to solve neural ordinary differential equations.

[0017] Furthermore, in step (3.1), the score function is used to calculate the scores of all quadruplets accompanying the candidate entity, ; in, For query time, represents transpose, is the set of candidate entities, and ; In step (3.2), the ranking of the prediction results is calculated As a basis for comparing the model effects, ; In step (3.3), the object with the highest score is selected as the fault prediction result of the model. ; Among them, arg max means finding The largest , is the final predicted fault entity.

[0018] The beneficial effects of the present invention are mainly manifested in: 1. Different methods are used to process the structured data and unstructured data of electromechanical equipment to provide data support for the construction of knowledge graphs.

[0019] 2. Define the ontology structure of the electromechanical equipment timing knowledge graph, and use corresponding methods to extract the entities, relationships, and time of the electromechanical equipment data according to the ontology structure to form a quadruple to realize the construction of the electromechanical equipment timing knowledge graph.

[0020] 3. Combine the neural ordinary differential equations with graph convolutional neural networks to build a dynamic graph structure embedding model, process the topological structure of the knowledge graph continuously in time and embed it into a hidden state vector, so as to achieve deep understanding and feature extraction of electromechanical equipment data. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the ontology structure of the electromechanical equipment timing knowledge graph defined in the present invention taking an elevator as an example. DETAILED DESCRIPTION

[0022] The present invention will be further described below in conjunction with the accompanying drawings.

[0023] Reference Figure 1 , a method for predicting mechanical and electrical equipment faults based on a dynamic graph structure embedding model, the method comprising the following steps: The first step is to build a knowledge graph of electromechanical equipment timing. The process is as follows: Step (1.1) ontology structure construction, such as Figure 1 As shown, the electromechanical equipment timing knowledge graph (ETKG) ontology is defined as follows: ; Among them, E represents the set of entities; R represents the set of relationships between these entities; T represents the set of time segments corresponding to entities and relationships. E includes three major categories: C represents the system and components of the equipment; F represents the failure of the equipment; M represents the maintenance of the equipment; Among them, the ontology structure is to define the standard terms and their hierarchical structures related to the operation of electromechanical equipment and identify the implicit associations between these terms. It is a commonly used method in the construction of knowledge graphs. Step (1.2) Data collection, the process is: Step (1.2.1) collects structured operation data in the electromechanical equipment control cabinet, including vibration signals, fault codes, operation data, electrical parameters, temperature and humidity; uses TCP / IP communication protocol to remotely obtain real-time operation data of electromechanical equipment in the sensor. Table 1 gives the corresponding data collection information using the elevator as an example; Table 1

