Digital-twin-based intelligent monitoring method and system for urban infrastructure

Abstract

The application relates to the technical field of urban infrastructure management, and discloses a city infrastructure intelligent supervision method and system based on digital twinning, which comprises a city infrastructure identification module, a city infrastructure diagnosis module and a city infrastructure feedback module; scientific dynamic diagnosis of abnormal hardware structure of a city infrastructure supervision node is carried out according to city infrastructure supervision node object entity model information and target city infrastructure supervision node normal three-dimensional entity model information, so that digital inspection of the external structure of the city infrastructure hardware is realized; target city infrastructure abnormal entity model identification data is transmitted to a city management platform through an Internet of Things communication network and is displayed on a display screen, so that visualized supervision of the city infrastructure based on digital twinning is realized, and the intuitiveness and applicability of the city infrastructure supervision are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of urban infrastructure management, in particular to an urban infrastructure intelligent supervision method and system based on digital twinning. BACKGROUND

[0002] Urban infrastructure is a general term for engineering infrastructure and social infrastructure necessary for the survival and development of a city. Urban infrastructure is generally divided into two categories: engineering infrastructure and social infrastructure. Engineering infrastructure generally refers to six systems: energy supply system, water supply and drainage system, road transportation system, communication system, environmental sanitation system, and urban disaster prevention system. Urban infrastructure supervision includes fault diagnosis and fault maintenance of urban infrastructure. The existing urban infrastructure supervision process cannot achieve digital intelligent inspection of urban infrastructure, nor can it achieve visual feedback of abnormal state of urban infrastructure, which reduces the efficiency and quality of urban infrastructure supervision.

[0003] Chinese patent application with publication number CN118446867A discloses a method and system for analyzing urban waterlogging based on a mathematical model. The method monitors changes in urban spatial layout based on real-time urban geographic spatial information data, analyzes urban terrain change information, develops a city infrastructure layout map, and obtains city change topology data. However, the above technical solution cannot monitor urban waterlogging objects based on digitalization and provide visual feedback. SUMMARY

[0004] (I) Technical problems solved

[0005] To solve the above problems that the existing urban infrastructure supervision process cannot achieve digital intelligent inspection of urban infrastructure, nor can it achieve visual feedback of abnormal state of urban infrastructure, which reduces the efficiency and quality of urban infrastructure supervision, the present application aims to accurately identify urban infrastructure supervision objects, intelligently determine urban infrastructure supervision node trajectories, scientifically and clearly construct urban infrastructure supervision object entity models, autonomously search for normal urban infrastructure supervision object entity models, intelligently diagnose abnormal state of urban infrastructure supervision objects, and visually display abnormal urban infrastructure objects.

[0006] (II) Technical solutions

[0007] The present application is implemented by the following technical solutions: an urban infrastructure intelligent supervision method based on digital twinning, comprising the following steps:

[0008] S1, collecting urban infrastructure feature text data;

[0009] S2, according to the city infrastructure feature text data and different city infrastructure object standard feature text data, the object recognition processing of city infrastructure is carried out, and city infrastructure object recognition data is generated;

[0010] S3, according to the city infrastructure object recognition data and city infrastructure supervision node standard trajectory data, the path trajectory search processing of city infrastructure supervision node is carried out, and target city infrastructure supervision node trajectory data is generated;

[0011] S4, based on the target city infrastructure supervision node trajectory data, the three-dimensional entity model construction processing of city infrastructure supervision node object is carried out, and city infrastructure supervision node object entity model data is generated;

[0012] S5, according to the city infrastructure object recognition data and city infrastructure supervision node object normal entity model data, the normal three-dimensional entity model information search processing of city infrastructure supervision node is carried out, and target city infrastructure supervision node normal three-dimensional entity model data is generated;

[0013] S6, according to the city infrastructure supervision node object entity model data and the target city infrastructure supervision node normal three-dimensional entity model data, the facility hardware structure abnormal diagnosis processing of city infrastructure supervision node is carried out, and target city infrastructure fault diagnosis result data is generated;

[0014] S7, the target city infrastructure abnormal entity model identification data is constructed, and the city infrastructure supervision result display operation is carried out.

[0015] Preferably, the operation steps of collecting city infrastructure feature text data are as follows:

[0016] S11, the feature description text information of the target city infrastructure supervised by the city management platform is input online, and city infrastructure feature text data is generated , city infrastructure includes power hardware infrastructure, gas hardware infrastructure, heating hardware infrastructure, track hardware infrastructure and fire hardware infrastructure; The city infrastructure feature text data includes geographical administrative region feature information, distribution geographical position feature information and function service range feature information of the target city infrastructure.

[0017] Preferably, according to the city infrastructure feature text data and different city infrastructure object standard feature text data, the object recognition processing of city infrastructure is carried out, and city infrastructure object recognition data is generated. The operation steps are as follows:

[0018] S21, establish different city infrastructure object standard feature text data set , ;in Indicates the first Standard feature text data of different urban infrastructure objects corresponding to each urban infrastructure regulatory object. This represents the maximum number of urban infrastructure regulatory objects; the standard feature text data for different urban infrastructure objects represents the standard feature description text information set for different urban infrastructure regulatory objects.

[0019] S22. Use the BERT language model search algorithm to extract the urban infrastructure feature text data. The standard feature text data set of the different urban infrastructure objects The text data of standard features of different urban infrastructure objects described in the article Perform feature text character matching to search for text data that matches the features of the urban infrastructure. Matching standard feature text data of the different urban infrastructure objects The corresponding text information of the urban infrastructure regulatory object number is used to generate urban infrastructure object identification data through data identification. .

