61 results about "Graph structured data" patented technology

The present application belongs to the field of desulfurization tower heat exchange performance monitoring and optimization control, and discloses a kind of power plant desulfurization tower heat exchanger heat exchange coefficient prediction method and system, the multivariate operating data of desulfurization tower heat exchanger system is collected by multiple sensors, and the graph structure data set containing the space relationship of measuring point is constructed after pretreatment;Design time decoupling module, decompose operating sequence into trend and periodic component, respectively extract long-term trend feature and periodic fluctuation feature and carry out fusion;Fusion operating condition mode, measuring point space attention and graph structure information, construct space-time feature matrix representing time correlation and multi-measuring point space correlation simultaneously;Through heat exchange coefficient prediction module, generate future multi-step heat exchange coefficient prediction sequence.The present application can effectively fuse time and space dependence and operating condition mode information in system operation, realize accurate, multi-step prediction of heat exchange coefficient, and provide reliable basis for desulfurization tower heat exchange performance evaluation and operation optimization control.

This disclosure provides a data processing method and related equipment based on deep learning, relating to the field of intelligent recommendation technology. It includes: acquiring graph structure data corresponding to a set of items, and determining a target item node and a set of neighboring item nodes based on multiple item nodes in the graph structure; identifying a subset of feature data from the feature data of the item set that is inconsistent with the neighboring feature data of the target item node; matching digital fingerprints to the feature data in the feature data subset and determining the similarity between the feature data and the digital fingerprints; filtering based on similarity, storing feature data that meets preset conditions in a uniform pool corresponding to the target item node, and aggregating the feature data in the uniform pool to generate updated target feature data for the target item node; and determining user preferences based on the updated target feature data of at least one target item node. This disclosure can reduce computational complexity and improve computational efficiency.

The application relates to an integrated circuit technology field, and discloses a delay prediction method and a computer readable storage medium. The method comprises the following steps: obtaining a to-be-calibrated circuit, and converting the to-be-calibrated circuit into graph structure data; calculating the to-be-calibrated circuit by using a timing analysis tool to determine a to-be-calibrated delay of the to-be-calibrated circuit; encoding the graph structure data corresponding to the to-be-calibrated circuit to determine a cell feature vector, a node feature matrix and an edge feature matrix; fusing the node feature matrix and the edge feature matrix by using a feature extraction model to obtain a fused node feature matrix; aggregating the fused node feature matrix to obtain a graph-level feature vector; integrating the cell feature vector and the graph-level feature vector to determine a context feature vector; splicing the context feature vector and the to-be-calibrated delay to determine a combined feature vector; performing residual prediction on the combined feature vector, and correcting the to-be-calibrated delay based on the predicted residual to obtain a calibrated delay.

The disclosure provides a method for training a graph neural network. The method includes: obtaining initial graph structure data corresponding to the terminal device, initial graph structure data respectively obtained by distributed training terminals being derived from the same sample graph structure data; and performing a graph structure data processing stage and graph neural network training stage cyclically, until a target neural network satisfying a training requirement is obtained: determining a processing opportunity for currently performing a graph structure data processing stage based on historical execution data of historically performing a graph structure data processing stage and a graph neural network training stage; performing, based on the processing opportunity, graph structure data processing on initial graph structure data in the graph structure data processing stage, to generate target graph structure data; and training, based on target graph structure data, the target neural network in the graph neural network training stage.

The present application belongs to the technical field of intelligent processing of electric shovels, and discloses an electric shovel intelligent excavation prompting system based on a hybrid encoder and a Transformer decoder, comprising: a data acquisition module, a sensor is arranged on the electric shovel to acquire real-time operating parameters of the electric shovel and real-time environmental information parameters of the working environment of the electric shovel; a hybrid encoder module, which dynamically constructs the relationship between nodes and edges through a graph neural network for the real-time operating parameters and the environmental information parameters, and converts the relationship into graph structure data; and removes noise information from the graph structure data through a cross-modal variational autoencoder to obtain encoded structure data; and a Transformer decoder module, which captures the correlation of the encoded structure data based on a self-attention mechanism, acquires working condition information of the electric shovel, and outputs excavation prompting information. The present application realizes intelligent processing of real-time operating parameters and environmental information parameters of the electric shovel, outputs excavation prompting information to assist the electric shovel driver in operation, and improves the efficiency and safety of electric shovel operation.

