Multistage alarm management system and method based on digital twinning

Abstract

The invention relates to the technical field of exception management, in particular to a digital twinning-based multi-level alarm management system and method, and the system comprises an information collection unit which is used for collecting the monitoring information of a target plant area; the digital twinborn body construction unit is used for performing bidirectional mapping on the geometric model, the physical model and the intelligent prediction model of the target plant area and constructing a digital twinborn body of the target plant area; the grid division unit is used for performing grid division on the digital twinborn body and establishing a position relation mapping table between the position of each object in the actual target factory area and the position of each object in the digital twinborn body; and the abnormality verification unit is used for judging the abnormal position of the actual target factory according to the abnormal information and the position of the grid where the abnormality occurs in the position relation mapping table, and performing verification according to the abnormal information and the real scene image where the abnormal position is located. According to the invention, the abnormal condition of the target factory can be monitored.

Description

Technical Field

[0001] The present invention relates to the field of abnormality management technology, and in particular to a multi-level alarm management system and method based on digital twins. Background Art

[0002] Digital twin is a new physical system modeling method based on virtual simulation technology. It refers to obtaining the operating status of the actual physical system through technical means such as sensors, data acquisition and information processing, and then building a corresponding virtual model in computer simulation software to achieve accurate simulation of the actual physical system. In the prior art, traditional alarm management systems can usually only provide real-time monitoring data, lack the ability to predict future situations, cannot detect potential anomalies and faults in advance, have weak security authentication and management mechanisms, are susceptible to illegal access and tampering, and cannot effectively guarantee security and credibility. Therefore, the multi-level alarm management system constructed using the digital twin method in the prior art has poor recognition effect and low accuracy for anomalies and faults. Summary of the invention

[0003] In order to solve the technical problem that the multi-level alarm management system constructed by the digital twin method in the prior art has poor recognition effect on abnormalities and faults, the purpose of the present invention is to provide a multi-level alarm management system and method based on digital twins. The technical scheme adopted is as follows: In a first aspect, the present invention provides a multi-level alarm management system based on digital twins, comprising: An information collection unit, used to collect monitoring information of a target plant area, wherein the monitoring information includes environmental information, dynamic information and static information; A digital twin construction unit is used to construct a geometric model of the target plant area by using three-dimensional animation rendering and production according to static information in the monitoring information; construct a physical model of the target plant area according to dynamic information in the monitoring information; construct an intelligent prediction model of the target plant area according to the monitoring information, and bidirectionally map the geometric model, physical model and intelligent prediction model of the target plant area to construct a digital twin of the target plant area; and correct the digital twin of the target plant area according to a first correction parameter and a second correction parameter; A grid division unit, used to grid the digital twin and establish a position relationship mapping table between the position of each object in the actual target factory area and the position of each object in the digital twin; The abnormality verification unit is used to determine the abnormal position of the actual target plant area according to the abnormal information and the position of the grid where the abnormality occurs in the position relationship mapping table, and verify it according to the abnormal information and the real scene image where the abnormal position is located.

[0004] Preferably, the system further comprises: The security authentication unit is used to register and manage the digital twin in the private chain of the blockchain, synchronize the real-time received monitoring data in the private blockchain, and transmit the digital twin after security authentication to the dynamic correction unit and the simulation unit; A dynamic correction unit is used to compare the real-time operation data of the target plant area after safety authentication with the output information of the digital twin, obtain a first correction parameter, and generate a second correction parameter according to the abnormal information in the verification result transmitted by the abnormal verification unit; correct the geometric model, physical model and intelligent prediction model of the digital twin according to the comparison result; and transmit the first correction parameter and the second correction parameter to the digital twin construction unit; The simulation unit is used to predict the abnormal information existing in the operation process of the target plant area according to the intelligent prediction model in the digital twin of the target plant area after safety authentication; and transmit the abnormal information to the abnormal verification unit.