[0020] Preferably, the steps for generating target urban infrastructure monitoring node trajectory data by performing path trajectory search processing on the urban infrastructure monitoring nodes based on the urban infrastructure object identification data and the standard trajectory data of urban infrastructure monitoring nodes are as follows:

[0021] S31. Establish a standard trajectory data set for urban infrastructure monitoring nodes. ,in Indicates the first Standard trajectory data of urban infrastructure supervision nodes corresponding to urban infrastructure supervision objects, wherein the standard trajectory data of urban infrastructure supervision nodes represents the spatial coordinate information of the supervision trajectory of urban infrastructure supervision nodes set for different urban infrastructure supervision objects;

[0022] S32. Use a bidirectional search algorithm to identify the urban infrastructure object data. With the standard trajectory data set of the urban infrastructure monitoring nodes Standard trajectory data of urban infrastructure monitoring nodes described in the article Perform character matching of urban infrastructure object numbers to retrieve the identification data of the urban infrastructure objects. The corresponding standard trajectory data of the urban infrastructure monitoring nodes And generate trajectory data of infrastructure monitoring nodes in the target city by identifying data. The target city infrastructure monitoring node trajectory data represents the spatial coordinate information of the monitoring trajectory of all monitoring nodes of the target city infrastructure monitoring object.

[0023] Preferably, the steps for generating entity model data of urban infrastructure monitoring node objects by performing 3D entity model construction processing on the trajectory data of the target urban infrastructure monitoring nodes are as follows:

[0024] S41. Using a drone equipped with a 3D laser scanner, based on the trajectory data of the target city's infrastructure monitoring nodes. The system collects the spatial coordinates of the monitoring trajectories of all monitoring nodes for the corresponding target urban infrastructure monitoring objects online, and generates a dataset of urban infrastructure monitoring node object entity models, based on the external structure of the facility hardware of the monitoring nodes. , ;in The first item representing the target city's infrastructure regulatory object The entity model data of the urban infrastructure supervision node objects corresponding to the target urban infrastructure supervision nodes. This represents the maximum number of infrastructure monitoring nodes in the target city.

[0025] Preferably, the steps for generating normal 3D entity model data of the target urban infrastructure supervision node by searching and processing the normal 3D entity model information of the urban infrastructure supervision node based on the urban infrastructure object identification data and the normal entity model data of the urban infrastructure supervision node object are as follows:

[0026] S51. Establish a data set matrix of normal entity models for urban infrastructure regulatory node objects. ,in Indicates the first The set of normal entity model data of urban infrastructure supervision node objects corresponding to each urban infrastructure supervision object, wherein , ; This represents the normal entity model data set of the urban infrastructure regulatory node objects. The Middle Normal entity model data of urban infrastructure supervision node objects corresponding to each urban infrastructure supervision node. This represents the normal entity model data set of the urban infrastructure regulatory node objects. The maximum number of urban infrastructure regulatory nodes in China, where the normal entity model data of the urban infrastructure regulatory node object represents the spatial three-dimensional entity model parameters of the urban infrastructure regulatory node object standard setting for all regulatory nodes of the urban infrastructure regulatory object.

[0027] S52. Use the KMP search algorithm to identify the urban infrastructure object data. The data set matrix of the normal entity model of the urban infrastructure regulatory node object The set of normal entity model data of urban infrastructure regulatory node objects described in the article By performing character matching of the identification numbers of urban infrastructure regulatory objects, the identification data of the urban infrastructure objects can be retrieved. The corresponding urban infrastructure regulatory node object normal entity model data set And through data identification, a set of normal three-dimensional entity model data of the target city's infrastructure supervision nodes is generated. ,in The first item representing the target city's infrastructure regulatory object Normal 3D entity model data of the target city infrastructure supervision node corresponding to the target city infrastructure supervision node.

[0028] Preferably, the steps for performing anomaly diagnosis processing on the facility hardware structure of the urban infrastructure monitoring node based on the entity model data of the urban infrastructure monitoring node object and the normal three-dimensional entity model data of the target urban infrastructure monitoring node, and generating fault diagnosis result data of the target urban infrastructure, are as follows:

[0029] S61. Collect the entity model data set of the urban infrastructure supervision node objects. The entity model data of the urban infrastructure supervision node objects described in the article The set of normal three-dimensional entity model data of the target city infrastructure supervision node Normal 3D solid model data of the target city infrastructure monitoring node described in the article Entity model feature matching is performed in an orderly manner according to the infrastructure supervision node number of the target city. Based on the entity model feature matching results, a dataset of infrastructure fault diagnosis results for the target city is generated. The process generates a dataset of fault diagnosis results for the target city's infrastructure. The specific operating steps are as follows:

[0030] S611. Population initialization, setting the maximum number of update iterations T, and identifying the stray dog ​​population in the target city's infrastructure monitoring node's normal 3D entity model data set. The spatial position is randomly updated within the search space, and the formula for updating the spatial position is as follows: ,in, Indicates the first A set of normal 3D solid model data of individual stray dogs identified at the target city's infrastructure monitoring nodes. The search boundary position of the search space. and These represent facility malfunctions that identify individual stray dogs. The set of normal three-dimensional solid model data of the target city infrastructure supervision node The upper and lower boundaries in the search space, This represents a random number between the values ​​[0, 1].

[0031] S612, Group Attack Phase: Facility Fault Identification - Wild Dogs' Normal 3D Entity Model Data Set at the Target City Infrastructure Monitoring Node. Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure supervision node number to retrieve entity model data matching the target city's infrastructure supervision node objects. Matching the target city infrastructure monitoring node normal 3D entity model data The prey's location is identified and surrounded; the prey search simulation behavior formula is as follows: ,in This indicates the identification of individual stray dogs due to facility malfunction after the (t+1)th iteration in the group attack phase. The set of normal three-dimensional solid model data of the target city infrastructure supervision node A new location in the search space. Represents a random integer generated in reverse order of [2, N / 2], where N represents the size of the stray dog ​​population identified by facility failure; It is a subset of the wild dog population identified as a facility malfunction in preparation for an attack. This indicates the number of cumulative sums. This indicates the individual stray dog ​​identified due to facility malfunction after the t-th iteration. The set of normal three-dimensional solid model data of the target city infrastructure supervision node The position in the search space; This indicates the individual stray dog ​​identified due to facility malfunction after the t-th iteration. The set of normal three-dimensional solid model data of the target city infrastructure supervision node Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure supervision node number to retrieve entity model data matching the target city's infrastructure supervision node objects. The most matching normal 3D entity model data of the target city infrastructure supervision node Location, This represents a scale factor random number within the range [-2, 2] used to change the size of the stray dog ​​trajectory for facility fault identification.

