249 results about "Spatial graph" patented technology

The invention relates to a large-scale road network traffic control method based on a deep reinforcement learning large model, and belongs to the technical field of intelligent traffic control. The method comprises the following steps: sensing real-time multi-modal road network information including urban road intersections, highway entrance ramps and emergency lanes, and generating a space-time fusion representation vector representing a current traffic network state by fusing a space diagram construction method and a time sequence embedding method; the space-time fusion representation vector and historical state memory are spliced to serve as input, a backbone network of a pre-training large language model is used for state feature distillation so as to enhance state representation, and a traffic control decision is output through a strategy network with a layered action space; through cross-modal knowledge migration and a progressive course learning strategy, a training process of a deep reinforcement learning algorithm is guided and optimized so as to improve model training efficiency and generalization ability. According to the method, the generalization performance and the accuracy of the control strategy are improved while the real-time response speed is ensured.

The invention provides a pedestrian trajectory prediction method based on window attention and a space diagram interaction network, and belongs to the technical field of computer vision. The technical problems of difficult long-time dependence modeling and complex space interaction are solved. According to the technical scheme, the method comprises the following steps: S1, acquiring data of a data set; s2, in the time dimension, designing a window mask mechanism, and adjusting the attention receptive field at each moment; s3, constructing a hierarchical heterogeneous graph convolutional network according to a spatial dimension, and combining a pedestrian dynamic interaction graph with a scene static semantic graph; and S4, inputting the time dimension features and the space dimension features into a multi-scale expansion convolutional network to generate a multi-modal trajectory. The method has the beneficial effects that the model is subjected to experimental verification on a public data set ETH / UCY, the experimental result proves the effectiveness of the model, the superior performance of the model is shown on key indexes, and the generalization ability in processing different scenes is also excellent.

The invention discloses a time-space diagram neural network water quality prediction method based on DEM composition, and the method comprises the steps: obtaining and processing DEM data of a target region, extracting a river topological structure in the target region, and constructing a graph structure reflecting the spatial relation of water quality monitoring stations; acquiring water quality data, preprocessing the data, and eliminating the influence of missing values and abnormal values; a water quality prediction model is constructed based on an STGCN model, the STGCN model comprises a plurality of space-time convolution blocks, and each space-time convolution block comprises a time gating convolution layer, a space graph convolution layer and a time gating convolution layer which are arranged in sequence; and detecting the water quality to be detected based on the constructed water quality prediction model. According to the method, the accuracy of water quality prediction is improved by capturing the spatial dependency relationship between the water quality monitoring stations. A river network topological structure is extracted by using DEM data, a graph structure capable of truly reflecting hydrological relation between monitoring stations is constructed, and the accuracy of water quality prediction is effectively improved by combining effective capture of the STGCN model on spatio-temporal characteristics.

The invention discloses a smart power grid state estimation method and system based on a space-time diagram convolutional network, and a medium, and the method comprises the steps: building a topological graph model of a power system according to the topological structure and measurement data of a smart power grid; an STGCN model containing a plurality of space-time convolution blocks is constructed, each space-time convolution block captures time dependence of time sequence data through a time gating convolution layer, and space features of the smart power grid are extracted through a space graph convolution layer; and establishing a power grid state estimation model based on the STGCN model to perform power system state estimation. According to the method, the accuracy and the reliability of an estimation result are ensured by fusing a space and time convolution technology. State estimation challenges under various topological structures can be effectively handled, stable estimation performance can be maintained, and a powerful guarantee is provided for safe operation of a smart power grid. And the operation efficiency and the response speed of the smart power grid are greatly improved.

