61 results about "Adjacency list" patented technology

The invention discloses a curved surface quadrilateral dominant grid generation method, device and equipment for an aircraft and a storage medium, and relates to the technical field of computational fluid mechanics, and the method comprises the steps: creating a surface unstructured grid based on surface grid generation parameters of a CAD file of the aircraft, carrying out the conformal mapping of the surface unstructured grid to a parameter domain plane, and calculating the vertex compression ratio of each grid vertex; selecting the current starting point with the minimum compression ratio and the maximum to-be-compared vertex in the adjacency list, and determining the current to-be-processed vertex with the minimum compression ratio in the adjacent vertexes and the advancing direction; determining to-be-processed discrete points according to the current compression ratio and the coordinate information to generate a current isoline, if the current compression ratio does not exceed a threshold value, iteratively updating the current starting point, and skipping to the step of determining the current to-be-processed vertex again; the discrete points are projected to the outer-layer isolines according to the isoline sequence, and the curved-surface quadrilateral dominant grid is generated based on the distance proportion between each discrete point and the projection point, so that the generation efficiency of the curved-surface quadrilateral grid is improved.

The invention discloses a CAD two-dimensional and three-dimensional constraint system graph decomposition method and system and a medium, and relates to the CAD technology, and the method comprises the steps that a CAD constraint system is constructed, and the CAD constraint system comprises geometric primitives and constraint objects; geometric primitives and constraint objects in the CAD constraint system are stored by adopting an adjacency list, and the CAD constraint system is converted into a constraint adjacency graph structure; segmenting the constrained adjacent graph structure into a plurality of double-connected sub-graphs; and for each double-connected sub-graph, decomposing the double-connected sub-graph into SPQR three-connected sub-graphs. According to the embodiment of the invention, the large-scale constraint system is decomposed into a subsystem set with a good local structure through combination of cut point segmentation of the constraint system diagram and SPQR decomposition, and association between subsystems is effectively decoupled.

The invention discloses an information retrieval method, device and equipment and a computer readable storage medium, and relates to the technical field of natural language process.The method includes the steps that after to-be-retrieved text data is received, vectorization processing is conducted on the to-be-retrieved text data, the storage space and the calculation overhead are greatly reduced, a hierarchical index structure is constructed in advance, and the retrieval efficiency is improved. An adjacency list corresponding to each vector node in a hierarchical index structure is set to contain a fixed preset number of similar vectors, a parallel computing architecture of the graphic processing unit is adapted, parallel information retrieval of the graphic processing unit is facilitated, rough information retrieval from a coarse-grained index end is transited to fine information retrieval layer by layer, and the information retrieval efficiency of the graphic processing unit is improved. The information retrieval process is accelerated, the technical problem that the parallel computing capacity of the image processing unit cannot be fully utilized is solved, and the technical effects that the parallel processing capacity of the image processing unit is fully utilized, the performance is greatly improved, and the information retrieval efficiency is improved are achieved.

The present disclosure provides a method for processing a secret-shared adjacency list of a graph. The method includes partitioning the secret-shared adjacency list into blocks, adding a vertex identifier to each tuple in the secret-shared adjacency list, defining a boolean variable for each block, computing a secret-shared vector of row identifiers for each tuple, shuffling tuples and additional tuples, extracting components from the shuffled tuples to form a d-normalized replicated adjacency list, and storing the d-normalized replicated adjacency list in a non-volatile storage device. The method utilizes parallel processing capabilities and a hardware-based random number generator for shuffling. The method also includes securely transmitting and receiving encrypted data related to adjacency lists through a network interface.

The invention provides an anomaly detection method and system for a streaming traceability graph based on iterative prediction correction. The anomaly detection method comprises the following steps: S1, collecting streaming data; S2, sampling and coding coding information of each node of the streaming traceability graph based on the streaming data; and S3, predicting an abnormal node based on coding information, adopting an adjacency list to cache a streaming event, adopting a sliding time window mechanism to update the adjacency list, adopting a Gaussian mixture model to predict a predicted node state of the node, and then correcting the predicted node state with a current node state. According to the method, the current node is detected, the coding information of the current node is obtained through weighted aggregation by taking the event frequency as the weight of the neighbor node, the abnormal point is judged based on the coding information, real-time detection is realized by adopting a stream processing mode through the stream graph neural network, and the problems of system scale and detection precision are solved.