[0032] S613, Persecution and Attack Phase: Facility Fault Identification Wilddogs use normal 3D entity model data sets of the target city's infrastructure monitoring nodes. Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure monitoring node number until entity model data matching the target city's infrastructure monitoring node object is captured. Matching the target city infrastructure monitoring node normal 3D entity model data The formula for simulated hunting behavior up to the prey is as follows: ,in This indicates that after the (t+1)th iteration in the persecution attack phase, the facility malfunctions and stray dog ​​individuals are identified. The set of normal three-dimensional solid model data of the target city infrastructure supervision node A new location in the search space; This represents a scale factor random number within the range [-2, 2]. This represents a random number within the range [-1, 1]. Represents random numbers The exponential parameter r represents a random number that takes values ​​in the interval [1, N]. This represents the set of normal 3D solid model data of the infrastructure monitoring node in the target city, randomly selected after the t-th iteration to identify individual stray dogs (r) for facility fault identification. The position in the search space, where i ≠ r;

[0033] S614, Cleaning Phase: The cleaning behavior is defined as the process when a facility malfunction is identified as a stray dog ​​in the target city infrastructure monitoring node's normal three-dimensional entity model data set. While traversing freely within the search space, entity model feature matching is performed in an orderly manner according to the target city's infrastructure monitoring node number to find entity model data matching the target city's infrastructure monitoring node object. Matching the target city infrastructure monitoring node normal 3D entity model data The formula for the behavior of eating carrion and the simulation behavior of cleaning is as follows: ,in This indicates the identification of individual stray dogs in the facility after iteration t+1 during the cleaning phase. The set of normal three-dimensional solid model data of the target city infrastructure supervision node A new location in the search space. This represents a binary number randomly generated by the algorithm. ∈{0,1};

[0034] S615. When the algorithm satisfies the maximum number of iterations T, output the entity model data of the urban infrastructure supervision node object. The normal three-dimensional solid model data of the target city infrastructure monitoring node The matching results of entity model features;

[0035] S616. Based on the entity model data of the urban infrastructure supervision node object described in step S615. The normal three-dimensional solid model data of the target city infrastructure monitoring node The entity model feature matching results generate a dataset of target city infrastructure fault diagnosis results. ,in The first item representing the target city's infrastructure regulatory object Data on fault diagnosis results of infrastructure in each target city corresponding to a target city infrastructure monitoring node;

[0036] when and Successful entity model feature matching indicates that the first... If the external structure of the infrastructure hardware of the target city's infrastructure monitoring node is normal, then the fault diagnosis result data of the target city's infrastructure will be output. This is normal;

[0037] when and The entity model feature matching failed, indicating that the first... If the external structure of the facility hardware of a target city infrastructure monitoring node fails, the fault diagnosis result data of the target city infrastructure will be output. This is abnormal.

[0038] Preferably, the steps for constructing the target city infrastructure abnormal entity model identification data and performing the city infrastructure supervision result display operation are as follows:

[0039] S71. Using the highlighting module in the 3D model rendering software, based on the target city's infrastructure fault diagnosis result data set... The target city infrastructure fault diagnosis result data described in the article The set of entity model data of the urban infrastructure supervision node The entity model data of the urban infrastructure regulatory node object that is abnormal in the fault diagnosis result of urban infrastructure. Perform color-coded rendering of the 3D solid model and generate a dataset of anomaly entity model identifiers for the target city's infrastructure. ,in The entity model data of the urban infrastructure monitoring node object indicating that the fault diagnosis result of urban infrastructure is abnormal. The corresponding target city infrastructure abnormal entity model identification data is generated after the three-dimensional entity model is rendered in the same color; the three-dimensional model rendering software includes any one of Navisworks, Three.js, and Unity3D.

[0040] S72, Collect the target city infrastructure abnormal entity model identification data set The data is transmitted online to the city management platform via the Internet of Things (IoT) communication network and displayed on a screen.

[0041] A digital twin-based intelligent monitoring system for urban infrastructure is used to implement the digital twin-based intelligent monitoring method for urban infrastructure. The system includes an urban infrastructure identification module, an urban infrastructure diagnosis module, and an urban infrastructure feedback module.

[0042] The urban infrastructure identification module includes an urban infrastructure feature information collection unit, an urban infrastructure standard feature information storage unit, and an urban infrastructure object identification unit.

[0043] The urban infrastructure feature information collection unit collects urban infrastructure feature text data through the urban management platform; the urban infrastructure standard feature information storage unit stores standard feature text data of different urban infrastructure objects; the urban infrastructure object identification unit performs urban infrastructure object identification processing based on the urban infrastructure feature text data and the standard feature text data of different urban infrastructure objects to generate urban infrastructure object identification data.

[0044] The urban infrastructure diagnostic module includes a standard trajectory storage unit for urban infrastructure monitoring nodes, a trajectory search unit for urban infrastructure monitoring nodes, an entity model acquisition unit for urban infrastructure monitoring objects, a normal entity model storage unit for urban infrastructure monitoring objects, a normal entity model search unit for urban infrastructure monitoring objects, and an anomaly diagnosis unit for urban infrastructure monitoring objects.

[0045] The urban infrastructure monitoring node standard trajectory storage unit is used to store urban infrastructure monitoring node standard trajectory data; the urban infrastructure monitoring node trajectory search unit performs path trajectory search processing of urban infrastructure monitoring nodes based on the urban infrastructure object identification data and the urban infrastructure monitoring node standard trajectory data to generate target urban infrastructure monitoring node trajectory data; the urban infrastructure monitoring object entity model acquisition unit performs 3D entity model construction processing of urban infrastructure monitoring node objects based on the target urban infrastructure monitoring node trajectory data and a UAV-mounted 3D laser scanner to generate urban infrastructure monitoring node object entity model data; the urban infrastructure monitoring object normal entity model storage unit is used to store urban infrastructure monitoring node object normal entity model data; the urban infrastructure monitoring object normal entity model search unit performs normal 3D entity model information search processing of urban infrastructure monitoring nodes based on the urban infrastructure object identification data and the urban infrastructure monitoring node object normal entity model data to generate target urban infrastructure monitoring node normal 3D entity model data; the urban infrastructure monitoring object anomaly diagnosis unit performs facility hardware structure anomaly diagnosis processing of urban infrastructure monitoring nodes based on the urban infrastructure monitoring node object entity model data and the target urban infrastructure monitoring node normal 3D entity model data to generate target urban infrastructure fault diagnosis result data;

[0046] The urban infrastructure feedback module includes an abnormal entity model identification unit for urban infrastructure supervision objects and an urban infrastructure supervision result display unit.