The invention discloses a photovoltaic power generation capability prediction method and system based on a space-time diagram neural network, and a medium. The method comprises the following steps: preprocessing historical data of a power system; a plurality of power stations are divided into a plurality of clusters, a power system is abstracted into a topological graph, and each cluster is regarded as a node of the topological graph; learning time features of the topological graph by adopting a time self-attention network model; learning spatial features of the topological graph by adopting a spatial graph convolutional network model; stacking the time self-attention network and the space graph convolutional network by adopting a factor type structure; and performing prediction by using a full connection layer. According to the method, the photovoltaic power generation system is modeled into the graph model, so that the spatial relationship and the time dependence among different power stations can be effectively captured, more efficient data processing is realized, redundant information is reduced, data streams can be optimized, and the requirements on storage resources are reduced by reducing direct operation on original data, so that the system is more efficient and more reliable. And the processing speed of the hardware is improved.

The invention relates to the technical field of slope catastrophe intelligent early warning and data modeling, and particularly discloses a strip mine high and steep slope multi-source monitoring and early warning method based on a graph neural network and Transform, and the method comprises the following steps: S01, carrying out the data preprocessing and disturbance variable construction of monitoring data; s02, constructing a heterogeneous space diagram structure by taking the monitoring points as nodes and taking geography, lithology and dynamic response relationships as edges; s03, constructing a space-time end-to-end multilayer coding framework based on the graph attention network and the integrated deep neural structure; s04, on the basis of graph coding and time sequence output, introducing a disturbance variable embedding mechanism, and designing a joint attention fusion structure; and S05, generating a deformation trend prediction value of the slope in a future period of time and performing corresponding risk grade judgment. The invention aims to solve the key technical problem of weak adaptability and interpretability of an early warning system.

The invention discloses a knowledge graph automatic construction and updating method based on a graph neural network, and the method mainly comprises the steps: carrying out the preprocessing of multi-source heterogeneous data, and mapping the text, image and sensor data to a unified feature space; performing graph neural network driven entity recognition and relation extraction, and outputting a structured triple; incremental fusion and graph embedding representation updating are carried out, and efficient connection of new and old knowledge is realized; self-adaptive plastic weight adjustment is carried out, and node connection is dynamically strengthened or attenuated based on an improved Hebb rule; and event-driven closed-loop feedback control is combined with a PID strategy to maintain the stability and accuracy of the system in a high-frequency updating scene. According to the method, multi-modal fusion, autonomous evolution and real-time updating are considered, the intelligent level and application expandability of the knowledge graph are greatly improved, and the method is suitable for the fields of medical treatment, finance, industry and the like.

The present invention is directed to a Spatiotemporal scene-graph embedding methodology that models scene-graphs and resolves safety-focused tasks for autonomous vehicles. The present invention features a computing system comprising instructions for accepting the one or more images, extracting one or more objects from each image, computing an inverse-perspective mapping transformation of the image to generate a bird's-eye view (BEV) representation of each image, calculating relations between each object for each image, and generating a scene-graph for each image based on the aforementioned calculations. The system may further comprise instructions for calculating a confidence value for whether or not a collision will occur through the generation of a spatio-temporal graph embedding based on a spatial graph embedding and a temporal model.

The invention discloses a physical information enhanced space-time diagram network trusted computing method for traffic flow anomaly prediction, and the method comprises the steps: building an LWR model physical constraint through an embedded traffic flow LWR equation, correcting critical density and free flow speed parameters through a Kruithof curve, and improving the flow density of an LWR model; carrying out road discretization and establishing a macroscopic CTM model; carrying out dynamic weight fusion on a prediction result of the macroscopic CTM model and a correction factor generated by the microscopic IDM model, and calculating a fusion result; constructing a space-time diagram convolution network model and a loss function, constructing a space diagram convolution module and a time sequence convolution module, and extracting a space feature matrix and a time sequence feature matrix; fusing and updating the spatio-temporal characteristics by using a gating circulation unit; performing dynamic anomaly detection and hierarchical response; a TEE is deployed in a traffic network sensing device, and evidence storage is carried out through a block chain. According to the invention, the traffic flow change is predicted in real time, the abnormal fluctuation in the traffic flow is identified, and the traffic flow prediction accuracy is improved.

