Digital Twin Management Method for Urban Infrastructure Supported by Beidou Time-Space Base

Abstract

The present invention relates to the technical field of smart cities, and specifically relates to a digital twin management method for urban infrastructure supported by a Beidou spatio-temporal base, including the following steps: S1, spatio-temporal reference grid construction and coding assignment: Assign Beidou grid codes to each urban infrastructure entity; S2, multi-modal data collection and spatio-temporal alignment processing: Generate a perception data stream with spatio-temporal tags; S3, feature extraction: Generate an infrastructure status feature matrix; S4, spatio-temporal rule knowledge base establishment and anomaly detection: Generate spatio-temporal anomaly vectors; S5, spatio-temporal memory enhanced prediction and early warning interval generation: Generate a spatio-temporal early warning interval for a prediction deviation threshold; S6, superposition analysis of anomalies and early warnings and causal reasoning optimization: Generate a maintenance decision instruction with Beidou grid code positioning; The present invention improves the reliability and maintenance efficiency of urban infrastructure and provides strong technical support for the intelligent operation and maintenance of urban infrastructure.

Description

Technical Field

[0001] The present invention relates to the field of smart city technology, and in particular to a digital twin management method for urban infrastructure supported by a Beidou space-time base. Background Art

[0002] With the acceleration of urbanization, the scale of urban infrastructure is expanding, involving a wide variety of facilities, including transportation, energy, communications and other facilities. In order to ensure the efficient, safe and reliable operation of these infrastructures, traditional urban infrastructure management methods can no longer meet the needs of modern urban management. The monitoring, early warning and maintenance of urban infrastructure urgently need to rely on advanced technical means, especially the integrated application of technologies such as the Internet of Things, big data and artificial intelligence. Digital twin technology, as a technology that maps physical entities through virtual models, can provide more accurate and real-time data support for the management of urban infrastructure, and is one of the key technologies for realizing smart city management.

[0003] Although the existing urban infrastructure management systems have made certain progress in some aspects, they generally have the following shortcomings. On the one hand, most traditional infrastructure management systems use static models and lack accurate perception and prediction of the real-time status of infrastructure, resulting in poor timeliness of fault prediction and maintenance decisions. On the other hand, the existing systems have limited capabilities in data fusion and spatiotemporal correlation analysis, and are unable to process multi-source and multi-dimensional sensor data, nor can they effectively perform spatiotemporal alignment and optimization of real-time data. In addition, traditional methods rely on experience and manual judgment in anomaly detection, fault diagnosis and maintenance decisions, resulting in frequent misjudgments and missed judgments. Existing technologies also lack adaptive learning mechanisms based on dynamic data, making it difficult to adapt to the rapidly changing urban infrastructure environment.

[0004] The purpose of this invention is to provide a digital twin management method for urban infrastructure based on the Beidou space-time base support, realize an adaptive intelligent management system, and effectively improve the operation efficiency and maintenance quality of urban infrastructure. Summary of the invention

[0005] The present invention provides a digital twin management method for urban infrastructure supported by a Beidou space-time base.

[0006] The urban infrastructure digital twin management method supported by the BeiDou space-time base includes the following steps:

[0007] S1, spatiotemporal reference grid construction and code allocation: Based on the satellite-based ground-based augmented data of the BeiDou-3 system, a centimeter-level spatiotemporal reference grid covering the target area is generated, and a BeiDou grid code with spatiotemporal uniqueness is allocated to each urban infrastructure entity;

[0008] S2, Multimodal Data Acquisition and Spatiotemporal Alignment Processing: Collect real-time status data through multimodal sensors deployed on urban infrastructure, align the real-time status data with the Beidou grid code in terms of space and time, and generate a perception data stream with spatiotemporal tags;

[0009] S3, Feature Extraction: Construct a digital twin evolution model based on a spatiotemporal graph neural network (ST-GNN), input the perception data stream with spatiotemporal tags as input data, and generate an infrastructure status feature matrix with spatiotemporal correlation characteristics;

[0010] S4, Establishment of Spatiotemporal Rule Knowledge Base and Anomaly Detection: Establish a spatiotemporal rule knowledge base for the evolution of urban infrastructure status, detect the abnormal status of urban infrastructure by comparing the infrastructure status feature matrix with the historical evolution patterns in the spatiotemporal rule knowledge base, and generate a spatiotemporal anomaly vector;

[0011] S5, Spatiotemporal Memory-Enhanced Prediction and Generation of Early Warning Intervals: Construct a spatiotemporal memory-enhanced prediction model, based on the spatiotemporal distribution characteristics of the current infrastructure status feature matrix and the Beidou grid code, predict the evolution trajectory of the infrastructure status within a set future time window, and generate a spatiotemporal early warning interval with a prediction deviation threshold;

[0012] S6, Superposition Analysis of Anomaly and Early Warning and Optimization of Causal Reasoning: Conduct superposition analysis on the spatiotemporal anomaly vector and the spatiotemporal early warning interval, dynamically adjust the search space of the causal reasoning engine, output a fault diagnosis report with spatiotemporal coordinates, and generate a maintenance decision instruction with Beidou grid code positioning according to the fault diagnosis report.

[0013] Optionally, the spatiotemporal reference grid construction and coding assignment in S1 include:

[0014] S11, Acquisition of Space-Based Augmentation Data: Obtain the original observation data of the target area through the space-based augmentation service (BDSBAS) of the Beidou-3 system and the ground-based augmentation station network (CORS);

[0015] S12, Spatiotemporal Reference Grid Division and Hierarchical Design: Design a multi-level spatiotemporal grid division scheme according to the spatial scope and application requirements of the target area, specifically including:

[0016] Spatial Hierarchy: Divide the target area into regular grid cells with adjustable side lengths, and the minimum cell size is 0.5m × 0.5m;

[0017] Time Hierarchy: Define time slices based on the Beidou timing signal, and the minimum time resolution is 1 second;

[0018] S13, Generation of Grid Coding and Spatiotemporal Binding: Adopt the Beidou grid coding algorithm to generate a four-dimensional code including longitude, latitude, altitude, and time for each grid cell, and the four-dimensional code structure meets the conditions:

[0019] ;

[0020] Among them, , , , are respectively the longitude, latitude, elevation and time of the grid center point, , , , are the coordinates of the spatio-temporal reference origin, , , , are the grid space and time resolutions, is the Beidou standard hash function, is the Beidou grid code;

[0021] S14, Infrastructure Entity Coding Mapping: Map the spatial coordinates and the temporal state of the urban infrastructure entity to the corresponding grid code through the dynamic interpolation algorithm.