[0047] The abnormal entity model identification unit for urban infrastructure supervision objects constructs abnormal entity model identification data for target urban infrastructure based on the fault diagnosis results information of the target urban infrastructure, the entity model information of the urban infrastructure supervision node objects, and 3D model rendering software; the urban infrastructure supervision result display unit transmits the abnormal entity model identification data of the target urban infrastructure to the urban management platform online through the Internet of Things communication network and displays it on the display screen.

[0048] (III) Beneficial Effects

[0049] This invention provides a method and system for intelligent monitoring of urban infrastructure based on digital twins. It has the following beneficial effects:

[0050] I. Accurately obtain characteristic description information of urban infrastructure through the urban management platform to provide reliable data support for scientifically locating the identity information of urban infrastructure supervision objects; based on the characteristic text information of urban infrastructure combined with intelligent search algorithms and standard characteristic text information of different urban infrastructure objects based on big data storage, intelligent identification of the identity of urban infrastructure supervision objects is carried out to realize intelligent location and identification of urban infrastructure supervision objects and improve the scientific nature of urban infrastructure supervision.

[0051] Second, by combining urban infrastructure object identification information with intelligent search algorithms and scientifically pre-set standard trajectory information of urban infrastructure supervision nodes, the system accurately matches the specific supervision node path trajectory of urban infrastructure supervision objects, achieving precise and efficient planning and processing of supervision trajectories for different urban infrastructures, thus improving the reliability of urban infrastructure supervision; based on the trajectory information of target urban infrastructure supervision nodes, and using UAVs equipped with 3D laser scanners to autonomously and efficiently construct 3D entity models of urban infrastructure supervision node objects, the system achieves intelligent dynamic acquisition of 3D entity models of urban infrastructure supervision objects, providing real data support for accurate diagnosis of urban infrastructure hardware structure faults; based on urban infrastructure object identification information combined with intelligent search algorithms and normal entity model information of urban infrastructure supervision node objects, the system efficiently searches for standard normal 3D entity model information of urban infrastructure supervision objects; based on the entity model information of urban infrastructure supervision node objects and the normal 3D entity model information of target urban infrastructure supervision nodes, the system scientifically and dynamically diagnoses abnormalities in the hardware structure of urban infrastructure supervision nodes, realizing digital inspection of the external structure of urban infrastructure hardware and improving the work efficiency of urban infrastructure supervision.

[0052] Third, by combining the fault diagnosis results of the target city's infrastructure with the entity model information of the urban infrastructure supervision node objects and the 3D model rendering software, the abnormal entity model identification information of the target city's infrastructure is scientifically constructed, realizing the rendering processing of the abnormal facility entity model in the target city's infrastructure supervision objects; the abnormal entity model identification data of the target city's infrastructure is transmitted online to the city management platform through the Internet of Things communication network and displayed on the display screen, realizing the visualization supervision of urban infrastructure based on digital twin, and improving the intuitiveness and applicability of urban infrastructure supervision. Attached Figure Description

[0053] Fig. 1 A schematic diagram of the modules of the intelligent urban infrastructure monitoring system based on digital twins provided by the present invention;

[0054] Fig. 2 A flowchart of the intelligent monitoring method for urban infrastructure based on digital twins provided by the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] An example of the intelligent monitoring method and system for urban infrastructure based on digital twins is as follows:

[0057] Example 1:

[0058] Please see Figs. 1-2 A smart regulatory method for urban infrastructure based on digital twins, comprising the following steps:

[0059] S1. Collect textual data on urban infrastructure characteristics;

[0060] S2. Based on the text data of urban infrastructure features and the text data of standard features of different urban infrastructure objects, perform object recognition processing of urban infrastructure to generate urban infrastructure object recognition data.

[0061] S3. Based on the urban infrastructure object identification data and the standard trajectory data of urban infrastructure supervision nodes, perform path trajectory search processing of urban infrastructure supervision nodes to generate trajectory data of target urban infrastructure supervision nodes.

[0062] S4. Based on the trajectory data of the target city's infrastructure monitoring nodes, perform 3D entity model construction processing of the city's infrastructure monitoring node objects to generate entity model data of the city's infrastructure monitoring node objects.

[0063] S5. Based on the urban infrastructure object identification data and the normal entity model data of urban infrastructure supervision node objects, perform normal three-dimensional entity model information search and processing for urban infrastructure supervision nodes to generate normal three-dimensional entity model data of target urban infrastructure supervision nodes.

[0064] S6. Based on the entity model data of the urban infrastructure supervision node object and the normal three-dimensional entity model data of the target urban infrastructure supervision node, perform abnormal diagnosis and processing of the facility hardware structure of the urban infrastructure supervision node, and generate fault diagnosis result data of the target urban infrastructure.

[0065] S7. Construct the abnormal entity model identification data of the target city's infrastructure and execute the task of displaying the urban infrastructure supervision results.

[0066] For further details, please refer to Figs. 1-2The steps for collecting textual data on urban infrastructure features are as follows:

[0067] S11. Input the characteristic description text information of the target urban infrastructure to be monitored online through the urban management platform, and generate urban infrastructure characteristic text data. Urban infrastructure includes power hardware infrastructure, gas hardware infrastructure, heating hardware infrastructure, rail hardware infrastructure, and fire protection hardware infrastructure; the textual data on urban infrastructure features includes geographic administrative region features, geographical location features, and functional service range features of the target city's infrastructure.