[0005] Preferably, the system further comprises: A real-scene image acquisition unit, used to acquire real-scene images of the target factory area through multiple groups of cameras set in the target factory area, and transmit the real-scene images to the abnormality verification unit; An abnormality sorting unit, used for receiving abnormality information in the verification result transmitted by the abnormality verification unit; Construct a related knowledge graph based on the abnormal information, calculate the similarity between each abnormal information through a similarity calculation algorithm, and compare the calculated similarity between each abnormal information with a preset threshold to determine whether there is any correlation between the abnormal information; If there is a correlation, the associated abnormal information is merged in the associated knowledge graph; the corresponding abnormal processing level is increased according to the number of abnormal information in each node in the associated knowledge graph, and the content in each node in the associated knowledge graph and the corresponding abnormal processing level are transmitted to the emergency processing unit; The emergency processing unit is used to generate a troubleshooting list based on the content of each node in the associated knowledge graph and the corresponding exception handling level, and to troubleshoot the occurred faults according to the troubleshooting list to identify potential fault hazards.

[0006] Preferably, the digital twin construction unit includes a geometric model construction module, a physical model construction module, an intelligent prediction model construction module, a digital twin construction module and a correction model application module; The geometric model building module is used to build a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; The physical model building module is used to build a physical model of the target plant area according to the dynamic information in the monitoring information, and the physical model includes an equipment operation model, a production line operation model and a logistics transportation model; The intelligent prediction model building module is used to build an intelligent prediction model of the target plant area according to the monitoring information, so as to predict the occurrence of abnormalities; The digital twin construction module is used to bidirectionally map the geometric model, physical model and intelligent prediction model of the target plant area to construct the digital twin of the target plant area; The correction model application module is used to correct the digital twin of the target plant area according to the first correction parameter and the second correction parameter transmitted by the dynamic correction unit.

[0007] Preferably, the grid division unit includes a grid determination module, a grid division module, a grid numbering module and a grid association module; The grid determination module is used to determine the size and number of grids according to the size and shape of the actual target plant area and the accuracy and resolution of the digital twin; The grid division module is used to model and simulate the actual target plant area using digital twin technology, dividing it into a number of grids, each grid representing a small area; The grid numbering module is used to number each grid; The grid association module is used to match and associate each object in the digital twin with the objects in the actual target plant, determine their position and size in the digital twin and the actual target plant, associate the grids where the two are located with the numbers, and record them in the position relationship mapping table.

[0008] Preferably, the abnormality verification unit is further used to receive the abnormality information transmitted by the simulation unit, the position relationship mapping table transmitted by the grid division unit, and the real scene image transmitted by the real scene image acquisition unit; The verification based on the abnormal information and the real-scene image of the abnormal location includes: Verify whether the anomaly has occurred through the real-life image of the anomaly location, and determine whether the anomaly type is correct based on the anomaly information; If an exception occurs and the exception type is correct, the exception information is transmitted to the exception sorting unit; If an exception occurs and the exception type is incorrect, the exception type is modified according to the real scene image, and the modified exception information is transmitted to the exception sorting unit; If the abnormality does not occur, the abnormality information is transmitted to the dynamic correction unit.

[0009] Preferably, the environmental information includes the temperature, humidity, air quality and noise level in the target factory area; the dynamic information includes the equipment operating status, production line operation status, personnel activity trajectory and logistics transportation status in the target factory area; the static information includes the building structure, equipment layout, safety facilities, entrance and exit locations and camera layout in the target factory area.

[0010] Preferably, the similarity calculation algorithm is specifically:

[0011]

[0012] in, represents the edit distance, represents the minimum edit distance size, Represents the two-dimensional coordinates of the computation matrix.

[0013] In a second aspect, the present invention provides a multi-level alarm management method based on digital twins, comprising: Step S1, information collection, including collecting monitoring information of the target plant area, wherein the monitoring information includes environmental information, dynamic information and static information; Step S2, data transmission; Step S3, data preprocessing; Step S4, digital twin construction, including constructing a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; constructing a physical model of the target plant area according to the dynamic information in the monitoring information; constructing an intelligent prediction model of the target plant area according to the monitoring information, bidirectionally mapping the geometric model, physical model and intelligent prediction model of the target plant area to construct a digital twin of the target plant area; and correcting the digital twin of the target plant area according to the first correction parameter and the second correction parameter; Step S5, grid division, including gridding the digital twin, and establishing a position relationship mapping table between the position of each object in the actual target factory area and the position of each object in the digital twin; Step S6, safety authentication, includes determining the abnormal position of the actual target plant area according to the abnormal information and the position of the grid where the abnormality occurs in the position relationship mapping table, and verifying according to the abnormal information and the real scene image where the abnormal position is located.