The invention discloses a multi-index dynamic fusion traffic hidden danger identification and completion method and system, and relates to the technical field of intelligent traffic systems, and the method comprises the steps: obtaining a multi-dimensional traffic data sequence containing missing values in a target monitoring region, the multi-dimensional traffic data sequence comprises a traffic flow index sequence, a driving behavior characteristic sequence and a road environment parameter sequence; and interpolating missing values in the multi-dimensional traffic data sequence based on a spatio-temporal dynamic graph completion network to generate a completed spatio-temporal data tensor. According to the method, the missing data are accurately completed through a space-time dynamic graph completion network, a space graph convolution layer of road network topology, a gating circulation unit of a modeling time sequence rule and a cross-index attention module, multi-index dynamic association is fully fused, and propagation delay and a space-time evolution rule are considered; and deep fusion features are extracted by using a shared feature encoder, hidden danger identification branches are linked, traffic flow, driving behaviors and environmental parameters are dynamically fused, and hidden danger dynamic formation paths are captured.

The invention relates to the technical field of digital education, in particular to a digital teaching material intelligent generation system based on digital education, which comprises an education resource acquisition module, a knowledge graph construction module, a teaching material content generation module, a knowledge tracking module and a teaching material presentation module, according to the method, a graph embedding method based on a hyperbolic space graph convolutional network and a hierarchical clustering algorithm with improved robustness are introduced, low-dimensional embedding and multilayer structure optimization are performed on an initial knowledge graph, and an educational knowledge graph with a semantic hierarchical structure is constructed; a graph structured learning behavior modeling method is introduced, a student-knowledge point heterogeneous graph structure is constructed, a heterogeneous graph convolutional network is adopted to model interaction behaviors of students and knowledge points, a knowledge state graph of each student is obtained, and the accuracy of individual knowledge mastery degree modeling is improved.

The invention provides a human skeleton behavior recognition method based on an improved space-time diagram convolutional network (ST-GCN). According to a traditional ST-GCN method, structural relations of joint points are modeled through space graph convolution, dynamic changes are captured in combination with time convolution, however, the problems that a topological structure is fixed, time sequence modeling is limited and channels have no difference weighting exist, and the recognition precision and adaptability of the traditional ST-GCN method are limited. According to the method, a joint channel attention module (JCA) and a multi-scale path attention residual network (MSPARN) are introduced on the basis of an ST-GCN architecture, so that the adaptive weighting capability of a feature channel and the multi-scale modeling capability of a time sequence dynamic feature are respectively improved. Specifically, the JCA module strengthens the response of the key channel by dynamically adjusting the channel weight; the MSPARN module utilizes multi-scale cavity convolution and a path attention mechanism to enhance the processing ability of the model to complex time sequence dynamics. Experiments show that the method has strong robustness in a complex background and a multi-person scene, and has wide application prospects, especially in the fields of intelligent security and protection, health monitoring, human-computer interaction and the like.

The invention provides a traffic flow prediction method and system for a dynamic association information fusion space-time network, and the method employs a time attention mechanism to adaptively adjust the association strength between different time points, and distributes a relative weight for each time point, so as to achieve the precise modeling of a time sequence dependence relation. And the space-time dynamic module captures directional information and position sensing features of the traffic flow by introducing a space-time feature enhancement mechanism so as to improve the expression ability of the traffic flow data and the extraction effect of the space-time features. Meanwhile, a gating attention unit is provided, the deep dynamic trend in the traffic sequence data is effectively mined, and a potential time sequence change mode is captured. By constructing a dynamic feature matrix, spatial correlation and node features are fused in real time, so that richer and more accurate spatial context information is provided. Associated information extracted by a dynamic feature matrix is introduced into a spatial graph convolutional network to deeply learn dynamic spatial topological structure features in a traffic network.