The invention discloses a rapid data segmentation method for large-scale parallel graph neural network calculation, and relates to the field of data segmentation. The method comprises the steps that when streaming graph processing is carried out on vertexes, a super adjacency list of k-Hop is constructed, and therefore the complex dependency relationship in the GNN calculation process can be better captured. The influence of different types of neighbors on vertex distribution is particularly concerned, in the GNN training process, nodes are usually divided into training points, test points and verification points, and the influences of different types of neighbors on vertexes are different. Due to the fact that multiple rounds of iteration are usually carried out in the training stage and only one round is carried out in the testing stage, for a certain node, the node and a training neighbor are more expected to be located in the same partition, and communication overhead is reduced. Through partition information of the distributed nodes, expected values of neighbors in the partitions are increased. Namely, through the combined action of the distribution information of the inner neighbors of each node and the distribution information of the out neighbors in the k-Hop adjacency list, the effectiveness of division is enhanced.

The invention provides a path texture driven 3D surface processing track generation method and device, and the method comprises the steps: 1, obtaining the coordinates of three vertexes of a triangular patch and a first-order adjacency list of each vertex based on an STL model; step 2, constructing mapping from a curved surface path to a two-dimensional image path based on the vertex coordinates and the first-order adjacency list obtained in the step 1; 3, processing the path texture image to obtain a point sequence, and then reversely mapping the point sequence to the STL model based on mapping to obtain an initial track point set; and step 4, performing optimization processing on the initial track point set to generate a path point set. According to the method, the path texture image and the STL model serve as input, and the continuous track capable of being used for precision machining tasks such as industrial grinding is generated.

The present invention discloses a compression index and query method for an attribute graph, which is applied to an attribute graph database system. The attribute graph maintained by the attribute graph database system includes a plurality of vertices, edges, and their corresponding attributes. The method includes: constructing an adjacency list for each vertex according to the attribute graph; constructing a compressed structure of the graph structure according to the adjacency list of each vertex; calculating vertex / edge numbers according to the compressed structure of the graph structure, obtaining the location of the attribute in the attribute compression index according to the vertex / edge numbers and the corresponding bit arrays, and extracting the attribute values from the attribute compression index; the attribute compression index includes a vertex attribute compression index and an edge attribute compression index; the bit arrays correspondingly record the starting positions of all edge attributes / vertex attributes; querying the attribute graph according to the compressed structure of the graph structure and the attribute compression index. The method of the present invention improves the query efficiency of the attribute graph database and reduces the storage space.

The invention belongs to the field of complex document processing, and provides a complex document-oriented dynamic DAG task decomposition and scheduling method and system in order to solve the problem of low complex document generation efficiency. The complex document-oriented dynamic DAG task decomposition and scheduling method comprises the following steps of: constructing an initial directed acyclic graph to form an initial task chain; receiving a node insertion request, calculating a dependency probability between any two nodes in all the non-ring adjacency lists by using a graph neural network model, and determining whether to add an implicit dependency edge according to the dependency probability so as to form an implicit dependency edge set; calculating task resource requirements of a to-be-generated document according to the video memory prediction model, dynamically allocating hardware resources and executing corresponding tasks in combination with node task types and resource requirement labels of all implicit dependency edges in the implicit dependency edge set, and aggregating execution results of all the tasks to generate a final document. The reliability and efficiency of a large model in a complex document generation task can be effectively improved.

The invention provides a line fault positioning method and device, equipment, a medium and a computer program product. The method comprises the following steps: analyzing acquired pipeline information to obtain a resource mapping relationship; performing multi-dimensional analysis on the fault original data according to the resource mapping relation to obtain a pre-positioned pipeline section set; converting the pipeline segment set to obtain a weighted undirected graph, and converting the weighted undirected graph to obtain an adjacency list; and determining a post-positioned target fault path based on the weighted undirected graph, the adjacency list and the OTDR measurement distance. The method comprises the following steps: performing multi-dimensional analysis on line fault information to obtain the breakpoint probability of each pipeline section in a fault area, and screening according to the probabilities to obtain a pre-positioned pipeline section set; and then, by inputting OTDR positioning conditions and results, marking out a route on a fault point path, displaying a pipeline position where a fault point is located, and improving the optical cable fault positioning precision through path visualization.