[0068] The steps for generating urban infrastructure object identification data by performing object recognition processing on urban infrastructure feature text data and standard feature text data of different urban infrastructure objects are as follows:

[0069] S21. Establish a set of standard feature text data for different urban infrastructure objects. , ;in Indicates the first Standard feature text data of different urban infrastructure objects corresponding to each urban infrastructure regulatory object. This represents the maximum number of urban infrastructure regulatory objects; the standard feature text data for different urban infrastructure objects represents the standard feature description text information set for different urban infrastructure regulatory objects.

[0070] S22. Use the BERT language model search algorithm to extract urban infrastructure feature text data. Text data sets of standard features of different urban infrastructure objects Standard feature text data of different urban infrastructure objects in China Perform feature text character matching to search for text data related to urban infrastructure features. Matching standard feature text data of different urban infrastructure objects The corresponding text information of the urban infrastructure regulatory object number is used to generate urban infrastructure object identification data through data identification. .

[0071] The urban infrastructure feature information collection unit uses the urban management platform to accurately obtain the feature description information of urban infrastructure, providing reliable data support for scientifically locating the identity information of urban infrastructure regulatory objects; the urban infrastructure object identification unit uses the urban infrastructure feature text information combined with intelligent search algorithms and different urban infrastructure object standard feature text information based on big data storage to intelligently identify the identity of urban infrastructure regulatory objects, thereby realizing intelligent location and identification of urban infrastructure regulatory objects and improving the scientific nature of urban infrastructure supervision.

[0072] For further details, please refer to Figs. 1-2 The steps for generating target urban infrastructure monitoring node trajectory data by performing path trajectory search processing on urban infrastructure monitoring nodes based on urban infrastructure object identification data and standard trajectory data of urban infrastructure monitoring nodes are as follows:

[0073] S31. Establish a standard trajectory data set for urban infrastructure monitoring nodes. ,in Indicates the first Standard trajectory data of urban infrastructure supervision nodes corresponding to each urban infrastructure supervision object. The standard trajectory data of urban infrastructure supervision nodes represents the spatial coordinate information of the supervision trajectory of urban infrastructure supervision nodes set for different urban infrastructure supervision objects.

[0074] S32. Use a bidirectional search algorithm to identify urban infrastructure objects. Standard trajectory data set of urban infrastructure regulatory nodes Standard trajectory data of urban infrastructure supervision nodes Perform character matching of urban infrastructure object numbers to retrieve urban infrastructure object identification data. Corresponding standard trajectory data of urban infrastructure supervision nodes And generate trajectory data of infrastructure monitoring nodes in the target city by identifying data. The target city infrastructure monitoring node trajectory data represents the spatial coordinate information of the monitoring trajectory of all monitoring nodes of the target city infrastructure monitoring object.

[0075] The steps for generating the entity model data of urban infrastructure monitoring node objects by constructing a 3D solid model of the urban infrastructure monitoring node objects based on the trajectory data of the target urban infrastructure monitoring nodes are as follows:

[0076] S41. Using a drone equipped with a 3D laser scanner to analyze the trajectory data of the target city's infrastructure monitoring nodes. The system collects the spatial coordinates of the monitoring trajectories of all monitoring nodes for the corresponding target urban infrastructure monitoring objects online, and generates a dataset of urban infrastructure monitoring node object entity models, based on the external structure of the facility hardware of the monitoring nodes. , ;in The first item representing the target city's infrastructure regulatory object The entity model data of the urban infrastructure supervision node objects corresponding to the target urban infrastructure supervision nodes. This represents the maximum number of infrastructure monitoring nodes in the target city.

[0077] The steps for generating normal 3D entity model data for target urban infrastructure regulatory nodes by searching and processing the normal 3D entity model information of urban infrastructure regulatory nodes based on urban infrastructure object identification data and normal entity model data of urban infrastructure regulatory nodes are as follows:

[0078] S51. Establish a data set matrix of normal entity models for urban infrastructure regulatory node objects. ,in Indicates the first The set of normal entity model data of urban infrastructure supervision node objects corresponding to each urban infrastructure supervision object, wherein , ; This represents the normal entity model data set of urban infrastructure regulatory node objects. The Middle Normal entity model data of urban infrastructure supervision node objects corresponding to each urban infrastructure supervision node. This represents the normal entity model data set of urban infrastructure regulatory node objects. The maximum number of urban infrastructure regulatory nodes in China; the normal entity model data of urban infrastructure regulatory node objects represents the spatial three-dimensional entity model parameters of the urban infrastructure regulatory node objects standard settings for all regulatory nodes of urban infrastructure regulatory objects.

[0079] S52. Use the KMP search algorithm to identify urban infrastructure object data. Data set matrix of normal entity model of urban infrastructure regulatory node objects Normal entity model data set of urban infrastructure regulatory node objects By matching the characters of the identification numbers of urban infrastructure regulatory objects, the identification data of urban infrastructure objects can be retrieved. The corresponding urban infrastructure regulatory node object normal entity model data set And through data identification, a set of normal three-dimensional entity model data of the target city's infrastructure supervision nodes is generated. ,in The first item representing the target city's infrastructure regulatory object Normal 3D entity model data of the target city infrastructure supervision node corresponding to the target city infrastructure supervision node.

[0080] The steps for performing fault diagnosis and processing of the infrastructure hardware structure of urban infrastructure monitoring nodes based on the entity model data of urban infrastructure monitoring nodes and the normal 3D entity model data of the target urban infrastructure monitoring nodes, and generating fault diagnosis result data of the target urban infrastructure, are as follows:

[0081] S61. Collect entity model data of urban infrastructure supervision node objects. Entity model data of urban infrastructure regulatory nodes Data set of normal 3D entity models of the target city's infrastructure regulatory nodes Normal 3D entity model data of the target city infrastructure supervision node Entity model feature matching is performed in an orderly manner according to the infrastructure supervision node number of the target city. Based on the entity model feature matching results, a dataset of infrastructure fault diagnosis results for the target city is generated. Generate a dataset of infrastructure fault diagnosis results for the target city. The specific operating steps are as follows:

[0082] S611. Population initialization: Set the maximum number of update iterations T; Fault identification: Wild dog population in the target city infrastructure monitoring node normal 3D entity model data set. The spatial position is randomly updated within the search space, and the formula for updating the spatial position is as follows: ,in, Indicates the first A set of normal 3D solid model data of individual stray dogs at the target city's infrastructure monitoring nodes for facility fault identification. The search boundary position of the search space. and These represent facility malfunctions that identify individual stray dogs. Normal 3D solid model data set of the target city infrastructure supervision node The upper and lower boundaries in the search space, This represents a random number between the values ​​[0, 1].