[0014] The embodiments of the present invention have at least the following beneficial effects: The present invention can collect the environment, dynamic and static information of the target plant area in real time through the information collection unit, including temperature, humidity, equipment operation status, etc., provide comprehensive and accurate monitoring data, and use the pre-processed monitoring information and correction parameters to construct the digital twin of the target plant area through the digital twin construction unit, including geometric model, physical model and intelligent prediction model, to achieve comprehensive simulation and prediction of the target plant area, and mesh the digital twin through the mesh division unit, and associate it with the actual target plant area, establish a position relationship mapping table, provide accurate object position information, and provide basic data support for subsequent abnormal verification and processing. According to the comparison results, the geometric model, physical model and intelligent prediction model are corrected to maintain the consistency of the digital twin with the actual situation. The abnormal verification unit receives abnormal information, position relationship mapping table and real-life image, determines the type and location of the abnormality, and verifies whether the abnormality occurs. The alarm management system of the present invention has better recognition effect and higher accuracy for abnormalities and faults, and can troubleshoot and avoid potential hidden dangers for the faults that occur. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a structural schematic diagram of a multi-level alarm management system based on digital twins provided by an embodiment of the present invention; Figure 2 It is an internal module block diagram of a digital twin construction unit provided in an embodiment of the present invention; Figure 3 is a block diagram of the internal modules of a grid division unit provided in an embodiment of the present invention; Figure 4 is a block diagram of the internal modules of the abnormal sorting unit provided by an embodiment of the present invention; Figure 5 It is a flowchart of a multi-level alarm management method based on digital twins provided by an embodiment of the present invention; In the figure: 1. Information acquisition unit; 2. Data transmission unit; 3. Data preprocessing unit; 4. Digital twin construction unit; 401. Geometric model construction module; 402. Physical model construction module; 403. Intelligent prediction model construction module; 404. Digital twin construction module; 405. Correction model application module; 5. Security authentication unit; 6. Dynamic correction unit; 7. Simulation unit; 8. Abnormal verification unit; 9. Grid division unit; 901. Grid determination module; 902. Grid division module; 903. Grid numbering module; 904. Grid association module; 10. Real-life image acquisition unit; 11. Abnormal sorting unit; 1101. Associated knowledge graph construction module; 1102. Similarity calculation module; 1103. Correlation judgment module; 1104. Exception handling level improvement module; 12. Emergency handling unit. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a multi-level alarm management system and method based on digital twins proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following is a detailed description of a multi-level alarm management system and method based on digital twins provided by the present invention in conjunction with the accompanying drawings.

[0020] See also Figure 1 , which shows a structural schematic diagram of a multi-level alarm management system based on digital twins provided by an embodiment of the present invention, the system comprising: an information acquisition unit 1, a data transmission unit 2, a data preprocessing unit 3, a digital twin construction unit 4, a security authentication unit 5, a dynamic correction unit 6, a simulation unit 7, an abnormality verification unit 8, a grid division unit 9, a real-scene image acquisition unit 10, an abnormality sorting unit 11 and an emergency processing unit 12.

[0021] The information collection unit 1 is used to collect monitoring information of the target plant area, and the monitoring information includes environmental information, dynamic information and static information.

[0022] The information collection unit 1 transmits the collected monitoring information to the data transmission unit 2. The environmental information includes the temperature, humidity, air quality and noise level in the target factory area; the dynamic information includes the equipment operation status, production line operation status, personnel activity trajectory and logistics transportation status in the target factory area; the static information includes the building structure, equipment layout, safety facilities, entrance and exit locations and camera layout in the target factory area.

[0023] The data transmission unit 2 receives the monitoring information collected by the information collection unit 1 and transmits the monitoring information to the data preprocessing unit 3 .

[0024] The data preprocessing unit 3 receives the monitoring information transmitted by the data transmission unit 2, and stores the monitoring information in the database using high-speed bandwidth transmission, performs data preprocessing through the Z-score standardization method to obtain the normalized characteristics of the monitoring information, and transmits the preprocessed monitoring information to the digital twin construction unit 4.

[0025] The digital twin construction unit 4 is used to construct a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; to construct a physical model of the target plant area according to the dynamic information in the monitoring information; to construct a deep learning model of the target plant area according to the monitoring information based on data preprocessing, model design, training and optimization, and to perform bidirectional mapping of the geometric model, physical model and intelligent prediction model of the target plant area to construct a digital twin of the target plant area; and to correct the digital twin of the target plant area according to the first correction parameter and the second correction parameter.