The embodiment of the invention provides a data storage method, device and equipment based on graph division, and the method comprises the steps: obtaining a to-be-processed graph structure, and determining a corresponding adjacent matrix; determining a label dictionary and an adjacency list based on a vertex connection relation and vertex attributes in the adjacency matrix; determining initial partitions corresponding to the vertexes in the to-be-processed graph structure according to the vertex identifiers and a preset partition number, determining target partitions of the adjacent vertexes and connecting edges based on target similar attributes of each group of adjacent vertexes in the adjacent list and the initial partitions of the adjacent vertexes, and updating the adjacent list based on the target partitions; and storing the data information in the to-be-processed graph structure by using the updated adjacent list and label dictionary in a target storage mode. According to the technical scheme provided by the embodiment of the invention, load balancing among the distributed storage nodes is realized by accurately dividing the target partitions of the to-be-processed graph structure, and the data information is stored in the target storage mode, so that the data access efficiency is effectively improved.

The invention belongs to the field of molecular dynamics, and provides a molecular dynamics adjacency list construction optimization method, a molecular dynamics adjacency list construction optimization system and a super computer platform in order to solve the problem that the cache hit rate is reduced when a traditional adjacency list is constructed. The molecular dynamics adjacency list construction optimization method comprises the following steps: mapping atomic coordinates in each physical space to three-dimensional integer grid coordinates; carrying out one-dimensional mapping coding on the three-dimensional integer grid coordinates of all the atoms and sorting all the atoms; clustering all the sequenced atoms to obtain a plurality of atom clusters and initializing sharing states in the atom clusters; constructing a candidate neighbor region of each atomic cluster based on the initialized sharing state in each atomic cluster; and uniformly traversing the candidate neighbor regions of each atomic cluster according to a vectorization instruction, constructing an adjacency list of each atom of each atomic cluster, and writing the adjacency list into a global adjacency list, so that larger-scale and higher-efficiency molecular simulation is realized.

The present disclosure provides a distributed graph data processing system, method, apparatus and device, and a storage medium. The system includes: a computing node in a distributed architecture, wherein the computing node includes a computing unit and a persistent storage unit that are in a corresponding relationship, and a storage manager is deployed in each computing unit. The storage manager divides the graph data adjacency list into graph data blocks, determines a key value corresponding to each of the graph data blocks, and then records and stores a corresponding relationship between key values of the graph data blocks and storage location information.

The application provides a graph database query method and device, electronic equipment and a storage medium. The method comprises the following steps: according to a query condition, querying target edges corresponding to a starting vertex and attribute information of the target edges in a graph database; generating an adjacency list of the starting vertex according to the target edges and the attribute information; and querying a path in the graph database based on the adjacency list. By querying the target edges corresponding to the starting vertex in the graph database, the adjacency list is constructed based on the target edges and the attribute information. The adjacency list comprises path relationships between vertices. The path in the graph database is queried based on the adjacency list, and a required path is found, thereby improving the situation that the query operation fails when directly querying the graph database, and improving the accuracy of the query result.

An electric energy tracing method based on a network topology sorting algorithm relates to the technical field of smart power grids, and comprises the steps of collecting data, constructing a power grid topology directed graph by using an adjacency list, carrying out topology sorting, executing topology sorting according to a power grid topology graph, and carrying out electric energy component accurate accounting and tracing derivation by using an algorithm. The method has the beneficial effects that the hybrid energy topological graph is constructed, accurate calculation of electric energy components and accurate analysis of electric energy traceability are realized by utilizing an electric energy flow analysis technology and combining real-time operation data, synchronous planning, synchronous construction and synchronous commissioning of a matched power grid project and a new energy project are met, and the reliability of the system is improved. And favorable green electricity strategy requirements are created for large-scale new energy development and grid-connected consumption.

The present invention relates to a method for additive homomorphic encryption and cryptographic accumulator calculation. The present invention proposes a verifiable outsourced calculation method for graph data based on additive homomorphism and bilinear map accumulators. First, homomorphic cryptography is used to encrypt graph data instantiated by an adjacency list. Additive homomorphic encryption and obfuscation circuits are combined to implement a breadth-first Dijkstra algorithm based on a priority queue to calculate the exact shortest path on the encrypted graph data. Second, a binomial heap structure is constructed based on the encrypted graph data, and a bilinear map accumulator is used to calculate verification evidence for the encrypted shortest path result, thereby achieving verifiable correctness of the outsourced shortest path query calculation result for the encrypted graph data. In addition, the scheme also uses certificateless short signatures to ensure the authenticity and non-tamperability of the shortest path query result and verification result of the cloud server, thereby achieving probabilistic correctness verification of the outsourced calculation result of the shortest path query for encrypted graph data.