The invention discloses a dementia identification method based on brain computer network space cross attention fusion. The dementia identification method comprises the steps that resting-state electroencephalogram signals are acquired, preprocessing and brain network construction are carried out, the frequency band power ratio is calculated, and a training set and a test set are divided; the brain network fuses the spatial information, and spatial features are obtained through an isotropic graph neural network and a local feature enhancement module; the frequency band power ratio is coded by a multi-layer perceptron to obtain frequency spectrum statistical characteristics; a bidirectional cross attention module is input, and complementary fusion features are extracted; and performing mixed pooling and splicing on the complementary fusion features, then sending the fused features to a Chebyshev Kolmogorov-Arnold network classifier, and outputting an Alzheimer's disease / frontotemporal dementia / health control (AD / FTD / HC) classification result and a cognitive scale evaluation score (MMSE). According to the method, the multi-feature complementary information is effectively integrated through joint modeling of the spatial diagram features and the global spectrum features, and the accuracy, stability and generalization ability of AD and FTD classification in a complex brain network are improved.

The invention relates to the field of bioinformatics and computational biology, in particular to a method for realizing accurate alignment of biological slices by utilizing a spatial graph network and a cell communication network based on the biological slices of a graph attention network. The method mainly comprises the following steps: (1) acquiring a high-resolution biological slice image and spatial transcriptome data of a target tissue or brain region (such as DLPFC), and carrying out standardization processing on the high-resolution biological slice image and spatial transcriptome data; (2) constructing a space diagram network based on cell space distribution, and constructing a cell communication network by using a ligand-receptor database; (3) carrying out feature fusion and deep learning training by adopting GAT, extracting high-dimensional features, and designing a comprehensive loss function fusing a mean square error and triple loss based on MNN at the same time; and (4) globally optimizing rotation, translation and X-axis overturning parameters through a differential evolution algorithm, realizing precise alignment of the slices, and evaluating an alignment effect by a consistency index.

The invention discloses a passenger car partition air conditioner self-adaptive adjustment method based on in-car multi-source signal perception, relates to the technical field of passenger car air conditioner adjustment, and aims to solve the problem of poor passenger car partition air conditioner adjustment efficiency. According to the method, multi-source environment data in a carriage is collected, unified scale processing is carried out in combination with nonlinear compression, the data is mapped to a three-dimensional space to construct a microenvironment topology tensor dynamically updated along with time, and environment states of different areas are accurately expressed; constructing a space graph structure based on state similarity and disturbance sensitivity weight, executing clustering division and correcting boundaries, and forming thermal environment partitions with adjustment values; a fuzzy adjustment participation multi-channel control structure is constructed for each partition, a control factor is dynamically generated according to an error signal and a disturbance trend, an adjustment result is output and mapped into an air conditioner equipment control parameter, partition cooperative adjustment of air supply, air exchange and purification is achieved, and the space response capacity and comfort matching level of an air conditioner control system are improved.

The invention discloses a traffic prediction method based on an adaptive semantic enhancement space-time diagram network, and the method comprises the steps: firstly constructing an initial space diagram through traffic sensor data, and carrying out the standardization processing; secondly, introducing an adaptive node embedding method to dynamically learn a spatial dependency relationship among nodes in the traffic network; a double-branch structure is further adopted, a parallel gating network (PGN) is used for capturing a long-term traffic mode, and linear hierarchical aggregation is adopted for capturing a short-term local mode; and finally, fusing space and time features to perform end-to-end training to optimize the prediction model. The method can effectively improve the accuracy and robustness of traffic prediction, and is suitable for intelligent management and decision optimization of a complex urban traffic environment.