[0026] More specifically, the digital twin construction unit 4 receives the preprocessed monitoring information transmitted by the data preprocessing unit 3 and the first correction parameter and the second correction parameter transmitted by the dynamic correction unit 6. Figure 2 As shown, the digital twin construction unit 4 includes a geometric model construction module 401, a physical model construction module 402, an intelligent prediction model construction module 403, a digital twin construction module 404 and a correction model application module 405.

[0027] The geometric model building module 401 builds a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information.

[0028] The physical model building module 402 builds a physical model of the target plant area according to the dynamic information in the monitoring information, including an equipment operation model, a production line operation model and a logistics transportation model.

[0029] The intelligent prediction model building module 403 builds an intelligent prediction model of the target plant area based on the monitoring information, which is used to predict the occurrence of abnormalities.

[0030] The digital twin construction module 404 performs bidirectional mapping of the geometric model, physical model and intelligent prediction model of the target plant to construct a digital twin of the target plant. The digital twin is a virtual image of the actual target plant, and the digital twin reflects the operating status and characteristics of the actual target plant.

[0031] The correction model application module 405 corrects the digital twin of the target plant area according to the first correction parameter and the second correction parameter transmitted by the dynamic correction unit 6 .

[0032] The grid division unit 9 is used to perform grid division on the digital twin and establish a position relationship mapping table between the position of each object in the actual target factory area and the position of each object in the digital twin.

[0033] Specifically, Figure 3 As shown, the grid division unit 9 receives the digital twin transmitted by the digital twin construction unit 4, and divides the digital twin into grids, and proportionally associates each divided grid with the actual target factory area to establish a position relationship mapping table between the location of each object in the actual target factory area and the location of each object in the digital twin, and transmits the grid-divided digital twin to the security authentication unit 5, and transmits the established position relationship mapping table to the abnormality verification unit 8.

[0034] More specifically, the grid division unit 9 includes a grid determination module 901 , a grid division module 902 , a grid numbering module 903 and a grid association module 904 .

[0035] The grid determination module 901 is used to determine the size and number of grids according to the size and shape of the actual target plant area and the accuracy and resolution of the digital twin; The grid division module 902 is used to model and simulate the actual target plant area using the digital twin technology, dividing it into a number of grids, each grid representing a small area; The grid numbering module 903 is used to number each grid so as to record and search in the position relationship mapping table, and the grid numbering adopts a two-dimensional coordinate system to identify the position of each grid in the form of rows and columns; The grid association module 904 is used to match and associate each object in the digital twin with the objects in the actual target plant, determine their position and size in the digital twin and the actual target plant, associate the grids and numbers where the two are located, and record them in the position relationship mapping table.

[0036] The security authentication unit 5 is used to register and manage the digital twin in the private chain of the blockchain, and synchronize the real-time received monitoring data in the private blockchain.

[0037] More specifically, the security authentication unit 5 receives the digital twin after grid division transmitted by the grid division unit 9, and registers and manages the digital twin in the private chain of the blockchain, ensuring that only authorized digital twins can receive the monitoring data transmitted by the information collection unit 1, and synchronize the real-time received monitoring data in the private blockchain, and transmit the digital twin after security authentication to the dynamic correction unit 6 and the simulation unit 7.

[0038] The dynamic correction unit 6 is used to compare the real-time operation data of the target plant after safety authentication with the output information of the digital twin, obtain the first correction parameter, and generate the second correction parameter according to the abnormal information in the verification result transmitted by the abnormal verification unit 8; correct the geometric model, physical model and intelligent prediction model of the digital twin according to the comparison result; and transmit the first correction parameter and the second correction parameter to the digital twin construction unit 4.

[0039] More specifically, the dynamic correction unit 6 receives the digital twin after security authentication transmitted by the security authentication unit 5 and the abnormal information transmitted by the abnormal verification unit 8. The real-time operation data of the target plant is compared with the output information of the digital twin, and the error between the two is calculated, and the geometric model, physical model and intelligent prediction model of the digital twin are corrected according to the comparison result until the error between the two is stable and lower than the threshold, and a first correction parameter is generated based on the comparison result, and a second correction parameter is generated according to the received abnormal information, and the first correction parameter and the second correction parameter are transmitted to the digital twin construction unit 4.