[0024] Among them, all sensors, devices and gateways are connected by serial port, and data communication is carried out through Modbus and RS485 protocols; Step (1.2.2) collects unstructured operation and maintenance data in the electromechanical equipment IoT system, including repair records, maintenance records, fault causes, and fault phenomena; Use crawler technology to obtain unstructured data on fault information in the field of electromechanical equipment from the electromechanical equipment IoT system; Among them, Web Crawling, also known as web spider or web robot, is an automated program used to browse and collect Internet data; Step (1.3) Data preprocessing Use ETL (Extract, Transform, Load) technology to pre-process the collected data, remove invalid data, convert all raw data into data schemas, and store them uniformly in a relational database. ETL technology is used to extract data from one or more sources and load it into the target database after transformation; Among them, ETL technology is a common data integration technology used to extract data from one or more sources and load it into the target database or data warehouse after transformation; Step (1.4) Entity and relationship extraction: automatically extract entities and relationships from a large amount of structured and unstructured data; the process is: Step (1.4.1) for structured vibration data Continuous wavelet transform is used to extract entities and relationships. The process is as follows: Step (1.4.1.1) Vibration data After continuous wavelet transform, ; ; in, is a wavelet basis function with two parameters a and b, Represents the coefficients of wavelet transform, which are obtained by using wavelet basis functions Vibration data Perform integral transformation; represents the conjugate complex root of the wavelet basis function, is the scale parameter, is the translation parameter; ∫ is the integral operation, used in mathematics and statistics; Step (1.4.1.2) calculates the energy spectrum, ; Among them, the energy spectrum The energy distribution at different scales is described to identify the main frequency components; Step (1.4.1.3) extreme point detection realizes fault entity extraction; Through detection The local extreme point of the signal is used to determine the sudden events or abnormal vibrations in the signal, and the entity is extracted according to the ontology classification in the first step; Step (1.4.1.4) calculates the co-occurrence matrix to achieve relationship extraction, ; Construct the co-occurrence matrix of wavelet coefficients at different scales , analyze the relationship between different frequency components, such as Figure 1 As shown; Step (1.4.2) uses Word2vec (Word to Vector), Bi-LSTM (Bidirectional Long Short-Term Memory), and CRF (Conditional Random Field) technologies to extract entities and relationships for unstructured text data. The process is as follows: Step (1.4.2.1) Text data embedding: The text data needs to be vectorized and the corresponding character sequence is mapped into a word vector sequence. The present invention uses the word embedding method Word2vec to continuously train and learn to update the word embedding lookup table, and finally matches the input text through the lookup table to convert it into a corresponding word vector sequence. Among them, Word2vec is a commonly used algorithm for natural language processing. It generates word vectors by training a neural network model, maps words into a continuous vector space, and captures the semantic and grammatical relationships between words. Step (1.4.2.2) The text embedding vector is processed by the Bi-LSTM layer; First, the multiple initial parameters used by the Bi-LSTM layer are randomly initialized, and the text embedding vector sequence output by the embedding layer is input into the forward LSTM layer and the reverse LSTM layer model to learn the context information of the sequence vector. Finally, the output vectors in the two directions are concatenated as the output sequence of the Bi-LSTM layer, and the dropout technology is introduced to prevent overfitting. Among them, dropout is a commonly used regularization technique that prevents overfitting by randomly discarding neurons. LSTM is a commonly used recurrent neural network that controls the storage and reading of information by introducing memory units and gating mechanisms, thereby maintaining and updating information over a long time span. Bi-LSTM is an extended version of LSTM that captures the bidirectional context information of sequence data by using two LSTMs in the forward and reverse directions of the time series. Step (1.4.2.3) The output vector of the Bi-LSTM layer is processed by the CRF layer; The CRF layer considers the relationship between adjacent labels to obtain the global optimal sequence, models the dependencies between entities based on the output of the Bi-LSTM layer, and selects the sequence with the highest probability from the label space; Among them, CRF is a probabilistic graphical model for labeling and segmenting sequence data, which can consider contextual information and global features to improve labeling accuracy; Step (1.4.2.4) calculates the loss function of the model output to obtain the optimal result; The output of the CRF layer is used as the final label sequence of the model output, and compared with the actual labeled sequence to calculate the Loss function. ; in, It represents the probability of outputting sequence Y when given input sequence X. Represents a given input sequence and tag sequence The score function when , log is the logarithmic operation, and exp is the natural exponential operation.

[0025] By continuously adjusting the parameters of each layer to minimize the Loss value, the optimal recognition model is obtained; Step (1.4.3) follows Figure 1 The ontology defined in the Graph Database classifies entities. The graph database language can use Cypher, etc. to achieve effective query and update of the graph database. The graph database can use Neo4j, which is a high-performance graph database that uses a graph structure to store data and is suitable for processing highly connected data. Finally, the electromechanical equipment timing knowledge graph is drawn in Neo4j; the process is as follows:

[0026] Step (1.4.3.1): Structured data format conversion; Export the structured data processed in step (1.4.1) from the relational database (such as Oracle) into csv format. To express the temporal characteristics of process state data, convert the traditional triple form into Expanded to a quadruple with timing information , where t represents the time information when the triple is established. However, currently graph databases (such as Neo4j) only support triple storage, so the present invention converts the timing information and embeds it into the relationship between entities to achieve the establishment of a timing knowledge graph of electromechanical equipment based on the graph database;

[0027] A triplet in data structures and databases refers to a set of three elements, which is often used to represent entity relationships in relational databases. Step (1.4.3.2) unstructured data format conversion; Convert the entities and entity relationships extracted in step (1.4.2) into a csv file format. Since the extracted data are all text, the present invention converts the csv file encoding format into UTF-8 encoding; Step (1.4.3.3) Import the quad data into Neo4j; Use graph database languages ​​(such as Cypher) to import csv files. For example, first store the organized entity and relationship csv files in the import folder under the Neo4j root path, and use Cypher's "LOAD CSV" statement to import the quadruple csv file into Neo4j;