[0022] Optionally, the multi-modal data acquisition and spatio-temporal alignment processing in S2 includes:

[0023] S21, Multi-modal Sensor Deployment and Data Acquisition: Deploy a multi-modal sensor network on urban infrastructure, including structural health monitoring sensors, environmental sensors and operating status sensors, and collect multi-modal data in real time, including structural health data, environmental data and operating status data, where;

[0024] The structural health data includes strain, vibration, displacement;

[0025] The environmental data includes temperature, humidity, wind speed;

[0026] The operating status data includes load, flow, pressure;

[0027] S22, Data Preprocessing: Preprocess the collected multi-modal data, including outlier removal, missing value filling, and noise filtering;

[0028] S23, Spatio-temporal Reference Unification and Data Alignment: Align the preprocessed multi-modal data with the Beidou grid code in terms of space and time, specifically including:

[0029] Spatial Alignment: Map the data in the local coordinate system of the sensor to the Beidou grid coordinate system through the coordinate transformation matrix;

[0030] Time Alignment: Synchronize and calibrate the sensor timestamps based on the Beidou time signal;

[0031] S24, Spatiotemporal Tag Generation and Data Stream Encapsulation: Add spatiotemporal tags to the aligned multimodal data to generate a standardized perception data stream.

[0032] Optionally, the data preprocessing in S22 includes:

[0033] S221, Outlier Removal: Remove abnormal data in the multimodal data based on criteria.

[0034] S222, Missing Value Filling: Use spatiotemporal interpolation to supplement the missing data in the multimodal data;

[0035] S223, Noise Filtering: Remove high-frequency noise in the multimodal data through wavelet transform.

[0036] Optionally, the feature extraction in S3 includes:

[0037] S31, Spatiotemporal Graph Structure Construction: Construct a dynamic spatiotemporal graph structure with Beidou grid codes as nodes and spatiotemporal correlation relationships as edges , where the node features are represented as , is the perception data of the th grid, is the embedding vector of the Beidou grid code, and the edge weight is represented as , is the spatial attenuation factor;

[0038] S32, Spatial Graph Convolution Calculation: Use a gated graph convolutional layer to extract spatial features;

[0039] S33, Temporal Gated Recurrent Unit: Model the temporal features of each node;

[0040] S34, Spatiotemporal Attention Mechanism: Dynamically calculate the importance weights of nodes ;

[0041] S35, Feature Matrix Generation: Sort the output of the spatiotemporal graph neural network according to the Beidou grid code to generate an infrastructure status feature matrix ;

[0042] S36, Model Training: Use a hybrid loss function for end-to-end training.

[0043] Optionally, the establishment of the spatiotemporal rule knowledge base and anomaly detection in S4 include:

[0044] S41. Construction of spatio-temporal rule knowledge base: Extract spatio-temporal patterns of the evolution of urban infrastructure status based on historical status data. Using multi-dimensional association analysis technology, abstract the spatio-temporal relationship of infrastructure status changes into rule triples (trigger conditions, spatio-temporal constraints, evolution results), and perform dimensionality reduction storage on high-dimensional relationships through spatio-temporal tensor decomposition technology to form a spatio-temporal rule knowledge base containing normal status evolution paths and fault modes;

[0045] S42. Spatio-temporal anomaly detection and vector generation: Compare the generated infrastructure status feature matrix with historical evolution patterns in the spatio-temporal rule knowledge base in multiple dimensions. Through spatio-temporal similarity calculation and dynamic threshold judgment, identify abnormal states deviating from the normal evolution path. For the detected abnormal states, generate spatio-temporal anomaly vectors including anomaly types, spatio-temporal coordinates, and deviation degrees, and at the same time mark potentially associated anomaly propagation paths.

[0046] Optionally, the construction of the spatio-temporal rule knowledge base in S41 includes:

[0047] S411. Preprocessing of historical data and generation of spatio-temporal sequences: Perform spatio-temporal alignment and normalization processing on historical status data, divide spatio-temporal units according to Beidou grid codes, and generate a spatio-temporal sequence data set of infrastructure status evolution , where a single spatio-temporal sequence is expressed as:

[0048] ;

[0049] Among them, is the feature matrix of the th spatio-temporal unit at time , , is the total number of spatio-temporal units, is the number of time steps;

[0050] S412. Mining of multi-dimensional association rules: Use an improved Apriori algorithm to extract spatio-temporal association rules. The rule form is a triple , where is the trigger condition, is the spatio-temporal constraint, is the evolution result;

[0051] S413. Spatio-temporal tensor decomposition and dimensionality reduction storage: Construct a four-dimensional spatio-temporal tensor , the dimensions represent spatial grid, time window, status feature, and association rule respectively, and perform dimensionality reduction through high-order singular value decomposition (HOSVD);

[0052] S414. Dynamic update of the rule knowledge base: Design an incremental tensor decomposition algorithm. When new data is added, update the rule tensor.

[0053] Optionally, the spatio-temporal anomaly detection and vector generation in S42 include:

[0054] S421, spatio-temporal similarity calculation: Using the spatio-temporal weighted cosine similarity algorithm, calculate the similarity between the current feature matrix and the historical feature matrix in the knowledge base ; ;

[0055] S422, dynamic threshold judgment: Based on the statistical characteristics of the historical similarity distribution in the knowledge base, dynamically set the anomaly judgment threshold ;

[0056] S423, anomaly vector generation: Generate spatio-temporal anomaly vectors for the anomaly states that meet ; ;

[0057] S424, anomaly propagation path marking: Based on the spatio-temporal graph structure and the causal rules in the knowledge base, use the random walk algorithm to predict the anomaly propagation path .

[0058] Optionally, the spatio-temporal memory enhanced prediction and early warning interval generation in S5 include:

[0059] S51, spatio-temporal memory pool construction: Construct a spatio-temporal memory pool based on the historical infrastructure state feature matrix ;

[0060] S52, gated spatio-temporal attention mechanism: Calculate the attention weights between the current feature matrix and the historical feature matrix in the memory pool ;

[0061] S53, prediction trajectory generation: Generate the prediction trajectory for the future time window through the gated recurrent unit (GRU) ;

[0062] S54, spatio-temporal early warning interval calculation: Generate a spatio-temporal early warning interval based on the confidence of the prediction trajectory .