[0040] The simulation unit 7 is used to predict the abnormal information existing in the operation process of the target plant area according to the intelligent prediction model in the digital twin of the target plant area after safety authentication; and transmit the abnormal information to the abnormal verification unit 8.

[0041] More specifically, the digital twin of the target plant area after security authentication transmitted by the security authentication unit 5 is received, and the operation process of the target plant area is displayed based on the digital twin of the target plant area. The abnormal information existing in the operation process of the target plant area is predicted through the intelligent prediction model in the digital twin of the target plant area, and the predicted abnormal information is transmitted to the abnormality verification unit 8.

[0042] The abnormality verification unit 8 is used to determine the abnormal position of the actual target plant area according to the abnormality information and the position of the grid where the abnormality occurs in the position relationship mapping table, and verify according to the abnormality information and the real scene image where the abnormal position is located.

[0043] More specifically, the abnormality verification unit 8 receives the abnormality information transmitted by the simulation unit 7, the position relationship mapping table transmitted by the grid division unit 9, and the real scene image transmitted by the real scene image acquisition unit 10. The abnormality type is determined according to the abnormality information, and the abnormal position of the actual target plant area is determined according to the position of the grid where the abnormality occurs in the position relationship mapping table, and the real scene image of the abnormal position is retrieved according to the abnormal position.

[0044] The verification based on the abnormal information and the real-scene image of the abnormal location includes: The abnormality verification unit receives the abnormality information transmitted by the simulation unit, and verifies the accuracy of the abnormality information by combining the real scene image provided by the real scene image acquisition unit and the position relationship mapping table provided by the grid division unit, and determines whether the abnormality type including equipment failure and environmental abnormality is correct based on the simulation results and actual monitoring data.

[0045] If an exception occurs and the exception type is correct, the exception information is transmitted to the exception sorting unit 11; if an exception occurs and the exception type is incorrect, the exception type is modified according to the real scene image to ensure that the modified exception type is correct, and the modified exception information is transmitted to the exception sorting unit 11; if the exception does not occur, the exception information is transmitted to the dynamic correction unit 6.

[0046] The real scene image acquisition unit 10 is used to acquire the real scene image of the target factory area by setting a plurality of groups of cameras in the target factory area, and transmit the acquired real scene image of the target factory area to the abnormality verification unit 8 .

[0047] The abnormality sorting unit 11 is used to receive abnormality information in the verification result transmitted by the abnormality verification unit 8.

[0048] An associated knowledge graph is constructed based on the abnormal information to determine the sequential relationship, concurrent relationship and conflict relationship of the abnormal information. The similarity between each abnormal information is calculated through a similarity calculation algorithm, and the calculated similarity between each abnormal information is compared with the preset threshold to determine whether there is a correlation between the abnormal information.

[0049] When the similarity between any two pieces of abnormal information is greater than a preset threshold, there is an association between the any two pieces of abnormal information; otherwise, there is no association.

[0050] If there is a correlation, the associated exception information will be merged in the associated knowledge graph; the corresponding exception handling level will be increased according to the amount of exception information in each node in the associated knowledge graph, and the content in each node in the associated knowledge graph and the corresponding exception handling level will be transmitted to the emergency processing unit 12.

[0051] The similarity calculation method is specifically as follows:

[0052]

[0053] in, represents the edit distance, represents the minimum edit distance size, Represents the two-dimensional coordinates of the computation matrix.

[0054] More specifically, if Figure 4 As shown, the exception sorting unit 11 includes an associated knowledge graph construction module 1101, a similarity calculation module 1102, a correlation judgment module 1103 and an exception handling level enhancement module 1104.

[0055] The associated knowledge graph construction module 1101 is used to construct an associated knowledge graph based on the received exception information, so as to determine the sequential relationship, concurrent relationship and conflict relationship between the exception information.

[0056] The similarity calculation module 1102 is used to calculate the similarity between each abnormal information so as to determine whether there is a correlation between the abnormal information.

[0057] The correlation judgment module 1103 is used to compare the calculated similarities between each abnormal information with the threshold set according to the set information including system sensitivity and historical false alarm rate parameters to determine whether there is a correlation between the abnormal information. If there is a correlation, the abnormal information with the correlation will be merged in the associated knowledge graph.

[0058] The exception handling level enhancement module 1104 is used to enhance the corresponding exception handling level according to the amount of exception information in each node in the associated knowledge graph.