This invention discloses a method, device, equipment, and program product for early warning of hydrogen leakage in fuel cell vehicles. The method includes: acquiring hydrogen concentration, temperature, and gas flow rate at multiple monitoring nodes of the hydrogen supply pipeline of the fuel cell vehicle to obtain multimodal node sensing data; determining the graph structure data of the hydrogen supply pipeline based on the pipeline structure and the multimodal node sensing data; inputting the graph structure data into a pre-trained hydrogen leakage risk prediction model to obtain the hydrogen leakage risk value for each monitoring node; determining the hydrogen leakage risk level of the hydrogen supply pipeline based on the hydrogen leakage risk value; and issuing an early warning based on the hydrogen leakage risk level. This invention improves the accuracy and timeliness of hydrogen leakage early warning, enhances the safety performance of fuel cell vehicles, and can be applied to the field of vehicle monitoring technology.

The application discloses a digital core permeability prediction method and device, equipment and a storage medium, and relates to the oil and gas field, which comprises the following steps: acquiring voxel data of a target digital core, and performing feature extraction on the voxel data by using a target pore network extraction algorithm to obtain corresponding graph structure data; obtaining target hidden features corresponding to the graph structure data by using a target encoder network, updating the target hidden features by using a target graph neural network to obtain updated hidden features, and obtaining micro physical quantities corresponding to the voxel data at a pore scale by using a target decoder network based on the updated hidden features; determining a target pressure field corresponding to the target digital core based on the micro physical quantities, determining total flow data corresponding to the target digital core based on the target pressure field and the micro physical quantities, and determining a target permeability corresponding to the target digital core based on the total flow data. The application realizes deep coupling of digital core permeability micro physical field prediction and macroscopic physical property derivation.

This invention relates to an intelligent flow sensing method and system for cold-pressing amorphous alloy powder, belonging to the field of flow measurement technology. The method includes constructing an acoustic emission feature vector; inputting the original capacitance measurement tensor and acoustic emission feature vector into a capacitance tomography reconstruction algorithm to construct a multi-frequency dielectric-mechanical coupled Gaussian mixture model; performing soft classification decoupling on the dielectric constant distribution matrix to generate a shear band density field and a porosity distribution field; introducing the shear band density field to perform free volume correction on the powder's apparent density distribution field, and solving to obtain a real-time flow characteristic scalar; inputting this time-series graph structure data into a graph neural network-long short-term memory coupled model; comparing the shear band network connectivity with a threshold in real time; and constructing a multi-objective cost function to achieve coordinated control of high-precision flow tracking and active suppression of shear band instability during cold pressing. This invention improves the flow sensing accuracy, instability early warning timeliness, and process control coordination of cold-pressing amorphous alloy powder.

The application discloses a kind of based on joint simulation underwater acoustic channel modeling and signal evaluation method, it is related to channel modeling technical field, including: in the first physical state generation first working condition received energy, and in the second physical state generation second working condition received energy;According to first working condition received energy and second working condition received energy determine heuristic information parameter;Determine the starting node and the terminal node of the graph structure data in submarine pipeline, in each iteration of the plurality of ants from the starting node to the terminal node in graph structure data, according to heuristic information parameter updates the travel path of each ant, to obtain the breathing dominant transmission path set;The application is favorable to provide more reliable technical basis for the design and operation of underwater acoustic communication in submarine pipeline corridor.