[0028] The second step is to use the dynamic graph convolution embedding model to realize the dynamic embedding of the electromechanical equipment operation and maintenance time series knowledge graph. The process is as follows: Step (2.1) uses the dynamic graph convolutional model to obtain The tail entity embedding vector of the knowledge graph , the process is: Step (2.1.1) The tail entity is embedded based on the traditional graph convolution model. ; in, is the adjoint matrix, is the tail entity at time t The hidden state vector at layer k+1 is is the entity in the subgraph at time t that is related to the tail entity Head entity connected by relationship r The collection of , is the normalization constant, Represents the head entity at level k The hidden state vector of is the diagonal relationship weight matrix when it is k layers, Represents the hidden state vector embedded in the initial state, and the activation function uses ReLU, is a trainable parameter matrix; Among them, traditional graph convolution is a commonly used method for knowledge graph embedding; Step (2.1.2) improves the traditional graph convolution model based on the Neural Ordinary Differential Equation. The process is as follows: Step (2.1.2.1) decomposes the propagation depth k of the traditional graph convolution into the integration time and step length The combination of , the Neural ODE representation of graph convolution is realized by adding a variable integration step size dt, ; in, is the hidden state vector of the tail entity at time t, is the hidden state vector embedded in the initial state of the tail entity, For the head entity, the range is 0 to The hidden state vector at time t, is the diagonal relationship weight matrix.

[0029] Step (2.1.2.2) uses Taylor expansion to convert the ordinary differential equation into a differential equation form. ; in, is the hidden state vector of the head entity at time t. Taylor expansion approximates the function with a polynomial near a certain point, which is used to approximate the solution of complex functions in differential equations. is the identity matrix.

[0030] Step (2.1.2.3) The present invention is carried out by making , a set of parameters Θ is used to characterize the entire continuous time aggregation process, and the complete tail entity embedding vector is obtained by summing up, ; in, It is a black-box ODE solver used to solve neural ordinary differential equations. Step (2.2) uses the dynamic graph convolutional model in step (2.1) to obtain The embedding vector of the head entity of the knowledge graph ; The third step is to obtain the optimal fault prediction result through the score function. The process is as follows: Step (3.1) uses the score function to calculate the scores of all quadruplets accompanying the candidate entity, ; in, For query time, represents transpose, is the set of candidate entities, and ; Step (3.2) calculates the ranking of the prediction results As a basis for comparing the model effects, ; Step (3.3) selects the object with the highest score as the fault prediction result of the model. ; Among them, arg max means finding The largest , is the final predicted fault entity.

[0031] The implementation process of the electromechanical equipment fault prediction method based on the dynamic graph structure embedding model of this embodiment includes the following steps: Step 1: Provide 10 knowledge graph reasoning models: TransE: This method is an embedding-based static knowledge graph reasoning model that learns vector representations of entities and relations by minimizing the difference between the vectors of the head and tail entities of a triple.

[0032] ComplEx: This method is a static knowledge graph embedding model that uses complex embeddings (real and imaginary parts) to represent entities and relations.

[0033] DistMult: This method is a simple but effective static knowledge graph embedding model that captures the relationships between entities by using symmetric trilinear products.

[0034] TTransE: This method is an extension of the TransE model and is a temporal knowledge graph inference method. By taking the time constraint as part of the triple, it can effectively predict past or future relations.

[0035] TNTComplEx: This method is an extension of the ComplEx model and is a temporal knowledge graph inference method. By introducing the time factor into the embedding of entities and relationships, the model can effectively predict relationships and events in the knowledge graph that change over time.

[0036] TA-DistMult: This method is an extension of the DistMult model and is a temporal knowledge graph inference method.

[0037] It introduces a time factor for each triple so that the embedding can reflect the changes of entities and relations over time.

[0038] RE-NET: This method is a model specially designed for temporal knowledge graph extrapolation, which captures the evolution of entities and relations over time by integrating recurrent neural networks.

[0039] RE-GCN: This method combines the advantages of graph convolutional networks and recurrent neural networks. It can capture the structural information and time evolution information of the knowledge graph at the same time, and is a temporal knowledge graph extrapolation model.