[0059] The emergency processing unit 12 is used to generate a troubleshooting list according to the content in each node in the associated knowledge graph and the corresponding exception processing level, and to troubleshoot the occurred faults according to the troubleshooting list to identify potential fault hazards.

[0060] The emergency processing unit 12 receives the content and corresponding exception processing level of each node in the associated knowledge graph transmitted by the exception sorting unit 11, and identifies and avoids potential fault hazards.

[0061] To summarize, the present invention can collect the environmental, dynamic and static information of the target plant area in real time, including temperature, humidity, equipment operating status, etc., through the information collection unit, provide comprehensive and accurate monitoring data, and preprocess and normalize the monitoring information through the data transmission unit and the data preprocessing unit to ensure the accuracy and consistency of the data, and store the data in a database for subsequent analysis and application. The digital twin of the target plant area is constructed by the digital twin construction unit using the preprocessed monitoring information and correction parameters, including a geometric model, a physical model and an intelligent prediction model, to achieve comprehensive simulation and prediction of the target plant area, and the digital twin is gridded by the grid division unit, and associated with the actual target plant area, to establish a position relationship mapping table, to provide accurate object location information, and to provide basic data support for subsequent abnormality verification and processing.

[0062] The present invention registers and manages digital twins through a private chain through a security authentication unit, ensuring that only authorized digital twins can receive monitoring data, ensuring the security and credibility of the system, and correcting the digital twins through a dynamic correction unit, comparing the actual operation data with the output information, and correcting the geometric model, physical model and intelligent prediction model according to the comparison results, maintaining the consistency of the digital twins with the actual situation, and displaying the operation process of the target plant through the digital twin through the simulation unit, and predicting abnormal information, providing decision support, and judging the type and location of the abnormality through the abnormal verification unit by receiving abnormal information, position relationship mapping table and real-life image, and verifying whether the abnormality occurs. The abnormal sorting unit sorts and correlates the abnormal information, improves the abnormality processing level, provides reference for emergency processing, and generates a troubleshooting list according to the associated knowledge map through the emergency processing unit, and troubleshoots the occurred faults and avoids potential hidden dangers.

[0063] See also Figure 5 , which shows a step flow chart of a multi-level alarm management method based on digital twins provided by an embodiment of the present invention, the method comprising the following steps: Step S1, information collection, including collecting monitoring information of the target plant area, wherein the monitoring information includes environmental information, dynamic information and static information; Step S2, data transmission, includes data transmission of monitoring information through 5G high-speed mobile transmission equipment.

[0064] Step S3, data preprocessing, includes performing data preprocessing by a Z-score standardization method to obtain normalized features of the monitoring information.

[0065] Step S4, digital twin construction, includes constructing a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; constructing a physical model of the target plant area according to the dynamic information in the monitoring information; constructing an intelligent prediction model of the target plant area according to the monitoring information, bidirectionally mapping the geometric model, physical model and intelligent prediction model of the target plant area to construct a digital twin of the target plant area; and correcting the digital twin of the target plant area according to the first correction parameter and the second correction parameter.

[0066] Step S5, grid division, includes gridding the digital twin and establishing a position relationship mapping table between the location of each object in the actual target factory area and the location of each object in the digital twin.

[0067] Step S6, safety authentication, includes determining the abnormal position of the actual target plant area according to the abnormal information and the position of the grid where the abnormality occurs in the position relationship mapping table, and verifying according to the abnormal information and the real scene image where the abnormal position is located.

[0068] Step S7, dynamic correction, includes comparing the real-time operating data of the target plant with the output information of the digital twin, and correcting the geometric model, physical model and intelligent prediction model of the digital twin according to the comparison results until the error between the two is stable and lower than the threshold, generating a first correction parameter based on the comparison result, and generating a second correction parameter based on the received abnormal information.

[0069] Step S8, simulation, includes displaying the operation process of the target plant area according to the digital twin, and predicting abnormal information existing in the operation process of the target plant area through the intelligent prediction model in the digital twin.

[0070] Step S9, real scene image acquisition, includes acquiring real scene images of the target factory area by setting up multiple groups of cameras in the target factory area.