The application provides an article classification method and device based on graph attention diffusion, the article classification method comprising: obtaining article information of an article to be classified, and constructing graph structure data of the article to be classified; determining an attention coefficient between two adjacent article nodes in the graph structure data based on a latent representation vector of each article node and using a graph attention mechanism; performing iterative attention diffusion on an attention matrix based on an adjacency matrix of the graph structure data and a jump back probability coefficient, to obtain a deep-level attention matrix; and performing feature aggregation on the latent representation vector based on the deep-level attention matrix, to obtain an attention aggregation feature of each article node, so as to determine a classification result corresponding to each article to be classified. Through the above method, the information from a high-order neighborhood is quantitatively aggregated, and an over-smoothing phenomenon is avoided, thereby improving the accuracy and stability of processing an article classification task.

Providing a new data generation system. [Solution] The system includes a first data processing device that has the function of converting a hardware description language into graph structure data and generating a netlist by performing calculations based on Monte Carlo tree search; a second data processing device that has the function of converting the graph structure data into matrix data based on a graph neural network; and a third data processing device that has a transformer model and generates a first command according to the input token sequence. The first data processing device has the function of generating random numbers, and the command to be executed in Monte Carlo tree search is selected based on the relationship between the random number and a specified value, either as a first command or as a second command selected based on a calculation formula for action selection.

This invention relates to the field of industrial production data management technology, specifically disclosing an automatic transfer and transposition method and system for electronic product assembly. It generates a continuous event flow by collecting and converting production events from each process on the production line. Based on this, it constructs and maintains a graph structure data, defining products, components, workstations, and label carriers as nodes, and their associations as relational edges. The core of this invention lies in its response to labeling requests. The system performs a traversal query based on the graph data. Before the physical labeling action is executed, it dynamically parses the target label carrier information and forcibly verifies the integrity of the product data chain. Only when the verification is successful is a control command generated to drive the actuator to complete the precise transfer, transposition, and labeling of the label carrier. After successful labeling, the system atomically updates the graph data, establishing a binding relationship between the product and the label carrier, achieving forward prediction and reverse full-link traceability. This invention realizes intelligent closed-loop production management from passive response to proactive prediction.

This application provides a software defect prediction method based on a dynamic path adaptive graph convolutional network. The method includes: constructing graph structure data of software modules; inputting the graph structure data into a graph convolutional network to learn the dependencies between modules; adaptively assigning weights to multi-hop paths through a dynamic path weighted convolutional layer and suppressing long path noise interference based on a path length decay coefficient; optimizing the adjacency matrix in real time based on a multi-head attention mechanism, enhancing key dependency edges and pruning redundant connections; embedding dynamic path scoring and an adaptive graph update algorithm into a classification objective function, updating training weights with a balancing strategy, and outputting defect prediction results through the optimized model. This application can effectively solve the performance degradation problems of traditional graph convolutional networks in scenarios with insufficient dependency modeling, noise accumulation, and class imbalance under the static graph assumption.

The application belongs to the technical field of data processing, and particularly relates to an intelligent prediction and evaluation system based on wire and cable insulation aging. The system comprises: a data acquisition preprocessing unit, which is used for mapping a line topology into graph structure data; a physical priori knowledge base management unit, which converts an aging mechanism into a logical constraint item; a double-flow feature extraction processor, which utilizes a graph neural network and a long short-term memory network to capture space-time features and combines the logical constraint to perform physical reduction; an intelligent prediction core execution mechanism, which performs transfer learning through meta-learning and eliminates environmental interference by using causal inference; and a multi-dimensional health degree fusion evaluation output unit, which adaptively adjusts feature weights and generates a health degree curve and a residual life distribution. The application improves prediction robustness and interpretability, and provides precise support for cable life cycle management.