[0040] CENET: This method is a temporal knowledge graph event prediction model based on a historical contrastive learning framework. It can effectively distinguish potential entities matching the query and is a temporal knowledge graph extrapolation model.

[0041] ETKGFP: The time series knowledge graph extrapolation method of the present invention integrates the neural ordinary differential equation and the graph convolutional neural network, dynamically embeds the time series knowledge graph and realizes the fault prediction of electromechanical equipment.

[0042] Step 2: Experimental Dataset The experimental data set is provided by the electromechanical equipment timing knowledge graph constructed by the ontology construction method proposed in this invention, which contains 12792 entities and 154 semantic relations.

[0043] Step 3: Define evaluation metrics In order to verify the prediction performance of each model, the present invention uses the following evaluation indicators for experiments: Hit@n: This metric measures the proportion of correct predictions that are within the top n highest rankings given by the model among all test samples; smaller n values ​​emphasize the accuracy of the model in predicting the most relevant options; larger n values ​​take more into account the comprehensiveness and diversity of the model. ; in, represents the ranking number of the quadruple, and N is the total number of candidate entities. In this embodiment, Hit@1, Hit@3, and Hit@10 are selected for evaluation;

[0044] MRR: An internationally accepted mechanism for evaluating matching algorithms. The calculation principle is that the higher the correct result is ranked, the higher the score is. If there is no correct result, the score is 0. The formula is as follows: ; Step 4: Analyze and compare the results Refer to Table 2, which shows the comparison results of MRR, Hit@1, Hit@3 and Hit@10 for the fault category entity prediction task. Table 2

[0045] It can be observed that the three static reasoning models TransE, DistMult and ComplEx are not as good as the time series model in terms of effect, because the static model cannot consider the time information and the dependency of entities between different time snapshots. In the case of the time series reasoning model, the purpose of the internalization models TTtransE, TA-Distmult and TNTComplEx is to complete the missing relations and entities, rather than predict new events, so they show worse performance than the extrapolation model. Among the extrapolation models, CENET has the best effect in fault entity prediction among the baseline models, because CENET, like the model of the present invention, also makes predictions based on the historical interactions between entities, rather than being highly dependent on the periodicity or reproducibility of events like RE-GCN and RE-NET. The contrastive learning method adopted by CENET focuses on the similarities and differences between entities, which makes it completely inferior to ETKGFP, which introduces the Neural Ordinary Differential Equation, in terms of dynamic modeling ability. The method proposed by the present invention achieved the best results in all four indicators, which were 41.54%, 32.84%, 48.86% and 57.65% respectively.

[0046] The contents described in the embodiments of this specification are merely enumerations of implementation forms of the inventive concept and are for illustrative purposes only. The protection scope of the present invention should not be considered to be limited to the specific forms described in this embodiment, and the protection scope of the present invention also extends to equivalent technical means that can be thought of by ordinary technicians in this field based on the inventive concept.

Claims

1. A method for predicting mechanical and electrical equipment faults based on a dynamic graph structure embedding model, characterized in that: The method comprises the following steps: The first step is to build a time series knowledge graph of electromechanical equipment, that is, to achieve parallel structural information modeling of electromechanical equipment through ontology construction method, and to extract quadruples from structured data and unstructured data to build a time series knowledge graph of electromechanical equipment; The second step is to use the dynamic graph convolution embedding model to realize the dynamic embedding of the electromechanical equipment operation and maintenance time series knowledge graph. The process is as follows: Step (2.1) uses a dynamic graph convolutional model to obtain the tail entity embedding vector of the knowledge graph at query time; Step (2.2) uses the dynamic graph convolutional model in step (2.1) to obtain the head entity embedding vector of the knowledge graph at query time; The third step is to obtain the optimal fault prediction result through the score function. The process is as follows: Step (3.1) uses the score function to calculate the scores of all quadruplets accompanying the candidate entity; Step (3.2) calculates the ranking of the prediction results as a basis for comparing the model effects; Step (3.3) selects the object with the highest score as the fault prediction result of the model.

2. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 1, characterized in that: The process of the first step is as follows: Step (1.1) Ontology structure construction, the electromechanical equipment timing knowledge graph ETKG ontology is defined as follows: ; Among them, E represents the set of entities; R represents the set of relationships between these entities; T represents the set of time segments corresponding to entities and relationships. E includes three major categories: C represents the system and components of the equipment; F represents the failure of the equipment; M represents the maintenance of the equipment; Step (1.2) data collection; Step (1.3) Data preprocessing: Use ETL technology to preprocess the data from the previous step, remove invalid data, convert all raw data into data schemas, and store them uniformly in a relational database. ETL technology is used to extract data from one or more sources and load it into the target database after conversion. Step (1.4) entity and relationship extraction, automatically extracting entities and relationships from large amounts of structured and unstructured data.

3. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 2, characterized in that: The process of step (1.2) is: Step (1.2.1) collects structured operating data in the electromechanical equipment control cabinet, including vibration signals, fault codes, operating data, electrical parameters, temperature and humidity; uses TCP / IP communication protocol to remotely obtain real-time operating data of electromechanical equipment in sensors; all sensors are connected to devices and gateways via serial ports, and data communication is carried out through Modbus and RS485 protocols; Step (1.2.2) collects unstructured operation and maintenance data in the electromechanical equipment IoT system, including repair records, maintenance records, fault causes and fault phenomena; and uses crawler technology to obtain unstructured data on fault information in the electromechanical equipment field from the electromechanical equipment IoT system.

4. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 2, characterized in that: The process of step (1.4) is: Step (1.4.1) for structured vibration data Continuous wavelet transform is used to extract entities and relations; Step (1.4.2) uses Word2vec, Bi-LSTM, and CRF technologies to extract entities and relationships for unstructured text data; Step (1.4.3) classifies entities according to the defined ontology, draws the electromechanical equipment timing knowledge graph in the graph database based on the acquired entities and relationships, and effectively queries and updates the entities and relationships of the graph database in combination with the graph database language.

5. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 4, characterized in that: The process of the step (1.4.1) is as follows: Step (1.4.1.1) Vibration data After continuous wavelet transform, ; ; in, is a wavelet basis function with two parameters a and b, Represents the coefficients of wavelet transform, which are obtained by using wavelet basis functions Vibration data Perform integral transformation; represents the conjugate complex root of the wavelet basis function, is the scale parameter, is the translation parameter; ∫ is the integral operation, used in mathematics and statistics; Step (1.4.1.2) calculates the energy spectrum, ; Among them, the energy spectrum The energy distribution at different scales is described to identify the main frequency components; Step (1.4.1.3) extreme point detection realizes fault entity extraction; Through detection The local extreme point of the signal is used to determine the sudden events or abnormal vibrations in the signal, and the entity is extracted according to the ontology classification in the first step; Step (1.4.1.4) calculates the co-occurrence matrix to achieve relationship extraction, ; Construct the co-occurrence matrix of wavelet coefficients at different scales , analyze the relationship between different frequency components.

6. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 4, characterized in that: The process of the step (1.4.2) is as follows: Step (1.4.2.1) Text data embedding: Text data needs to be vectorized, and the corresponding character sequence is mapped to a word vector sequence. Through the word embedding method Word2vec, the word embedding lookup table is updated through continuous training and learning, and finally the input text is matched through the lookup table and converted into the corresponding word vector sequence; Step (1.4.2.2) The text embedding vector is processed by the Bi-LSTM layer; First, the multiple initial parameters used by the Bi-LSTM layer are randomly initialized, and the text embedding vector sequence output by the embedding layer is input into the forward LSTM layer and the reverse LSTM layer model to learn the context information of the sequence vector. Finally, the output vectors in the two directions are concatenated as the output sequence of the Bi-LSTM layer, and the dropout technology is introduced to prevent overfitting. Step (1.4.2.3) The output vector of the Bi-LSTM layer is processed by the CRF layer; The CRF layer considers the relationship between adjacent labels to obtain the global optimal sequence, models the dependencies between entities based on the output of the Bi-LSTM layer, and selects the sequence with the highest probability from the label space; Step (1.4.2.4) calculates the loss function of the model output to obtain the optimal result; The output of the CRF layer is used as the final label sequence of the model output, and compared with the actual labeled sequence to calculate the Loss function. ; in, It represents the probability of outputting sequence Y when given input sequence X. Represents a given input sequence and tag sequence The score function when , log is the logarithmic operation, exp is the natural exponential operation; By continuously adjusting the parameters of each layer to minimize the Loss value, the optimal recognition model is obtained.

7. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 4, characterized in that: The process of the step (1.4.3) is as follows: Step (1.4.3.1) structured data format conversion; Export the structured data processed in step (1.4.1) from the relational database into csv format. To express the temporal characteristics of process state data, convert the traditional triple form into Expanded to a quadruple with timing information , where t represents the time information when the triple is formed. The graph database supports triple storage, and the timing information is converted and embedded into the relationship between entities to realize the establishment of the timing knowledge graph of electromechanical equipment based on the graph database; Step (1.4.3.2) unstructured data format conversion; Convert the entities and entity relationships extracted in step (1.4.2) into csv file format. Since the extracted data are all text, convert the csv file encoding format to UTF-8 encoding; Step (1.4.3.3) imports the csv file data into the graph database.

8. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to any one of claims 1 to 7, characterized in that: In step (2.1), a dynamic graph convolution model is used to obtain The tail entity embedding vector of the knowledge graph , the process is: Step (2.1.1) The tail entity is embedded based on the traditional graph convolution model. ; in, is the adjacency matrix, is the tail entity at time t The hidden state vector at layer k+1 is is the entity in the subgraph at time t that is related to the tail entity Head entity connected by relationship r The collection of , is the normalization constant, Represents the head entity at level k The hidden state vector of is the diagonal relationship weight matrix when it is k layers, Represents the hidden state vector embedded in the initial state, and the activation function uses ReLU, is a trainable parameter matrix; Step (2.1.2) improves the traditional graph convolution model based on the Neural Ordinary Differential Equation; In step (2.2), the dynamic graph convolution model in step (2.1) is used to obtain The embedding vector of the head entity of the knowledge graph .

9. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to claim 8, characterized in that: The process of step (2.1.2) is as follows: Step (2.1.2.1) decomposes the propagation depth k of the traditional graph convolution into the integration time and step length The combination of , the Neural ODE representation of graph convolution is realized by adding a variable integration step size dt, ; in, is the hidden state vector of the tail entity at time t, is the hidden state vector embedded in the initial state of the tail entity, For the head entity, the range is 0 to The hidden state vector at time t, is the diagonal relationship weight matrix; Step (2.1.2.2) uses Taylor expansion to convert the ordinary differential equation into a differential equation form. ; in, is the hidden state vector of the head entity at time t. Taylor expansion approximates the function with a polynomial near a certain point, which is used to approximate the solution of complex functions in differential equations. is the identity matrix; Step (2.1.2.3) , a set of parameters Θ is used to characterize the entire continuous time aggregation process, and the complete tail entity embedding vector is obtained by summing up, ; in, It is a black-box ODE solver used to solve neural ordinary differential equations.

10. The electromechanical equipment fault prediction method based on dynamic graph structure embedding model according to any one of claims 1 to 7, characterized in that: In step (3.1), the score function is used to calculate the scores of all quadruplets accompanying the candidate entity. ; in, For query time, represents transpose, is the set of candidate entities, and ; In step (3.2), the ranking of the prediction results is calculated As a basis for comparing the model effects, ; In step (3.3), the object with the highest score is selected as the fault prediction result of the model. ; Among them, arg max means finding The largest , is the final predicted fault entity.

Citation Information

Patent Citations

  • Knowledge reasoning method based on industrial mechinery fault diagnosis knowledge graph

    CN113961718A

  • Vulnerability meta-knowledge extraction and aggregation method and system based on knowledge graph

    CN117763554A

  • Topology awareness-based time sequence knowledge graph extrapolation method of dynamic relation graph and time sequence fusion

    CN118036749A

  • Spatiotemporal data processing apparatus and method based on graph neural controlled differential equation

    US20240169016A1

Cited By

  • Event causal relationship mining method based on CRF mark learning

    CN119293220A

  • Equipment maintenance fault intelligent analysis method and system based on Internet of Things

    CN120387553A

  • Intelligent analysis method and system for equipment maintenance failure based on internet of things

    CN120387553B

  • Elevator health assessment method based on time sequence knowledge graph and deep learning

    CN121902952A