[0071] Step S10, abnormality verification, including determining the abnormality type according to the abnormality information, and determining the abnormal position of the actual target plant area according to the position of the grid where the abnormality occurs in the position relationship mapping table, and retrieving the real scene image of the abnormal position according to the abnormal position, and verifying whether the abnormality occurs and whether the abnormality type is correct through the real scene image. If the abnormality occurs and the abnormality type is correct, the abnormality information is transmitted. If the abnormality occurs and the abnormality type is incorrect, the abnormality type is modified according to the real scene image, and the modified abnormality information is transmitted; Step S11, exception sorting, includes constructing an associated knowledge graph based on the exception information to determine the sequential relationship, concurrent relationship and conflict relationship of the exception information, calculating the similarity between each exception information through a similarity calculation algorithm, and comparing the calculated similarity between each exception information with a preset threshold to determine whether there is a correlation between the exception information. If there is a correlation, the associated exception information will be merged in the associated knowledge graph, and the corresponding exception handling level will be increased according to the number of exception information in each node in the associated knowledge graph.

[0072] Step S12, emergency handling, includes generating a troubleshooting list based on the content in each node in the associated knowledge graph and the corresponding exception handling level, and troubleshooting the occurred faults according to the troubleshooting list to identify and avoid potential fault hazards.

[0073] It should be noted that a multi-level alarm management method based on digital twins essentially executes the method steps of a multi-level alarm management system processing process based on digital twins. Since an embodiment of a multi-level alarm management system based on digital twins has been explained in detail, it will not be introduced in detail here.

[0074] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A multi-level alarm management system based on digital twins, characterized in that: The system includes: An information collection unit (1) is used to collect monitoring information of a target plant area, wherein the monitoring information includes environmental information, dynamic information and static information; The digital twin construction unit (4) is used to construct a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; construct a physical model of the target plant area according to the dynamic information in the monitoring information; construct an intelligent prediction model of the target plant area according to the monitoring information, and perform bidirectional mapping between the geometric model, the physical model and the intelligent prediction model of the target plant area to construct a digital twin of the target plant area; and correct the digital twin of the target plant area according to the first correction parameter and the second correction parameter; A grid division unit (9) is used to divide the digital twin into grids and establish a position relationship mapping table between the position of each object in the actual target factory area and the position of each object in the digital twin; The abnormality verification unit (8) is used to determine the abnormal position of the actual target plant area according to the abnormality information and the position of the grid where the abnormality occurs in the position relationship mapping table, and to verify according to the abnormality information and the real scene image where the abnormal position is located.

2. According to claim 1, a multi-level alarm management system based on digital twins is characterized in that: The system also includes: A security authentication unit (5) is used to register and manage the digital twin in the private chain of the blockchain, synchronize the real-time received monitoring data in the private blockchain; and transmit the digital twin after security authentication to the dynamic correction unit (6) and the simulation unit (7); A dynamic correction unit (6) is used to compare the real-time operation data of the target plant area after safety authentication with the output information of the digital twin, obtain a first correction parameter, generate a second correction parameter according to the abnormal information in the verification result transmitted by the abnormal verification unit (8); correct the geometric model, physical model and intelligent prediction model of the digital twin according to the comparison result; and transmit the first correction parameter and the second correction parameter to the digital twin construction unit (4); The simulation unit (7) is used to predict abnormal information existing in the operation process of the target plant area according to the intelligent prediction model in the digital twin of the target plant area after safety authentication; and transmit the abnormal information to the abnormal verification unit (8).

3. According to claim 2, a multi-level alarm management system based on digital twins is characterized in that: The system also includes: A real scene image acquisition unit (10), used to acquire a real scene image of the target factory area through a plurality of cameras arranged in the target factory area, and transmit the real scene image to the abnormality verification unit (8); An abnormality sorting unit (11) is used to receive abnormality information in the verification result transmitted by the abnormality verification unit (8); Construct a related knowledge graph based on the abnormal information, calculate the similarity between each abnormal information through a similarity calculation algorithm, and compare the calculated similarity between each abnormal information with a preset threshold to determine whether there is any correlation between the abnormal information; If there is a correlation, the associated abnormal information is merged in the associated knowledge graph; the corresponding abnormal processing level is increased according to the amount of abnormal information in each node in the associated knowledge graph, and the content in each node in the associated knowledge graph and the corresponding abnormal processing level are transmitted to the emergency processing unit (12); The emergency processing unit (12) is used to generate a troubleshooting list based on the content of each node in the associated knowledge graph and the corresponding exception processing level, and to troubleshoot the occurred faults according to the troubleshooting list to identify potential fault hazards.