The invention provides a multi-slice spatial group data alignment and data enhancement method based on deep learning, which comprises the following steps: collecting slices of different tissues in a plurality of development stages, and obtaining gene expression data of a single cell level of each slice and spatial position information of each cell in the tissues by utilizing a spatial transcriptomics technology; based on the gene expression data and the spatial position information of each slice, constructing a spatial diagram by combining a K-nearest neighbor method and a circular neighborhood construction method, and mapping the spatial diagram to a shared embedding space by using a DGCNN network to obtain embedding distribution; respectively sampling the embedding representation of each cell point from the embedding distributions of the two slices, and outputting the probability of a real sample by using a discriminator of the generative adversarial network; spatial alignment of the two slices is achieved by minimizing the difference between the probability distributions of the two slices. According to the method, accurate spatial alignment is realized by integrating spatial information of different slices, and data enhancement is carried out by aggregating information of adjacent slices into a target slice.

The invention provides an electrocardiosignal anomaly detection method fused with electrocardiosignal propagation simulation. The method comprises the steps that a space diagram structure is established based on ECG data; performing spatial position mapping on each electrode and establishing a heart simulation physiological model; feature extraction is carried out on each electrode node, and a dynamic space diagram structure is constructed; based on a dynamic graph structure, through multi-layer graph convolution updating and time aggregation, global space-time feature vectors are output, electrocardiosignal detection categories are output through a classifier, a space dynamic neural network model is constructed, a total loss function is established for model optimization, the electrocardiosignal detection categories are predicted, and quantitative recognition of abnormal propagation paths is achieved. And iteratively optimizing the dynamic graph structure. According to the method, unified modeling is carried out through time dynamics and spatial relevance of physiological signals, the defect that a static graph cannot capture abnormal changes of a propagation path in a pathological state is overcome, and the accuracy of cardiovascular disease diagnosis and the effectiveness of clinical decision support are remarkably improved.

The invention relates to the technical field of micro-grids, in particular to a micro-grid source load prediction matching method based on a graph neural network and an attention mechanism. The method comprises the following steps: acquiring multi-dimensional data of a micro-grid, and constructing a heterogeneous graph based on the multi-dimensional data; based on the heterogeneous graph, space graph convolution features and time sequence features are extracted, and unified space-time features are obtained through space-time graph attention network fusion; constructing a hierarchical attention network, and outputting the enhanced spatial-temporal characteristics of each node; respectively predicting an available power output sequence, a power demand sequence and a cost matrix through three decoders; calculating an optimal matching matrix with the minimum total transmission cost; and converting the optimal matching matrix into an equipment control instruction, issuing the equipment control instruction to an equipment controller, and taking an execution result and the newest state of the power grid as new multi-dimensional data. According to the method, high-precision matching prediction is realized, and a real-time and optimal control strategy is generated, so that the operation economy, safety and new energy consumption capability of the micro-grid are remarkably improved.

The invention discloses a traffic state prediction method, device and equipment for a road side edge platform, and a medium, and relates to the technical field of computers, and the method comprises the steps: carrying out the data preprocessing and aggregation of traffic state observation data, extracting a space rule by using a first time sequence gating convolution layer and a space graph convolution layer in a target space-time convolution block of the initial traffic state prediction model, and extracting time sequence dynamic characteristics based on a second time sequence gating convolution layer; and performing feature enhancement by using a three-dimensional attention mechanism, determining an obtained output result as a new input feature, determining a new target space-time convolution block, and skipping to a step of extracting a time rule by using a first time sequence gating convolution layer in the target space-time convolution block of the initial traffic state prediction model based on the input feature, and obtaining a target feature output by the target traffic state prediction model, and determining a future traffic state prediction result based on the target feature. And the accuracy of traffic state prediction is improved.

The invention discloses a pedestrian trajectory prediction method based on space-time interaction perception, and aims to solve the problem of complex space-time interaction behaviors of pedestrians and improve trajectory prediction precision. The method comprises the following steps: firstly, processing a pedestrian trajectory data set, and constructing a time diagram and a space diagram as input; extracting space and time features among pedestrians through a self-attention mechanism, and inputting the space and time features into a space-time interaction perception module to learn a dynamic coupling relationship between motion and interaction to obtain interaction perception features; and then inputting the interactive features into a graph convolutional network to obtain trajectory representation features, finally predicting two-dimensional Gaussian distribution of future trajectories through a full-dimensional dynamic time sequence convolutional network, and outputting a high-precision trajectory prediction result. According to the invention, through a spatio-temporal joint modeling framework, spatio-temporal feature dynamic fusion of pedestrian motion is realized, and the accuracy and robustness of pedestrian trajectory prediction are improved.