[0083] S612, Group Attack Phase: Facility Fault Identification - Wild Dogs in the Target City Infrastructure Monitoring Node's Normal 3D Entity Model Data Set Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure regulatory node number to retrieve entity model data matching the target city's infrastructure regulatory node objects. Matching target city infrastructure regulatory node normal 3D entity model data The prey's location is identified and surrounded; the prey search simulation behavior formula is as follows: ,in This indicates the identification of individual stray dogs due to facility malfunction after the (t+1)th iteration in the group attack phase. Normal 3D solid model data set of the target city infrastructure supervision node A new location in the search space. Represents a random integer generated in reverse order of [2, N / 2], where N represents the size of the stray dog ​​population identified by facility failure; It is a subset of the wild dog population identified as a facility malfunction in preparation for an attack. This indicates the number of cumulative sums. This indicates the individual stray dog ​​identified due to facility malfunction after the t-th iteration. Normal 3D solid model data set of the target city infrastructure supervision node The position in the search space; This indicates the individual stray dog ​​identified due to facility malfunction after the t-th iteration. Normal 3D solid model data set of the target city infrastructure supervision node Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure regulatory node number to retrieve entity model data matching the target city's infrastructure regulatory node objects. The most matching target city infrastructure regulatory node normal 3D entity model data Location, This represents a scale factor random number within the range [-2, 2] used to change the size of the stray dog ​​trajectory for facility fault identification.

[0084] S613, Persecution and Attack Phase: Facility Fault Identification - Wild Dogs' Normal 3D Entity Model Data Set at Target City Infrastructure Monitoring Nodes Within the search space, entity model features are matched systematically according to the target city's infrastructure regulatory node numbers until entity model data matching the target city's infrastructure regulatory node objects is captured. Matching target city infrastructure regulatory node normal 3D entity model data The formula for simulated hunting behavior up to the prey is as follows: ,in This indicates that after the (t+1)th iteration in the persecution attack phase, the facility malfunctions and stray dog ​​individuals are identified. Normal 3D solid model data set of the target city infrastructure supervision node A new location in the search space; This represents a scale factor random number within the range [-2, 2]. This represents a random number within the range [-1, 1]. Represents random numbers The exponential parameter r represents a random number that takes values ​​in the interval [1, N]. This represents the set of 3D solid model data of randomly selected facility fault identification stray dog ​​individuals r at the target city infrastructure monitoring node after the t-th iteration. The position in the search space, where i ≠ r;

[0085] S614, Cleaning Phase: The cleaning behavior is defined as the process where, during the facility malfunction identification phase, stray dogs are detected in the target city's infrastructure monitoring node's normal 3D entity model data set. While traversing freely within the search space, entity model features are matched sequentially according to the target city's infrastructure monitoring node numbers to find entity model data matching the city's infrastructure monitoring node objects. Matching target city infrastructure regulatory node normal 3D entity model data The formula for the behavior of eating carrion and the simulation behavior of cleaning is as follows: ,in This indicates the identification of individual stray dogs in the facility after iteration t+1 during the cleaning phase. Normal 3D solid model data set of the target city infrastructure supervision node A new location in the search space. This represents a binary number randomly generated by the algorithm. ∈{0,1};

[0086] S615. When the algorithm satisfies the maximum number of iterations T, output the entity model data of the urban infrastructure supervision node object. Normal 3D entity model data of the target city's infrastructure regulatory nodes The matching results of entity model features;

[0087] S616, Based on the entity model data of the urban infrastructure supervision node object in step S615. Normal 3D entity model data of the target city's infrastructure regulatory nodes The entity model feature matching results generate a dataset of target city infrastructure fault diagnosis results. ,in The first item representing the target city's infrastructure regulatory object Data on fault diagnosis results of infrastructure in each target city corresponding to a target city infrastructure monitoring node;

[0088] when and Successful entity model feature matching indicates that the first... If the external structure of the infrastructure hardware of the target city's infrastructure monitoring node is normal, then the fault diagnosis result data of the target city's infrastructure will be output. This is normal;

[0089] when and The entity model feature matching failed, indicating that the first... If the external structure of the infrastructure hardware of a target city's infrastructure monitoring node fails, the target city's infrastructure fault diagnosis result data will be output. This is abnormal.

[0090] The urban infrastructure monitoring node trajectory search unit accurately matches the specific monitoring node path trajectory of urban infrastructure monitoring objects with the object identification information, intelligent search algorithm, and scientifically preset standard trajectory information of urban infrastructure monitoring nodes. This enables precise and efficient planning and processing of monitoring trajectories for different urban infrastructures, improving the reliability of urban infrastructure monitoring. The urban infrastructure monitoring object entity model acquisition unit autonomously and efficiently constructs 3D entity models of urban infrastructure monitoring node objects based on the target urban infrastructure monitoring node trajectory information and using a drone equipped with a 3D laser scanner. This achieves intelligent and dynamic acquisition of 3D entity models of urban infrastructure monitoring objects, providing real data support for accurate diagnosis of urban infrastructure hardware structural faults. The urban infrastructure monitoring object normal entity model search unit efficiently searches for standard normal 3D entity model information of urban infrastructure monitoring objects based on the object identification information, intelligent search algorithm, and normal entity model information of urban infrastructure monitoring node objects. The urban infrastructure monitoring object anomaly diagnosis unit scientifically and dynamically diagnoses anomalies in the hardware structure of urban infrastructure monitoring nodes based on the entity model information of urban infrastructure monitoring node objects and the normal 3D entity model information of target urban infrastructure monitoring nodes. This enables digital inspection of the external structure of urban infrastructure hardware, improving the efficiency of urban infrastructure monitoring.