This invention relates to the field of structural explosion dynamic response calculation, specifically disclosing a heterogeneous graph data representation method for structural explosion dynamic response analysis. The method includes: establishing and validating a high-fidelity finite element baseline model of the structural explosion response, and obtaining a baseline data source for the structural explosion response; establishing and implementing an adaptive sampling algorithm based on physical field gradients to adaptively select key nodes representing the structural dynamic characteristics from the finite element mesh nodes; establishing a multi-attribute graph edge linking and weight quantization method to construct a multi-attribute weighted graph; and establishing an automated construction and storage method for spatiotemporal graph datasets to generate a spatiotemporal sequence graph dataset for training a graph neural network. This invention overcomes the limitations of traditional neural networks on the serialization and meshing of training data, achieving automated, high-fidelity conversion from continuous, heterogeneous finite element simulation data to sparse, discrete graph structure data with well-defined topological relationships, as well as data size simplification and increased physical information density.

This disclosure provides a graph cue learning method, apparatus, electronic device, storage medium, and program product. The method includes: determining the graph structure data of a target network; extracting vectors from the graph structure data to obtain local cue vectors and global cue vectors; enhancing the local and global cue vectors to obtain local and global cue-enhanced node embeddings; training a model based on the local and global cue-enhanced node embeddings to obtain a dual-cue task inference model; and inputting the graph structure data into the dual-cue task inference model for task inference to obtain the task classification result output by the model. This disclosure can fully utilize graph structure information and improve downstream task performance with only a few parameter updates.

The application discloses a social user classification method and system based on a graph neural network, electronic equipment and a computer readable storage medium, and belongs to the technical field of social user classification. The method comprises the following steps: constructing original graph structure data based on input social user data, obtaining node representation, performing oversampling operation based on the node representation, generating a synthetic node for a minority node in the data, obtaining adjacency information of the synthetic node based on the synthetic node, assigning a pseudo label to the synthetic node, combining the synthetic node, the adjacency information and a real node to construct a node balanced graph, and performing classification. The method can solve the imbalance classification problem, improve the accuracy of social user classification, and solve the problem of low accuracy and high calculation cost caused by class imbalance in social user classification in the prior art.

The application relates to the technical field of management and decision support information system, and discloses a comprehensive plan management information system and method, which comprises the following steps: acquiring time sequence dependent graph structure data of each project and plan time period and resource demand identification of each bidding task, generating a cross-project resource association graph as a resource layer of a double-layer graph atlas; receiving resource scheduling change event data, calculating time offset of each affected task, and generating a resource layer affected task set; mapping an indirect risk value of a supplier layer to an associated bidding task, superimposing the time offset influence of the resource layer, and identifying a high-risk bidding task with a composite risk index exceeding a threshold value. The application solves the technical problems that resource conflict analysis and supplier risk analysis are independently operated, cross-system interaction is ignored, and composite risk cannot be timely warned.

The application discloses a kind of sonar image classification methods of combined SLIC superpixel and graph attention network.It includes the following steps: according to the imaging principle of two-dimensional forward-looking sonar and side scan sonar respectively and prior information when imaging: the sonar image after correction is based on the image pre-segmentation of improved DeepLabV3+ network, SLIC superpixel algorithm is used to construct Graph (graph) structure data: construct the sonar image classification model based on GAT (graph attention network), the sonar graph structure data built is sent into network to complete the training and test of model;Verify the importance of pixel feature and spatial position feature.The application discloses a kind of sonar image classification methods of combined SLIC superpixel and graph attention network, which fully utilizes the space position relationship of acoustic shadow area, target area and shadow area by SLIC superpixel method and graph attention network, so as to realize higher accuracy of sonar image classification identification of sonar image by combining pixel feature and spatial geometric feature.

The application discloses a kind of based on WFGN's thermal error prediction method and system, comprising: step S1: obtaining temperature sequence data, and part error data, data is converted into graph structure data;Step S2: data set is constructed using graph structure data, data set is divided into training set, verification set and test set;Step S3: construct multi-domain fusion graph neural network, including the convolution network based on fourier transform for capturing global feature, the convolution network based on wavelet transform for capturing local feature and fully connected network, connect by feature fusion layer;Step S4: the multi-domain fusion graph neural network is trained;Step S5: according to the temperature sequence data detected when part processing, the corresponding error result is predicted;Using deep learning technology, by fusing fourier transform convolution network, wavelet transform convolution network and fully connected network, learn to automatically predict the thermal error generated under different machine tool processing temperature, extract global frequency characteristics by fourier transform, wavelet transform captures local multi-scale change characteristics, and high-precision prediction result is integrated and output by fully connected network, effectively improve the precision and efficiency of thermal error prediction.