[0063] Optionally, the anomaly and early warning superposition analysis and causal reasoning optimization in S6 include:

[0064] S61, anomaly and early warning superposition analysis: Perform spatio-temporal superposition analysis on the spatio-temporal anomaly vector and the spatio-temporal early warning interval to calculate the overlap coefficient of the anomaly within the early warning interval ;

[0065] S62, Dynamic Adjustment of the Search Space of the Causal Reasoning Engine: Based on the overlap coefficient Dynamically adjust the search space of the causal reasoning engine, specifically including:

[0066] Spatial Constraint: Limit the search range to the spatio-temporal region centered on the Beidou grid code where the abnormal event occurs with a radius , where (take 10 times the minimum unit of the Beidou grid code);

[0067] Rule Screening: Only retain the subset of causal rules related to the current abnormal type Type;

[0068] S63, Generation of Fault Diagnosis Report: Based on the output of the causal reasoning engine, generate a fault diagnosis report with spatio-temporal coordinates, including the Beidou grid code where the root cause event occurs (the spatial location where the fault or abnormality occurs), the timestamp when the root cause event occurs (the specific time when the fault occurs), and the diagnostic confidence (the reliability of the fault diagnosis);

[0069] S64, Generation of Maintenance Decision Instructions: Generate maintenance decision instructions with Beidou grid code positioning based on the fault diagnosis report, including priority sorting (sort the root cause events in descending order of confidence) and path planning (generate the optimal maintenance path based on the spatial distribution of the Beidou grid codes).

[0070] Advantages of the present invention:

[0071] In the present invention, through the construction of the spatio-temporal reference grid and the encoding assignment, a Beidou grid code with spatio-temporal uniqueness is assigned to each infrastructure entity, realizing high-precision spatio-temporal positioning and data association. Combined with multi-modal data collection and spatio-temporal alignment processing, it can obtain and accurately process the operation data of urban infrastructure in real time, significantly improving the integrity and consistency of data collection, and providing a reliable data basis for subsequent state feature extraction, anomaly detection, and prediction models.

[0072] In the present invention, through the spatio-temporal memory-enhanced prediction and early warning interval generation mechanism, combined with the spatio-temporal graph neural network model and the spatio-temporal attention mechanism, the accuracy of urban infrastructure state prediction and the timeliness of early warning are effectively improved. Through the spatio-temporal rule knowledge base constructed based on historical state data and the dynamically optimized causal reasoning engine, it can accurately detect abnormal states deviating from the normal evolution path, and provide timely fault diagnosis reports and maintenance decision instructions, not only improving the accuracy in the fault diagnosis process, but also enhancing the adaptability and robustness of the prediction model in long-term operation, ensuring the efficient operation of urban infrastructure.

[0073] In the present invention, through the closed-loop interaction between the digital twin and the physical entity, the adaptive optimization of the intelligent management system is achieved, enhancing the adaptability of the system to complex scenarios. During the maintenance decision-making process, based on the superimposed analysis and causal reasoning optimization of spatio-temporal anomalies and early warning intervals, the system can generate the optimal maintenance path according to the diagnostic results, provide accurate fault location and repair instructions, improving the fault response speed and decision-making efficiency of the infrastructure. Through this intelligent and real-time management method, the reliability and maintenance efficiency of urban infrastructure are significantly improved, providing strong technical support for the intelligent operation and maintenance of urban infrastructure. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0075] Figure 1 Schematic diagram of the management method process for the embodiment of the present invention;

[0076] Figure 2 Schematic diagram of spatio-temporal memory enhanced prediction and early warning interval generation for the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0077] The present invention will be described in detail below with reference to the drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; and the drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0078] It should be pointed out that in the specification, it is mentioned that "an embodiment", "embodiments", "exemplary embodiments", "some embodiments", etc. indicate that the described embodiments may include specific features, structures or characteristics, but not necessarily every embodiment includes such specific features, structures or characteristics. In addition, when combining embodiments to describe specific features, structures or characteristics, implementing such features, structures or characteristics in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0079] Generally, terms can be understood at least in part from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or property in the singular sense, or can be used to describe a combination of features, structures, or properties in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but rather can alternatively, depending at least in part on the context, allow for the existence of other factors that are not necessarily explicitly described.

[0080] As Figure 1 - Figure 2 shown, the digital twin management method for urban infrastructure supported by the Beidou spatio-temporal base includes the following steps:

[0081] S1, Spatio-temporal reference grid construction and coding assignment: Based on the satellite-based augmentation data of the Beidou-3 system, generate a centimeter-level spatio-temporal reference grid covering the target area, and assign a Beidou grid code with spatio-temporal uniqueness to each urban infrastructure entity;

[0082] S2, Multi-modal data acquisition and spatio-temporal alignment processing: Collect real-time status data through multi-modal sensors deployed on urban infrastructure, align the real-time status data with the Beidou grid code in space and time, and generate a perception data stream with spatio-temporal tags;

[0083] S3, Feature extraction: Construct a digital twin evolution model based on spatio-temporal graph neural network (ST-GNN), input the perception data stream with spatio-temporal tags as input data, and generate an infrastructure status feature matrix with spatio-temporal correlation characteristics;

[0084] S4, Spatio-temporal rule knowledge base establishment and anomaly detection: Establish a spatio-temporal rule knowledge base for the evolution of urban infrastructure status, detect the abnormal status of urban infrastructure and generate a spatio-temporal anomaly vector by comparing the infrastructure status feature matrix with the historical evolution patterns in the spatio-temporal rule knowledge base;

[0085] S5, Spatio-temporal memory enhanced prediction and warning interval generation: Construct a spatio-temporal memory enhanced prediction model, based on the spatio-temporal distribution characteristics of the current infrastructure status feature matrix and the Beidou grid code, predict the evolution trajectory of the infrastructure status within a set future time window, and generate a spatio-temporal warning interval with a prediction deviation threshold;

[0086] S6, Abnormality and warning superposition analysis and causal reasoning optimization: Perform superposition analysis on the spatio-temporal anomaly vector and the spatio-temporal warning interval, dynamically adjust the search space of the causal reasoning engine, output a fault diagnosis report with spatio-temporal coordinates, and generate a maintenance decision instruction with Beidou grid code positioning according to the fault diagnosis report;

[0087] Through the above, the precise alignment and spatio-temporal coupling analysis of multi-source data of urban infrastructure are realized. Combining the dynamic early warning interval generation and causal reasoning optimization mechanism, the accuracy of anomaly diagnosis and the timeliness of maintenance decision-making are significantly improved. At the same time, through the closed-loop interaction between the digital twin and the physical entity, an adaptive intelligent management system for urban infrastructure is formed.