The invention discloses a block file space topology aggregation and cross-modal indexing method based on HDF5, and relates to the technical field of electric digital data processing. The method comprises the following steps of point cloud space division, space data separation and storage, multi-modal data fusion and multi-level index construction. The method comprises the following steps: performing space division on a point cloud space through an octree algorithm to obtain octree nodes, analyzing the relevance between the space division and a memory in real time to determine whether to perform octree division dynamic adjustment, then encoding each octree node, performing space block data separation storage, and simultaneously performing point cloud data access optimization. According to the method, the multi-level index system of the point cloud space is established, then spatial attribute aggregation is performed to establish the spatial diagram, multi-modal data integration fusion is performed, and finally the multi-level index system of the point cloud space is established, so that the efficiency of accessing and retrieving the image block file space is improved, and the problem of low cross-modal retrieval efficiency of the point cloud space divided based on an octree algorithm in the prior art is solved.

The invention is suitable for the technical field of tourism prediction, and provides a tourism prediction method based on multi-source data fusion, and the method comprises the following steps: collecting historical tourist data, social media data, weather forecast data, hotel reservation data and ticket reservation data of each scenic spot / city; constructing a tourism prediction model to obtain preliminary prediction information; the method comprises the following steps: constructing a graph structure by taking each scenic spot / city as a node and taking an operation unit or a function role of the scenic spot / city as an edge; spreading the graph structure in the graph neural network, introducing space graph convolution, a gating mechanism and a dynamic attention mechanism in the spreading process to generate a nonlinear correction, and adjusting the preliminary prediction information by using the nonlinear correction to obtain tourism prediction information. According to the method, nonlinear correction is carried out through modeling space-time dependence, the implied complex space-time nonlinearity problem in multi-source data is solved, and the modeling capacity for emergencies and network effects is improved.

The invention discloses an arch bridge tensioning method and system based on a physical constraint space-time diagram attention network, and the method comprises the steps: constructing a dynamic topological graph of an arch bridge cantilever construction structure, taking a structure initial state parameter, a control force system parameter and a real-time environment parameter as node input features, and taking a structure state response parameter as an output label, generating a training data set based on finite element model simulation; designing a deep learning network model combined with space diagram attention and time sequence causal convolution, and modeling a space-time dependency relationship; defining a mixed loss function of the data fidelity item and the physical residual loss item of the structural mechanical control equation; and training the model through a back propagation algorithm and deploying the model to a field decision control system. According to the method, through graph structure abstraction, spatio-temporal joint modeling and physical constraint embedding, multi-source monitoring data driven structure state online prediction and tension control strategy intelligent optimization are achieved.

The invention discloses a cloud edge fusion photovoltaic power generation prediction method, system and device based on a graph attention network and probability autoregression, and a medium, and belongs to the technical field of photovoltaic power generation prediction, and the method comprises the steps: obtaining geographic information of each photovoltaic power station through a cloud server, generating an adjacent node list according to a preset adjacent rule, and issuing the adjacent node list to an edge node, the edge nodes construct a space graph structure based on neighborhood information and local historical multi-source data; constructing a joint prediction model containing a graph attention network and a probability long-short term memory network at the edge node; each edge node locally trains a joint prediction model and periodically uploads local model parameters to a cloud server, and the cloud server aggregates the edge model parameters based on a federated learning strategy to form and synchronously issue a global sharing model; and the edge node performs real-time power prediction based on the updated model, and outputs a probability distribution result of the generated power. According to the method, the prediction accuracy and efficiency are remarkably improved, and the real-time prediction capability is good.