The application belongs to the technical field of machine learning, and is especially based on the integrated algorithm research of Bayes and LSTM and GNN, including a graph data construction module, a graph neural network module and a Bayes long short-term memory network module; the graph data construction module is used for receiving original data and constructing graph structure data, the graph structure data including nodes, edges and node features; the graph neural network module is connected with the graph data construction module and is used for processing the graph structure data to generate node embedding vectors containing topological relationship information; in the application, the graph data construction module, the GNN module and the Bayes LSTM module are designed in a serial integrated manner, and a complete and coherent data flow from constructing a graph structure from original data, to extracting spatial features by using GNN and finally to time series modeling by Bayes LSTM is defined.

The application discloses a kind of based on graph neural network's solid oxide fuel cell system multi-fault diagnosis model training method and fault diagnosis method, obtains the multi-source time series operation data of SOFC system under different operating conditions, obtains data set;Operating condition includes normal state, single fault state and compound fault state, and data set is labeled with operating condition label;Sample data in data set is preprocessed, node feature matrix and corresponding adjacency matrix are generated according to each sample data, and graph structure data set is formed;Graph neural network model including multi-head graph attention layer, residual graph convolution layer and global pooling classification layer in turn is constructed, then the graph structure data in graph structure data set is as input, corresponding operating condition is as output, graph neural network model is trained, i.e. the multi-fault diagnosis model trained is obtained, and can be used for the multi-fault decoupling diagnosis of SOFC system, improves diagnostic accuracy and practicality.

The application provides a data processing method and device based on manifold learning embedding, equipment and medium, including: obtaining graph structure data representing the relationship between high-dimensional data points, and storing edge data in memory; rearranging the storage order of the edge data in the memory according to the vertex identifier associated with it, so that the edge data associated with the same vertex is continuously stored in the memory address space; based on the rearranged edge data, divide all vertices into multiple subsets, and perform iterative optimization of embedding coordinates in parallel by multiple processing units to obtain low-dimensional embedding coordinates of each vertex; based on the low-dimensional embedding coordinates, perform downstream data processing tasks, including at least one of data visualization, feature extraction or anomaly detection. Using the above method, the efficiency of data processing based on manifold learning embedding can be greatly improved, the data processing overhead can be reduced, the running performance of the whole data processing process can be optimized, and the actual needs of efficient processing of large-scale high-dimensional data can be met.

The present application belongs to the technical field of intelligent analysis and measurement, and discloses a kind of track geometric parameter intelligent analysis and adjustment quantity calculation method.The method synchronously collects laser radar point cloud and binocular sequence image, and constructs track three-dimensional real scene model by feature matching, beam adjustment and multi-modal registration fusion;Continuous mileage is converted into graph structure data, and parameter error is corrected by combining continuity constraint and Euler beam dynamics regularization with graph neural network;Based on finite element digital twinning and reinforcement learning agent, the optimal adjustment quantity is quickly solved by combining Gaussian process proxy model and active learning with geometric compliance and stress safety as constraints.The present application eliminates non-physical jump and stress rebound, improves detection accuracy and adjustment rationality, and is suitable for intelligent detection and fine maintenance operation of railway track.