4. According to claim 1, a multi-level alarm management system based on digital twins is characterized in that: The digital twin construction unit (4) comprises a geometric model construction module (401), a physical model construction module (402), an intelligent prediction model construction module (403), a digital twin construction module (404) and a correction model application module (405); The geometric model building module (401) is used to build a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; The physical model building module (402) is used to build a physical model of the target plant area according to the dynamic information in the monitoring information, and the physical model includes an equipment operation model, a production line operation model and a logistics transportation model; The intelligent prediction model building module (403) is used to build an intelligent prediction model of the target plant area according to the monitoring information, so as to predict the occurrence of abnormalities; The digital twin construction module (404) is used to perform bidirectional mapping of the geometric model, physical model and intelligent prediction model of the target plant area to construct a digital twin of the target plant area; The correction model application module (405) is used to correct the digital twin of the target plant area according to the first correction parameter and the second correction parameter transmitted by the dynamic correction unit (6).

5. According to claim 1, a multi-level alarm management system based on digital twins is characterized in that: The grid division unit (9) comprises a grid determination module (901), a grid division module (902), a grid numbering module (903) and a grid association module (904); The grid determination module (901) is used to determine the size and number of grids according to the size and shape of the actual target plant area and the accuracy and resolution of the digital twin; The grid division module (902) is used to model and simulate the actual target plant area using the digital twin technology, dividing it into a plurality of grids, each grid representing a small area; The grid numbering module (903) is used to number each grid; The grid association module (904) is used to match and associate each object in the digital twin with the objects in the actual target plant, determine their positions and sizes in the digital twin and the actual target plant, associate the grids where the two are located with the numbers, and record them in the position relationship mapping table.

6. A multi-level alarm management system based on digital twin according to claim 3, characterized in that: The abnormality verification unit (8) is also used to receive the abnormality information transmitted by the simulation unit (7), the position relationship mapping table transmitted by the grid division unit (9), and the real scene image transmitted by the real scene image acquisition unit (10); The verification based on the abnormal information and the real-scene image of the abnormal location includes: Verify whether the anomaly has occurred through the real-life image of the anomaly location, and determine whether the anomaly type is correct based on the anomaly information; If an exception occurs and the exception type is correct, the exception information is transmitted to the exception sorting unit (11); If an exception occurs and the exception type is incorrect, the exception type is modified according to the real scene image, and the modified exception information is transmitted to the exception sorting unit (11); If no abnormality occurs, the abnormality information is transmitted to the dynamic correction unit (6).

7. According to claim 1, a multi-level alarm management system based on digital twins is characterized in that: The environmental information includes the temperature, humidity, air quality and noise level in the target factory area; the dynamic information includes the equipment operating status, production line operation status, personnel activity trajectory and logistics transportation status in the target factory area; the static information includes the building structure, equipment layout, safety facilities, entrance and exit locations and camera layout in the target factory area.

8. The multi-level alarm management system based on digital twin according to claim 3 is characterized in that: The similarity calculation algorithm is specifically as follows: in, represents the edit distance, represents the minimum edit distance size, Represents the two-dimensional coordinates of the computation matrix.

9. A multi-level alarm management method based on digital twins, characterized in that: The method comprises the following steps: Step S1, information collection, including collecting monitoring information of the target plant area, wherein the monitoring information includes environmental information, dynamic information and static information; Step S2, data transmission; Step S3, data preprocessing; Step S4, digital twin construction, including constructing a geometric model of the target plant area by using three-dimensional animation rendering and production according to the static information in the monitoring information; constructing a physical model of the target plant area according to the dynamic information in the monitoring information; constructing an intelligent prediction model of the target plant area according to the monitoring information, bidirectionally mapping the geometric model, physical model and intelligent prediction model of the target plant area to construct a digital twin of the target plant area; and correcting the digital twin of the target plant area according to the first correction parameter and the second correction parameter; Step S5, grid division, including gridding the digital twin, and establishing a position relationship mapping table between the position of each object in the actual target factory area and the position of each object in the digital twin; Step S6, safety authentication, includes determining the abnormal position of the actual target plant area according to the abnormal information and the position of the grid where the abnormality occurs in the position relationship mapping table, and verifying according to the abnormal information and the real scene image where the abnormal position is located.

Images