[0088] The spatio-temporal reference grid construction and coding assignment in S1 include:

[0089] S11, Acquisition of satellite-based augmentation data: The original observation data of the target area are obtained through the satellite-based augmentation service (BDSBAS) of the Beidou-3 system and the ground-based augmentation network (CORS).

[0090] S12, Spatio-temporal reference grid division and hierarchical design: According to the spatial scope and application requirements of the target area, a multi-level spatio-temporal grid division scheme is designed, specifically including:

[0091] Spatial level: The target area is divided into regular grid cells with adjustable side lengths, and the minimum cell size is 0.5m × 0.5m.

[0092] Temporal level: Time slices are defined based on the Beidou time signal, and the minimum time resolution is 1 second.

[0093] S13, Grid code generation and spatio-temporal binding: The Beidou grid coding algorithm is used to generate a four-dimensional code including longitude, latitude, elevation, and time for each grid cell. The conditions satisfied by the four-dimensional code structure are:

[0094] ;

[0095] Among them, , , , are the longitude, latitude, elevation, and time of the grid center point respectively, , , , are the spatio-temporal reference origin coordinates, , , , are the grid space and time resolutions, is the Beidou standard hash function, with an output of a 128-bit code, is the Beidou grid code;

[0096] S14, Infrastructure entity coding mapping: The spatial coordinates and the temporal state of the urban infrastructure entity are mapped to the corresponding grid code through the dynamic interpolation algorithm, expressed as:

[0097] ;

[0098] wherein, is the Euclidean distance from the entity to the center of the th adjacent grid, is the code of the th adjacent grid, is the infrastructure entity code, is the Euclidean distance from the entity to the center of the th adjacent grid;

[0099] Through the above content, the satellite-based augmentation data of the Beidou-3 system constructs a centimeter-level spatio-temporal reference grid, and combines a multi-level grid division and a four-dimensional coding algorithm to assign a Beidou grid code with spatio-temporal uniqueness to each urban infrastructure entity, solving problems such as inconsistent spatio-temporal reference, coding redundancy, and difficult cross-system data fusion in traditional methods, significantly improving the accuracy and efficiency of urban infrastructure management. At the same time, a smooth mapping of entity codes is achieved through a dynamic interpolation algorithm to ensure coding consistency and scalability in complex scenarios, providing a reliable spatio-temporal reference support for the efficient operation of digital twin cities.

[0100] The multi-modal data acquisition and spatio-temporal alignment processing in S2 include:

[0101] S21, multi-modal sensor deployment and data acquisition: Deploy a multi-modal sensor network on urban infrastructure, including structural health monitoring sensors, environmental sensors, and operating state sensors, and collect multi-modal data in real time, including structural health data, environmental data, and operating state data, wherein;

[0102] Structural health data includes strain, vibration, and displacement;

[0103] Environmental data includes temperature, humidity, and wind speed;

[0104] Operating state data includes load, flow, and pressure;

[0105] S22, data preprocessing: Preprocess the collected multi-modal data, including outlier removal, missing value filling, and noise filtering;

[0106] S23, spatio-temporal reference unification and data alignment: Align the preprocessed multi-modal data with the Beidou grid code, specifically including:

[0107] Spatial alignment: Map the data in the local coordinate system of the sensor to the Beidou grid coordinate system through a coordinate transformation matrix, expressed as:

[0108] ;

[0109] wherein, , , are the coordinates in the grid coordinate system, coordinates, coordinates, coordinates, is the rotation matrix, , , are the coordinates in the local coordinate system, coordinates, coordinates, coordinates, , , are respectively direction, direction, the translation amounts in the direction;

[0110] Time alignment: Synchronize and calibrate the sensor timestamps based on the Beidou timing signal to eliminate the clock drift error;

[0111] S24, Spatiotemporal label generation and data stream encapsulation: Add spatiotemporal labels to the aligned multi-modal data to generate a standardized perception data stream, and the label format is:

[0112] ;

[0113] Among them, is the Beidou grid code corresponding to the data, is the Beidou synchronization timestamp, is the data type identifier (such as strain, temperature, etc.), is the data quality score, is the spatiotemporal label;

[0114] Through the above content, deploying a multi-modal sensor network and achieving high-precision spatiotemporal alignment of data based on the Beidou grid code solves the problems of inconsistent spatiotemporal benchmarks and difficult fusion of multi-source data in traditional methods, significantly improves the integrity and consistency of data collection. At the same time, by encapsulating the perception data stream with standardized spatiotemporal labels, it supports efficient data retrieval and quality control, provides a reliable data foundation for the digital twin management of urban infrastructure, and enhances the real-time performance, scalability and analysis accuracy of the system.

[0115] The data preprocessing in S22 includes:

[0116] S221, Outlier removal: Remove the abnormal data in the multi-modal data based on the criterion, expressed as:

[0117] ;

[0118] Among them, is the value of the th data point, is the mean value of the data sequence, is the standard deviation of the data sequence;

[0119] ;

[0120] ;

[0121] Among them, is the length of the data sequence;

[0122] S222, Missing value filling: The spatio-temporal interpolation method is used to supplement the missing data in the multimodal data, which is expressed as:

[0123] ;

[0124] ;

[0125] ;

[0126] Among them, is the missing value to be filled, is the value of the th adjacent data point, is the th weight of the adjacent data point, is the spatio-temporal distance between the missing point and the th adjacent point, is the timestamp of the missing point, is the th timestamp of the adjacent point, is the Beidou grid code of the missing point, is the th Beidou grid code of the adjacent point, is the weight coefficient of the spatial distance and the time distance (take ), is the number of adjacent points participating in the interpolation (take );

[0127] S223, Noise filtering: The high-frequency noise in the multimodal data is removed by wavelet transform, specifically including:

[0128] Wavelet decomposition: ;

[0129] Among them, is the original signal, is the wavelet basis function, is the scaling function, is the wavelet coefficient (high-frequency component), is the scale coefficient (low-frequency component), is the decomposition level;

[0130] Threshold denoising: Perform soft threshold processing on the wavelet coefficients which is expressed as:

[0131] ;

[0132] where is the threshold, taking ( is the noise standard deviation, is the signal length), is the wavelet coefficient after denoising;

[0133] Wavelet reconstruction: ;

[0134] where is the signal after denoising;

[0135] Through the above content, the quality and reliability of multi-modal data are improved. Outlier removal effectively removes obvious error data, missing value filling ensures the spatio-temporal continuity of data, and noise filtering accurately separates and removes high-frequency noise through wavelet transform, enhancing the robustness of the system and the accuracy of the analysis results.