The application discloses a kind of based on AI technology's spare part intelligent optimization design system, comprising: AI intelligent demand processing module, AI intelligent modeling module, AI intelligent evaluation module, AI intelligent optimization module and AI intelligent mapping module;Among them, AI intelligent demand processing module is used to analyze user demand data, generates the initial three-dimensional geometric model of spare part.AI intelligent modeling module is used to model according to the graph structure data of initial three-dimensional geometric model, generates multiple similar candidate geometric model scheme.AI intelligent evaluation module is used to evaluate the performance of each candidate geometric model scheme corresponding spare part, generates corresponding three-dimensional model multidisciplinary performance evaluation data.AI intelligent optimization module is used to carry out multidisciplinary topology optimization to three-dimensional model multidisciplinary performance evaluation data, generates optimization geometric model scheme.AI intelligent mapping module is used to carry out intelligent mapping according to optimization geometric model scheme, to generate spare part two-dimensional design drawing.

The application discloses a laser communication signal attenuation prediction and stable control method and system based on multi-modal data fusion, and relates to the technical field of laser communication, and comprises the following steps: acquiring multi-modal data of a laser communication link, taking a sampling point as a graph node, determining an edge weight according to environmental parameter similarity, and constructing graph structure data; inputting the graph structure data into a graph neural network to perform feature fusion and obtain a joint feature vector; inputting the joint feature vector into a clustering model containing an environmental variance compensation term to obtain an attenuation density distribution; and adjusting the defocusing amount of a transmitting end and the focal length of a receiving end according to the attenuation density distribution and a received power feedback closed loop, so that the received power is not lower than a preset threshold P_th and the cat-eye effect is suppressed. Through four key links of multi-source data synchronous acquisition, graph neural network (GNN) feature fusion, optimized K-means clustering and dynamic control of an optical tracking matrix, high-precision prediction and adaptive optimization of laser communication signal attenuation are realized, and the prediction precision can be improved.

The application provides a new energy station high-voltage voltage transformer error evaluation method and system, the method comprises the following steps: collecting voltage phasors of all measurement nodes in a predetermined time period in the substation and the new energy station simultaneously, obtaining voltage transformer equipment parameters and line parameters between measurement nodes, and constructing a measurement data set; based on the power system line topology structure and the data in the measurement data set, the corresponding adjacency matrix and node feature matrix are constructed by extracting the line impedance and node working condition characteristics, and the time series graph structure data set is obtained according to the adjacency matrix and the node feature matrix; a graph neural network is constructed, and the graph neural network is trained based on the time series graph structure data set; input the voltage phasor of each measurement node, and judge the voltage transformer state through the preset threshold based on the output result of the trained graph neural network. Through the scheme, the voltage transformer error state online evaluation can be realized, and the accuracy and interpretability of the error evaluation are improved.

The application discloses a network traffic data identification method and device for out-of-distribution detection of a general graph, a medium and a product, and relates to the technical field of network data identification. The method comprises the following steps: converting network traffic data into graph structure data and inputting the graph structure data into a pre-trained graph neural network; the graph neural network aggregates node neighborhood information through a graph convolution layer contained in the graph neural network, and outputs feature representation of a node or a graph and a classification logic value corresponding to a preset category; based on the classification logic value, an out-of-distribution score is calculated for each graph structure data through a preset energy function; the out-of-distribution score is compared with a preset determination threshold; if the out-of-distribution score is higher than the preset determination threshold, it is determined that the corresponding network traffic data is out-of-distribution traffic; wherein the preset determination threshold is obtained according to an out-of-distribution score distribution of known in-distribution traffic samples. The application can realize reliable and automatic identification of unknown threat traffic.

The application belongs to the field of computer data processing, and more particularly relates to a weakly supervised semantic segmentation method based on shape block semantic correlation degree. The method comprises the following steps: S1, inputting an original image into a classification network to obtain a class activation map; S2, obtaining graph structure data with shape blocks as nodes by dividing the original image through a shape division module; S3, performing shape block pooling on the class activation map by using the shape block division result in S2 to obtain a pooled class activation map; S4, training a semantic correlation degree network by using the confidence region in the pooled class activation map; S5, performing semantic classification on the graph nodes by using the adjacency matrix output by the semantic correlation degree network, and aggregating the nodes into pseudo labels; and S6, training a semantic segmentation network by using the pseudo labels, wherein the network receives an original image and outputs a predicted semantic segmentation result.