[0091] For further details, please refer to Figs. 1-2 The steps for constructing the abnormal entity model identification data of the target city's infrastructure and performing the operation of displaying the urban infrastructure supervision results are as follows:

[0092] S71. The highlighting module in the 3D model rendering software is used based on the target city's infrastructure fault diagnosis result data set. Data on infrastructure fault diagnosis results in target cities Data set of entity models of urban infrastructure regulatory nodes The entity model data of urban infrastructure regulatory nodes with abnormal fault diagnosis results in China. Perform color-coded rendering of the 3D solid model and generate a dataset of anomaly entity model identifiers for the target city's infrastructure. ,in This indicates that the entity model data of the urban infrastructure regulatory node is abnormal, representing the fault diagnosis result of urban infrastructure. The corresponding target city infrastructure anomaly entity model identification data is generated after the 3D solid model is rendered in the same color; the 3D model rendering software includes any one of Navisworks, Three.js, and Unity3D.

[0093] S72, Collect the target city infrastructure anomaly entity model identification data set The data is transmitted online to the city management platform via the Internet of Things (IoT) communication network and displayed on a screen.

[0094] The abnormal entity model identification unit for urban infrastructure supervision objects scientifically constructs the abnormal entity model identification information of the target urban infrastructure based on the fault diagnosis results information of the target urban infrastructure, the entity model information of the urban infrastructure supervision node objects, and 3D model rendering software, thereby realizing the rendering processing of abnormal facility entity models in the target urban infrastructure supervision objects. The urban infrastructure supervision result display unit transmits the abnormal entity model identification data of the target urban infrastructure to the urban management platform online through the Internet of Things communication network, and displays it on the display screen, thereby realizing the visualization supervision of urban infrastructure based on digital twins and improving the intuitiveness and applicability of urban infrastructure supervision.

[0095] Example 2:

[0096] Please see Figs. 1-2 A digital twin-based intelligent monitoring system for urban infrastructure is used to realize a digital twin-based intelligent monitoring method for urban infrastructure. The system includes an urban infrastructure identification module, an urban infrastructure diagnosis module, and an urban infrastructure feedback module.

[0097] The urban infrastructure identification module includes an urban infrastructure feature information collection unit, an urban infrastructure standard feature information storage unit, and an urban infrastructure object identification unit;

[0098] The urban infrastructure feature information collection unit collects urban infrastructure feature text data through the urban management platform; the urban infrastructure standard feature information storage unit stores standard feature text data of different urban infrastructure objects; and the urban infrastructure object identification unit performs object identification processing on urban infrastructure based on the urban infrastructure feature text data and the standard feature text data of different urban infrastructure objects to generate urban infrastructure object identification data.

[0099] The urban infrastructure diagnostic module includes a standard trajectory storage unit for urban infrastructure monitoring nodes, a trajectory search unit for urban infrastructure monitoring nodes, an entity model acquisition unit for urban infrastructure monitored objects, a normal entity model storage unit for urban infrastructure monitored objects, a normal entity model search unit for urban infrastructure monitored objects, and an anomaly diagnosis unit for urban infrastructure monitored objects.

[0100] The system comprises the following components: a standard trajectory storage unit for urban infrastructure monitoring nodes, used to store standard trajectory data of urban infrastructure monitoring nodes; an urban infrastructure monitoring node trajectory search unit, which performs path trajectory search processing on urban infrastructure monitoring nodes based on urban infrastructure object identification data and standard trajectory data, generating trajectory data of the target urban infrastructure monitoring node; an urban infrastructure monitoring object entity model acquisition unit, which performs 3D entity model construction processing on urban infrastructure monitoring node objects based on the trajectory data of the target urban infrastructure monitoring node and a 3D laser scanner mounted on a UAV, generating entity model data of the urban infrastructure monitoring node objects; a normal entity model storage unit for urban infrastructure monitoring objects, used to store normal entity model data of urban infrastructure monitoring node objects; a normal entity model search unit for urban infrastructure monitoring objects, which performs normal 3D entity model information search processing on urban infrastructure monitoring nodes based on urban infrastructure object identification data and normal entity model data of urban infrastructure monitoring node objects, generating normal 3D entity model data of the target urban infrastructure monitoring node; and an urban infrastructure monitoring object anomaly diagnosis unit, which performs facility hardware structure anomaly diagnosis processing on urban infrastructure monitoring nodes based on entity model data of urban infrastructure monitoring nodes and normal 3D entity model data of the target urban infrastructure monitoring node, generating fault diagnosis result data of the target urban infrastructure.

[0101] The urban infrastructure feedback module includes an abnormal entity model identification unit for urban infrastructure supervision objects and an urban infrastructure supervision result display unit;

[0102] The urban infrastructure supervision object abnormal entity model identification unit constructs the target urban infrastructure abnormal entity model identification data based on the target urban infrastructure fault diagnosis result information, urban infrastructure supervision node object entity model information, and 3D model rendering software; the urban infrastructure supervision result display unit transmits the target urban infrastructure abnormal entity model identification data to the urban management platform online through the Internet of Things communication network and displays it on the display screen.