[0136] The feature extraction in S3 includes:

[0137] S31, Construction of spatio-temporal graph structure: Using Beidou grid code as nodes and spatio-temporal correlation relationship as edges, construct a dynamic spatio-temporal graph structure where the node feature is expressed as is the perception data of the th grid, is the embedding vector of the Beidou grid code, and the edge weight is expressed as is the spatial attenuation factor;

[0138] S32, Spatial graph convolution calculation: Adopt gated graph convolution layer to extract spatial features, which is expressed as:

[0139] ;

[0140] where is the hidden state of the th layer and the th node, is the hidden state of the th layer and the th node, , are trainable weight matrices, ​​is the Hadamard product, is the Sigmoid function, is the th hidden state of the th node, is the degree of node is the degree of node ; is the activation function, is the set of neighbors of node ;

[0141] S33, Temporal Gated Recurrent Unit: Models the temporal features of each node, expressed as:

[0142] ;

[0143] ;

[0144] ;

[0145] ;

[0146] where, , , are the temporal gated weight matrices, is the time step, is the update gate, is the reset gate, is the candidate hidden state, is the updated hidden state, is the hidden state at the current time step, is the hidden state at the previous time step;

[0147] S34, Spatiotemporal Attention Mechanism: Dynamically calculates the node importance weights , expressed as:

[0148] ;

[0149] where, is the attention vector, is the attention weight matrix, is the attention weight between node and node , is the hidden state of node , is the hidden state of node , is the activation function, ;

[0150] S35, Feature Matrix Generation: Sort the output of the spatio-temporal graph neural network according to the Beidou grid code to generate the infrastructure status feature matrix , expressed as:

[0151] ;

[0152] Among them, is the final time step, is the total number of grids, is the node 's hidden state at the final time step, ;

[0153] S36, Model Training: Adopt a hybrid loss function for end-to-end training, expressed as:

[0154] ;

[0155] Among them, is the true state label, is the prior distribution of physical constraints, , are balance coefficients, is the hybrid loss function, is the mean absolute error, is the KL divergence, is the model output distribution;

[0156] Through the above content, the feature extraction model based on the spatio-temporal graph neural network, combined with spatial graph convolution, time-gated recurrent unit and spatio-temporal attention mechanism, realizes the efficient fusion and feature extraction of multi-source data of urban infrastructure. Spatial graph convolution captures the spatial dependence relationship between nodes, the time-gated recurrent unit models the temporal dynamic changes, and the spatio-temporal attention mechanism dynamically adjusts the node importance. Finally, an infrastructure status feature matrix with spatio-temporal correlation characteristics is generated, significantly improving the accuracy and robustness of feature expression.

[0157] The establishment of the spatio-temporal rule knowledge base and anomaly detection in S4 include:

[0158] S41, Spatio-Temporal Rule Knowledge Base Construction: Extract the spatio-temporal patterns of the evolution of urban infrastructure status based on historical state data, adopt multi-dimensional association analysis technology, abstract the spatio-temporal relationship of infrastructure status changes into interpretable rule triples (trigger conditions, spatio-temporal constraints, evolution results), and use spatio-temporal tensor decomposition technology to reduce the dimension of high-dimensional relationships for storage, forming a spatio-temporal rule knowledge base containing normal state evolution paths and fault patterns;

[0159] S42, Spatiotemporal Anomaly Detection and Vector Generation: Compare the generated infrastructure status feature matrix with the historical evolution patterns in the spatiotemporal rule knowledge base in multiple dimensions. Through spatiotemporal similarity calculation and dynamic threshold judgment, identify the abnormal states that deviate from the normal evolution path. For the detected abnormal states, generate spatiotemporal anomaly vectors including the anomaly type, spatiotemporal coordinates, and deviation degree, and at the same time mark the potentially associated abnormal propagation paths;

[0160] Through the above content, based on the spatiotemporal rule knowledge base of historical operation and maintenance data, combined with multi-dimensional correlation analysis and spatiotemporal tensor decomposition technology, the accurate modeling and rule extraction of the urban infrastructure status evolution are realized. Through multi-dimensional comparison and dynamic threshold judgment, the abnormal states that deviate from the normal evolution path can be efficiently identified, and spatiotemporal anomaly vectors including the anomaly type, spatiotemporal coordinates, and deviation degree are generated, significantly improving the accuracy and interpretability of anomaly detection. At the same time, it supports the dynamic update and adaptive optimization of the knowledge base, enhancing the intelligent level and long-term applicability of the system.

[0161] The construction of the spatiotemporal rule knowledge base in S41 includes:

[0162] S411, Historical Data Preprocessing and Spatiotemporal Sequence Generation: Perform spatiotemporal alignment and normalization processing on the historical state data, divide the spatiotemporal units according to the Beidou grid code, and generate a spatiotemporal sequence data set for the infrastructure status evolution , where a single spatiotemporal sequence is expressed as:

[0163] ;

[0164] Among them, is the feature matrix of the th spatiotemporal unit at time , , is the total number of spatiotemporal units, is the number of time steps;

[0165] S412, Multi-dimensional Association Rule Mining: Use the improved Apriori algorithm to extract spatiotemporal association rules. The rule form is a triple , specifically including:

[0166] Support calculation: ;

[0167] Confidence calculation: ;

[0168] Among them, is the triggering condition, is the spatiotemporal constraint, is the evolution result;

[0169] S413, Spatiotemporal Tensor Decomposition and Reduced-Dimension Storage: Construct a four-dimensional spatiotemporal tensor , where the dimensions represent spatial grid, time window, state feature, and association rule respectively. Perform dimensionality reduction through Higher-Order Singular Value Decomposition (HOSVD), expressed as:

[0170] ;

[0171] Among them, is the rule tensor, representing the core rule structure after dimensionality reduction, , , , are factor matrices, corresponding to the spatial, temporal, feature, and rule dimensions respectively, is -mode product;

[0172] S414, Dynamic Update of Rule Knowledge Base: Design an incremental tensor decomposition algorithm. When new data is added, update the rule tensor, expressed as:

[0173] ;

[0174] Among them, is the updated rule tensor, is the transpose of the factor matrix of the th dimension;

[0175] Through the above content, a spatiotemporal rule knowledge base based on historical state data is constructed. Combining multi-dimensional association rule mining and spatiotemporal tensor decomposition techniques, accurate modeling and rule extraction of the evolution of urban infrastructure status are realized. Through the dynamic update mechanism, it can learn new data online and optimize rule weights, significantly improving the adaptability and timeliness of the knowledge base. At the same time, it supports interpretable rule triples (trigger conditions, spatiotemporal constraints, evolution results), providing high-quality knowledge support for anomaly detection and fault diagnosis, and enhancing the intelligent level and long-term applicability of the system.