The invention relates to a BIM digital building operation management technology, and discloses a space splitting and merging method and system based on BIM vector modeling data, and the method comprises the steps: obtaining vector data of a target room model, obtaining a preliminary disassembly range drawn by a user and a target area input by the user, and obtaining a target area; and performing dynamic iterative optimization on the initial disassembly range based on a difference value between an intersection area of the initial disassembly range and the vector data and the target area to obtain a final disassembly range, and performing disassembly operation on the vector data by using the final disassembly range to obtain a spatial graph set. And according to the division results of the overlapping regions and the non-overlapping regions between the spatial graphs in the spatial graph set, performing merging processing on the spatial graphs in the spatial graph set to obtain a merging result, and performing topological optimization on the merging result to obtain an optimized merging result. According to the method, the efficiency and precision of space splitting and merging in the BIM model can be improved.

The invention discloses a wheeled robot fault diagnosis method and system based on a dynamic graph convolutional network, and belongs to the technical field of fault diagnosis. The method comprises the following steps: acquiring multi-channel sensor data of the wheeled robot, performing correlation analysis on the sensor data of different channels, and generating a sensor dynamic association graph; combining the sensor dynamic association diagrams corresponding to different time nodes along the time dimension, and constructing a dynamic space-time diagram; processing the dynamic space-time diagram through a trained fault diagnosis network to generate a fault diagnosis result; the fault diagnosis network alternately executes time sequence convolution and space diagram converter convolution on an input dynamic space-time diagram through a space-time diagram converter module, and integrates multi-modal time sequence information into feature representation of diagram nodes. The understanding and representation capability of the model on multi-sensor spatio-temporal data can be improved, and then the accuracy and robustness of fault diagnosis model prediction are improved; the problem that the robustness of fault diagnosis of an existing wheeled robot is low is solved.

The invention relates to a modeling method for accurate traffic flow prediction, which comprises the following steps of S1, integrally modeling all intersections and roads into a road network undirected graph, and defining an adjacent matrix and a distance matrix, S2, defining propagation time delay of traffic flow at the intersections, and S3, calculating the traffic flow of the intersections according to the propagation time delay. S3, combining propagation time delay and traffic flow abrupt change influence of adjacent intersections to confirm a comprehensive effect of each intersection at the moment t, and obtaining traffic flow representation of the intersection at the moment t, S3, aggregating features of each intersection through an adjacent matrix and the propagation time delay by a space graph convolutional layer, and capturing a dynamic rule through a time graph convolutional layer, the method comprises the steps of (S1) obtaining a graph convolution layer, combining the graph convolution layer and a time convolution layer to form a space-time graph convolution network, and outputting final traffic flow prediction, (S2) giving the final traffic flow prediction and a kernel density estimation matrix of GKDE and setting three learnable matrixes, and (S5) outputting through a feedforward neural network. The method has the advantage of accurate traffic flow prediction.

The invention discloses a building structure health monitoring method and system based on multi-source data fusion, and the method comprises the steps: obtaining multi-source physical parameters related to a building structure, and carrying out the preprocessing; and carrying out online quality evaluation by using a conditional variation auto-encoder model trained based on health reference data, and calculating the dynamic reliability weight of each sensor according to an error. And constructing a sensor space diagram fusing physical proximity and signal correlation, performing deep feature learning by using a space-time diagram attention network, and outputting a global feature vector and a key node feature vector representing the overall health state of the structure. And constructing a reliability weighting-dynamic Bayesian network model, and outputting the structural health state confidence evolved along with time. And outputting a deterministic health state recognition result and a corresponding confidence coefficient by adopting a dual-threshold decision-making mechanism in combination with state transition verification and persistence analysis. The damage position, the damage degree and uncertainty quantification are output, and the robustness of building structure health monitoring is improved.