[0103] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent supervision of urban infrastructure based on digital twins, characterized in that: The method includes the following steps: S1. Collect textual data on urban infrastructure characteristics; S2. Perform object recognition processing on urban infrastructure to generate urban infrastructure object recognition data; S3. Perform path trajectory search and processing for urban infrastructure monitoring nodes to generate trajectory data of target urban infrastructure monitoring nodes; S4. Based on the trajectory data of the target city infrastructure monitoring node, perform three-dimensional solid model construction processing of the city infrastructure monitoring node object to generate solid model data of the city infrastructure monitoring node object. S5. Perform normal three-dimensional entity model information search and processing for urban infrastructure supervision nodes to generate normal three-dimensional entity model data of target urban infrastructure supervision nodes. S6. Perform abnormal diagnosis and processing of the hardware structure of urban infrastructure supervision nodes, and generate fault diagnosis result data of the target city infrastructure. The specific steps include: S61, collecting entity model data of urban infrastructure supervision node objects. The first target of urban infrastructure supervision Entity model data of urban infrastructure supervision nodes corresponding to each target urban infrastructure supervision node. Data set of normal 3D entity models of the target city's infrastructure regulatory nodes The first target of urban infrastructure supervision Normal 3D solid model data of the target city infrastructure supervision node. Entity model feature matching is performed in an orderly manner according to the infrastructure supervision node number of the target city. Based on the entity model feature matching results, a dataset of infrastructure fault diagnosis results for the target city is generated. Execute to generate the The specific operating steps are as follows: S611, Population initialization, setting the maximum number of update iterations T, facility fault identification of wild dog population as described in the section... Randomly update spatial locations within the search space; S612, During the group attack phase, facility malfunction identification wild dogs are described in the following... Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure regulatory node number to find entities matching the specified features. The matching Locate the prey and surround it; S613, Persecution and Attack Phase, Facility Failure Identification Wild Dogs in the... Within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure regulatory node numbers until a match is found. The matching Until the prey is caught; S614, Cleaning Phase: The cleaning behavior is defined as the action taken when a facility malfunction is detected and a stray dog ​​is identified in the following... While roaming freely within the search space, entity model features are matched in an orderly manner according to the target city's infrastructure regulatory node numbers to find entities that match the... The matching The act of eating carrion; S615. When the algorithm satisfies the maximum number of iterations T, output the above. With the The matching results of entity model features; S616, As described in step S615 With the The entity model feature matching results generate a dataset of target city infrastructure fault diagnosis results. ,in The first item representing the target city's infrastructure regulatory object Data on fault diagnosis results of infrastructure in each target city corresponding to a target city infrastructure monitoring node; when and If entity model feature matching is successful, then the output is as follows. This is normal; when and If the entity model features are not matched successfully, then output the following: This is abnormal; S7. Construct the abnormal entity model identification data of the target city's infrastructure and execute the task of displaying the urban infrastructure supervision results.

2. The intelligent monitoring method for urban infrastructure based on digital twins according to claim 1, characterized in that: S1 includes the following steps: S11. Input the characteristic description text information of the target urban infrastructure to be monitored online through the urban management platform, and generate urban infrastructure characteristic text data. .

3. The intelligent monitoring method for urban infrastructure based on digital twins according to claim 2, characterized in that: S2 includes the following steps: S21. Establish a set of standard feature text data for different urban infrastructure objects. , ;in Indicates the first Standard feature text data of different urban infrastructure objects corresponding to each urban infrastructure regulatory object. This represents the maximum number of urban infrastructure regulatory targets. S22. Using the BERT language model search algorithm to search the... With the The above Perform feature text character matching to search for characters that match the given text. The matching The corresponding text information of the urban infrastructure regulatory object number is used to generate urban infrastructure object identification data through data identification. .

4. The intelligent monitoring method for urban infrastructure based on digital twins according to claim 3, characterized in that: S3 includes the following steps: S31. Establish a standard trajectory data set for urban infrastructure monitoring nodes. ,in Indicates the first Standard trajectory data of urban infrastructure supervision nodes corresponding to each urban infrastructure supervision object; S32, Using a bidirectional search algorithm to... With the The above Perform character matching of urban infrastructure object numbers to search for the aforementioned The corresponding And generate trajectory data of infrastructure monitoring nodes in the target city by identifying data. .

5. The intelligent monitoring method for urban infrastructure based on digital twins according to claim 4, characterized in that: S4 includes the following steps: S41. Using a drone equipped with a 3D laser scanner, based on the... The system collects the spatial coordinates of the monitoring trajectories of all monitoring nodes for the corresponding target urban infrastructure monitoring objects online, and generates 3D solid model information of the external structure of the facility hardware of the urban infrastructure monitoring node objects. , ;in The first item representing the target city's infrastructure regulatory object The entity model data of the urban infrastructure supervision node objects corresponding to the target urban infrastructure supervision nodes. This represents the maximum number of infrastructure monitoring nodes in the target city.

6. The intelligent monitoring method for urban infrastructure based on digital twins according to claim 5, characterized in that: S5 includes the following steps: S51. Establish a data set matrix of normal entity models for urban infrastructure regulatory node objects. ,in Indicates the first The set of normal entity model data of urban infrastructure supervision node objects corresponding to each urban infrastructure supervision object, wherein , ; Indicates the The Middle Normal entity model data of urban infrastructure supervision node objects corresponding to each urban infrastructure supervision node. Indicates the The maximum number of urban infrastructure regulatory nodes in China; S52, Using the KMP search algorithm to... With the The above Perform character matching of the regulatory object numbers for urban infrastructure to search for the aforementioned The corresponding And generated through data identification ,in The first item representing the target city's infrastructure regulatory object Normal 3D entity model data of the target city infrastructure supervision node corresponding to the target city infrastructure supervision node.

7. The intelligent monitoring method for urban infrastructure based on digital twins according to claim 6, characterized in that: S7 includes the following steps: S71, using the highlighting module in the 3D model rendering software according to the above... The above In the The above refers to the abnormal results of urban infrastructure fault diagnosis. Perform color-coded rendering of the 3D solid model and generate a dataset of anomaly entity model identifiers for the target city's infrastructure. ,in The statement indicating that the fault diagnosis result of urban infrastructure is abnormal The corresponding target city infrastructure abnormal entity model identification data is generated after the three-dimensional entity model is rendered in the same color. S72, the above The data is transmitted online to the city management platform via the Internet of Things (IoT) communication network and displayed on a screen.

8. A digital twin-based intelligent monitoring system for urban infrastructure, used to implement the digital twin-based intelligent monitoring method for urban infrastructure as described in any one of claims 1-7, characterized in that: The system includes an urban infrastructure identification module, an urban infrastructure diagnosis module, and an urban infrastructure feedback module.

Images