[0176] The spatiotemporal anomaly detection and vector generation in S42 include:

[0177] S421, Spatiotemporal Similarity Calculation: Adopt the spatiotemporal weighted cosine similarity algorithm to calculate the similarity between the current feature matrix and the historical feature matrix in the knowledge base, expressed as:

[0178] ;

[0179] Among them, is the spatiotemporal weight, is the attenuation coefficient, is the feature vector of the th grid of the current feature matrix, is the feature vector of the th historical pattern in the knowledge base for the th grid, is the total number of grids;

[0180] S422, dynamic threshold judgment: Dynamically set the anomaly judgment threshold based on the statistical characteristics of the historical similarity distribution in the knowledge base , expressed as:

[0181] ;

[0182] Among them, is the mean of historical similarities, is the standard deviation of historical similarities, is the sensitivity adjustment factor (taking );

[0183] S423, abnormal vector generation: Generate a spatio-temporal abnormal vector for the abnormal state that satisfies , expressed as:

[0184] ;

[0185] Among them, is the abnormal type code, is the Beidou grid code of the abnormal location, is the abnormal timestamp, is the deviation degree (defined as );

[0186] S424, abnormal propagation path marking: Based on the spatio-temporal graph structure and the causal rules in the knowledge base, use the random walk algorithm to predict the abnormal propagation path , expressed as:

[0187] ;

[0188] Among them, is the causal rule in the knowledge base, is to select the propagation path starting from the abnormal point according to the edge weight and causal rule probability;

[0189] Causal rules in the knowledge base Are set based on historical state data, specifically including:

[0190] Preprocessing of historical state data: Perform spatio-temporal alignment based on historical state data to ensure the spatio-temporal consistency of the data, and remove the noise in the data through normalization processing;​

[0191] Multi-dimensional association rule mining: Mining the causal relationships between the infrastructure states of different cities from historical data and constructing causal rules , expressed as:

[0192] ;

[0193] Among them, is the triggering condition, is the evolution result;

[0194] ;

[0195] ;

[0196] Among them, represents the number of simultaneous occurrences of condition and result , is the size of the dataset, is the number of occurrences of condition ;

[0197] Space-time tensor decomposition and dimensionality reduction storage: Storing the causal rule in an efficient space-time rule knowledge base for subsequent retrieval and application, and reducing the dimensionality of high-dimensional data through space-time tensor decomposition to extract the most important causal rules;

[0198] Incremental rule update: When new data is added, use the incremental tensor decomposition method to update the causal rules in the knowledge base ;

[0199] Through the above content, it is possible to accurately identify abnormal states deviating from the normal evolution path, generate space-time abnormal vectors containing abnormal types, space-time coordinates, and deviation degrees, significantly improving the accuracy and interpretability of anomaly detection. At the same time, combining the causal rule set to mark potential associated abnormal propagation paths realizes full-process coverage from anomaly detection to risk warning, providing efficient and reliable technical support for the intelligent operation and maintenance of urban infrastructure.

[0200] The space-time memory-enhanced prediction and warning interval generation in S5 includes:

[0201] S51, Space-time memory pool construction: Construct a space-time memory pool based on the historical infrastructure state feature matrix , expressed as:

[0202] ;

[0203] Among them, is the space-time memory pool The historical state feature matrix in ;

[0204] S52, Gated spatio-temporal attention mechanism: Calculate the attention weights of the current feature matrix and the historical feature matrix in the memory pool , expressed as:

[0205] ;

[0206] Among them, is the th historical feature matrix in the spatio-temporal memory pool, is the attention weight matrix, is the attention vector, is the spatial decay coefficient, is to calculate the current Beidou grid code and the th historical feature matrix of the Beidou grid code the spatial distance between them, is the total number of historical state matrices stored in the spatio-temporal memory pool, is the th historical feature matrix in the spatio-temporal memory pool, is to calculate the current Beidou grid code and the th historical feature matrix of the Beidou grid code the spatial distance between them;

[0207] S53, Prediction trajectory generation: Generate the prediction trajectory of the future time window , expressed as:

[0208] ;

[0209] Among them, is the predicted infrastructure state feature matrix, indicating the infrastructure state in the future time window , is the length of the prediction time window, that is, how many future time steps are predicted, is the time step;

[0210] S54, Spatio-temporal warning interval calculation: Generate the spatio-temporal warning interval based on the confidence of the prediction trajectory , expressed as:

[0211] ;

[0212] Among them, is the confidence level coefficient, is the standard deviation of the prediction error, is for calculating the historical state of the Beidou grid code and the reference grid code the spatial distance between them, is the Beidou grid code for calculating the current infrastructure state and the reference grid code the spatial distance between them;

[0213] Through the above, combining the historical infrastructure state feature matrix and the current state data, using the gated spatio-temporal attention mechanism to weight the historical patterns, so as to generate accurate prediction trajectories and reliable warning intervals, which can make full use of the historical data in the spatio-temporal memory pool, improve the adaptability of the model to complex spatio-temporal changes, support dynamic adjustment and real-time prediction. Through precise spatio-temporal correlation calculation, it can not only predict the future state of the infrastructure, but also generate spatio-temporal warning intervals according to the prediction confidence, identify potential anomalies or faults in advance, and ensure the intelligence and efficiency of urban infrastructure management. In addition, the model dynamic update mechanism enables the system to self-optimize as new data is continuously added, further improving the prediction accuracy and long-term applicability.

[0214] The anomaly and warning superposition analysis and causal reasoning optimization in S6 include:

[0215] S61, anomaly and warning superposition analysis: Perform spatio-temporal superposition analysis on the spatio-temporal anomaly vector and the spatio-temporal warning interval to calculate the overlap coefficient of the anomaly within the warning interval , expressed as:

[0216] ;

[0217] Among them, is the anomaly deviation, representing the amplitude or deviation degree of the anomaly, is the Beidou grid code where the anomaly event occurs, representing the spatial position of the anomaly event, is the Beidou grid code of the warning interval, representing the spatial position of the warning interval;

[0218] S62, dynamic adjustment of the causal reasoning engine search space: Dynamically adjust the search space of the causal reasoning engine according to the overlap coefficient , specifically including:

[0219] Spatial constraint: Limit the search range to the spatio-temporal region centered on the Beidou grid code where the anomaly event occurs and with a radius , where (take 10 times the minimum unit of the Beidou grid code);

[0220] Rule screening: Only retain the subset of causal rules related to the current exception type Type;

[0221] S63, Fault diagnosis report generation: Based on the output of the causal reasoning engine, generate a fault diagnosis report with spatio-temporal coordinates, including the Beidou grid code where the root cause event occurred (the spatial location where the fault or exception occurred), the timestamp when the root cause event occurred (the specific time when the fault occurred), and the diagnostic confidence (the reliability of the fault diagnosis);

[0222] Diagnostic confidence Expressed as:

[0223] ;

[0224] Wherein, is the weight of the causal rule;

[0225] S64, Maintenance decision instruction generation: Generate maintenance decision instructions with Beidou grid code positioning according to the fault diagnosis report, including priority sorting (sort the root cause events in descending order of confidence) and path planning (generate the optimal maintenance path based on the spatial distribution of the Beidou grid code);

[0226] Through the above content, the accuracy and real-time performance of fault diagnosis are effectively improved. By calculating the overlap coefficient between the exception and the warning interval, it can be accurately evaluated whether the exception event is within the warning range, and the influence of the exception is weighted according to the spatial distance attenuation characteristic, thereby reducing the risks of false alarms and missed alarms. At the same time, through rule screening based on the exception type and dynamic adjustment of the search space in causal reasoning optimization, it is ensured that the causal reasoning engine can focus on the most relevant regions and rules, thereby improving the efficiency of fault location and cause analysis, enhancing the system's adaptability to complex scenarios, improving the accuracy and response speed of infrastructure fault warning and maintenance decision-making, and realizing more intelligent and efficient urban infrastructure management.

[0227] This invention covers any substitutions, modifications, equivalent methods, and solutions made within the essence and scope of this invention. To enable the public to have a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments of this invention. However, those skilled in the art can fully understand this invention even without the description of these details. Additionally, to avoid unnecessary confusion to the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0228] The above are only the preferred embodiments of this invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of this invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this invention.

Claims

1. A digital twin management method for urban infrastructure supported by BeiDou space-time base, characterized in that: The following steps are involved: S1, spatiotemporal reference grid construction and code allocation: Based on the satellite-based ground-based augmented data of the BeiDou-3 system, a centimeter-level spatiotemporal reference grid covering the target area is generated, and a BeiDou grid code with spatiotemporal uniqueness is allocated to each urban infrastructure entity; S2, multimodal data collection and spatiotemporal alignment processing: real-time status data is collected through multimodal sensors deployed on urban infrastructure, and the real-time status data is spatiotemporally aligned with the Beidou grid code to generate a perception data stream with spatiotemporal labels; S3, feature extraction: construct a digital twin evolution model based on spatiotemporal graph neural network, take the spatiotemporal labeled perception data stream input as input data, and generate the infrastructure status feature matrix with spatiotemporal correlation characteristics; S4, establishment of spatiotemporal rule knowledge base and anomaly detection: establish a spatiotemporal rule knowledge base for the evolution of urban infrastructure status, detect abnormal status of urban infrastructure and generate spatiotemporal anomaly vectors by comparing the infrastructure status feature matrix with the historical evolution pattern in the spatiotemporal rule knowledge base; S5, spatiotemporal memory enhanced prediction and warning interval generation: Construct a spatiotemporal memory enhanced prediction model, based on the current infrastructure status feature matrix and the spatiotemporal distribution characteristics of the Beidou grid code, predict the evolution trajectory of the infrastructure status within the future set time window, and generate the spatiotemporal warning interval of the prediction deviation threshold; S6, anomaly and warning superposition analysis and causal reasoning optimization: superimpose the spatiotemporal anomaly vector and the spatiotemporal warning interval for analysis, dynamically adjust the search space of the causal reasoning engine, output a fault diagnosis report with spatiotemporal coordinates, and generate maintenance decision instructions with Beidou grid code positioning based on the fault diagnosis report.

2. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 1 is characterized in that: The construction and coding allocation of the spatiotemporal reference grid in S1 include: S11, satellite-based and ground-based augmentation data acquisition: obtain the original observation data of the target area through the satellite-based augmentation service of the BeiDou-3 system and the ground-based augmentation station network; S12, Spatiotemporal reference grid division and hierarchical design: Design a multi-level spatiotemporal grid division scheme based on the spatial scope and application requirements of the target area, including: Spatial level: Divide the target area into regular grid cells with adjustable side lengths, and the minimum cell size is 0.5m×0.5m; Time level: Time slices are defined based on Beidou timing signals, with a minimum time resolution of 1 second; S13, grid code generation and space-time binding: Beidou grid coding algorithm is used to generate a four-dimensional code including longitude, latitude, elevation and time for each grid unit. The four-dimensional code structure meets the following conditions: ； in, , , , are the longitude, latitude, elevation and time of the grid center point, , , , is the coordinate of the origin of the space-time reference, , , , is the grid spatial and temporal resolution, is the Beidou standard hash function, is the Beidou grid code; S14, Infrastructure entity coding mapping: The spatial coordinates of urban infrastructure entities are mapped With timing status Mapped to the corresponding grid code through a dynamic interpolation algorithm.

3. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 1 is characterized in that: The multimodal data acquisition and spatiotemporal alignment processing in S2 includes: S21, Multimodal sensor deployment and data collection: Deploy multimodal sensor networks on urban infrastructure, including structural health monitoring sensors, environmental sensors, and operating status sensors, to collect multimodal data in real time, including structural health data, environmental data, and operating status data, among which; The structural health data includes strain, vibration, and displacement; The environmental data include temperature, humidity, and wind speed; The operating status data includes load, flow, and pressure; S22, data preprocessing: preprocessing the collected multimodal data, including outlier removal, missing value filling, and noise filtering; S23, time-space benchmark unification and data alignment: the pre-processed multi-modal data is time-space aligned with the Beidou grid code, specifically including: Spatial alignment: Map the data in the sensor's local coordinate system to the Beidou grid coordinate system through the coordinate transformation matrix; Time alignment: Synchronize and calibrate sensor timestamps based on Beidou timing signals; S24, spatiotemporal label generation and data stream encapsulation: Add spatiotemporal labels to the aligned multimodal data to generate standardized perception data streams.

4. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 3 is characterized in that: The data preprocessing in S22 includes: S221, outlier removal: based on The criterion removes abnormal data from multimodal data; S222, missing value filling: using spatiotemporal interpolation to fill missing data in multimodal data; S223, Noise filtering: Remove high-frequency noise in multimodal data through wavelet transform.

5. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 1 is characterized in that: The feature extraction in S3 includes: S31, construction of space-time graph structure: using Beidou grid codes as nodes and space-time associations as edges to construct a dynamic space-time graph structure , where the node feature is expressed as , For the The perception data of a grid, is the embedding vector of Beidou grid code, and the edge weight is expressed as , is the spatial attenuation factor; S32, spatial graph convolution calculation: using gated graph convolution layer to extract spatial features; S33, time-gated recurrent unit: models the timing characteristics of each node; S34, Spatiotemporal Attention Mechanism: Dynamically Calculating Node Importance Weights ; S35, feature matrix generation: Sort the output of the spatiotemporal graph neural network by Beidou grid code to generate the infrastructure status feature matrix ; S36, model training: end-to-end training is performed using a hybrid loss function.

6. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 5 is characterized in that: The establishment of the spatiotemporal rule knowledge base and anomaly detection in S4 include: S41, construction of spatiotemporal rule knowledge base: extract the spatiotemporal pattern of the evolution of urban infrastructure status based on historical status data, use multidimensional correlation analysis technology to abstract the spatiotemporal relationship of infrastructure status changes into rule triples, and use spatiotemporal tensor decomposition technology to reduce the dimensionality of high-dimensional relationships and store them, forming a spatiotemporal rule knowledge base that includes normal state evolution paths and failure modes; S42, spatiotemporal anomaly detection and vector generation: The generated infrastructure status feature matrix is ​​compared with the historical evolution pattern in the spatiotemporal rule knowledge base in multiple dimensions. Through spatiotemporal similarity calculation and dynamic threshold judgment, the abnormal state that deviates from the normal evolution path is identified. For the detected abnormal state, a spatiotemporal anomaly vector including the anomaly type, spatiotemporal coordinates and degree of deviation is generated, and the potentially associated abnormal propagation path is marked.

7. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 6 is characterized in that: The construction of the spatiotemporal rule knowledge base in S41 includes: S411, Historical data preprocessing and spatiotemporal sequence generation: Perform spatiotemporal alignment and normalization on historical status data, divide spatiotemporal units according to Beidou grid codes, and generate a spatiotemporal sequence dataset of infrastructure status evolution , where a single space-time sequence is represented as: ； in, For the Space-time units in time The characteristic matrix of , is the total number of space-time units, is the number of time steps; S412, Multidimensional Association Rule Mining: Use the improved Apriori algorithm to extract spatiotemporal association rules in the form of triples ,in, As the trigger condition, For space-time constraints, As a result of evolution; S413, Spatiotemporal Tensor Decomposition and Dimensionality Reduction Storage: Constructing a Four-Dimensional Spatiotemporal Tensor , the dimensions represent the spatial grid, time window, state features, and association rules, respectively, and the dimensionality reduction is performed through high-order singular value decomposition; S414, Dynamic update of rule knowledge base: Design an incremental tensor decomposition algorithm, when new data is added , update the rule tensor.

8. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 7 is characterized in that: The spatiotemporal anomaly detection and vector generation in S42 include: S421, spatiotemporal similarity calculation: Use spatiotemporal weighted cosine similarity algorithm to calculate the current feature matrix and the historical feature matrix in the knowledge base Similarity ; S422, dynamic threshold judgment: based on the statistical characteristics of the historical similarity distribution in the knowledge base, dynamically set the abnormality judgment threshold ; S423, exception vector generation: for satisfying The abnormal state generates a spatiotemporal abnormal vector ; S424, Anomaly Propagation Path Labeling: Based on Spatiotemporal Graph Structure and the causal rules in the knowledge base, and uses the random walk algorithm to predict the abnormal propagation path .

9. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 8 is characterized in that: The generation of the spatiotemporal memory enhanced prediction and warning interval in S5 includes: S51, Spatiotemporal Memory Pool Construction: Constructing a spatiotemporal memory pool based on the historical infrastructure status feature matrix ; S52, Gated spatiotemporal attention mechanism: Calculate the current feature matrix Attention weights with historical feature matrix in memory pool ; S53, prediction trajectory generation: generating future time windows through gated recurrent units The predicted trajectory ; S54, spatiotemporal warning interval calculation: Generate spatiotemporal warning interval based on the confidence of the predicted trajectory .

10. The urban infrastructure digital twin management method supported by the Beidou space-time base according to claim 9 is characterized in that: The anomaly and warning superposition analysis and causal reasoning optimization in S6 include: S61, anomaly and warning superposition analysis: the spatiotemporal anomaly vector and the spatiotemporal warning interval Conduct spatiotemporal superposition analysis and calculate the overlap coefficient of anomalies within the warning interval ; S62, Dynamic adjustment of the search space of the causal inference engine: based on the overlap coefficient Dynamically adjust the search space of the causal inference engine, including: Spatial constraint: limit the search range to the Beidou grid code where the abnormal event occurred Center, radius The space-time region, where ; Rule filtering: only retain the subset of causal rules related to the current exception type Type; S63, fault diagnosis report generation: based on the output of the causal reasoning engine, a fault diagnosis report with time and space coordinates is generated, including the Beidou grid code of the root cause event, the timestamp of the root cause event, and the diagnosis confidence; S64, maintenance decision instruction generation: generating maintenance decision instructions with Beidou grid code positioning according to the fault diagnosis report, including priority sorting and path planning.

Images