The invention discloses a limited resource multi-task matching method based on a greedy strategy, and belongs to the field of scheduling optimization. The method comprises the following steps: screening all feasible solutions for task requirements from limited resources according to a multi-task requirement table, and constructing a multi-task feasible solution table; constructing a limited resource undirected weighted graph by using an adjacency list according to the multi-task feasible solution table; on the basis of a greedy strategy, traversing the multi-task feasible solution table to obtain a resource packet preliminary screening table; judging the number of feasible solutions required by each task in the resource packet preliminary screening table; and for task requirements of which matching is not completed, a breadth-first search algorithm is used for searching a channel in the limited resource undirected weighted graph and the multi-task feasible solution, and resource transfer and allocation are carried out. According to the method, the greedy strategy is ingeniously utilized for preliminary matching, and the problem scale is reduced; on the basis, the defect that a greedy strategy is prone to falling into a local optimal solution is overcome through a breadth-first search algorithm, and a global feasible solution can be found more efficiently.

The invention provides a full-coverage path planning method and system based on a minimum spanning tree of transverse and longitudinal control and an A * algorithm, and the method comprises the steps: (1), carrying out the rasterization of an original map of a robot, and obtaining a raster map, and enabling each raster to be the size of one robot; (2) in order to improve the time efficiency and reduce the calculation complexity, carrying out partition processing on the grid map; (3) each grid partition is subjected to composition according to the up-down-left-right relation, and a graph structure represented by an adjacency list method is obtained; (4) executing a minimum spanning tree algorithm on each grid partition to fulfill the aim of fully covering the partition; (5) in order to obviously reduce the problem of excessive turning in full-coverage path planning, executing transverse and longitudinal control on the minimum spanning tree algorithm; (6) different grid partitions are migrated through A *; and then the steps (3)-(6) are repeated until all the grid partitions are covered. The method introduces the minimum spanning tree algorithm of transverse and longitudinal control and A * to solve the full-coverage path planning, has the advantages of high coverage rate and low repetition rate, reduces the traversal time, reduces the turning times through transverse and longitudinal control, and improves the system efficiency. And meanwhile, through partition processing, the complexity of the adjacency list representation graph structure is reduced, and the time consumption of algorithm planning is remarkably reduced.

The invention relates to a cable shortest path optimization method and device based on an improved Astar algorithm, and a medium. The method comprises the following steps: constructing and initializing a record matrix B, a path matrix P, a minimum heap Q and an adjacency list matrix G; updating the adjacency list matrix G according to the bridge network adjacency matrix A; updating the minimum heap Q according to the starting point coordinate and the terminal point coordinate of the cable to be laid; updating the path matrix P in combination with the minimum heap Q and the adjacency list matrix G; and according to the updated path matrix P, obtaining the laying path of the cable to be laid. According to the invention, the operation efficiency of the Astar algorithm is greatly improved.

The invention discloses a human haplogroup hierarchical tracking and macro haplogroup normalization method, and relates to the field of group genetics and bioinformatics software methods. Comprising the following steps: establishing an adjacency list data structure based on an authoritative phylogenetic tree; constructing a synonym normalization dictionary for processing the heterogeneity problem of haplogroup naming; a hierarchical path backtracking algorithm is realized, backtracking is performed from a target haplogroup node to a root node based on a depth-first search principle, a complete evolutionary path is generated, the hierarchical path backtracking algorithm comprises a path integrity verification mechanism, a broken path or a circular reference error is automatically detected and reported, and the biological rationality of the generated path is ensured; a path output interface is provided, and a positive-sequence path representation mode and a negative-sequence path representation mode are supported. The method has the advantages of being high in standardization degree, high in automation level, high in reproducibility, good in universality and good in user experience.

The invention provides a method and a device for constructing a distributed approximate neighbor search graph index. The method comprises the following steps of: initializing each process, dividing batches for a read-only vector data set, and distributing a message buffer area; if the value of the counter is an even number, unprocessed data set batches are taken out, an adjacency list of each point is calculated in parallel, and a tuple is generated; if the counter value is an odd number, determining the affected points in the graph index updating process, and performing parallel pruning again and updating; the adjacency list length information is written into an adjacency list length array in parallel in the process; copying the information to a message sending buffer area in parallel by utilizing a parallel prefix and a calculation offset; broadcasting a sending message buffer area of each process; deserializing a tuple from the received message buffer area by each process, and updating a graph index in the process; generating tuples for the adjacency table of the updated points; printing related results by one process and outputting a graph index. According to the method, the index construction time is greatly shortened; and higher query accuracy can be provided in a query stage.

This application relates to the fields of power distribution network technology and deep learning technology. It discloses a method, apparatus, and device for generating power distribution network topology diagrams based on temporally explicit networks. The method includes: a power distribution network automation system generating an adjacency list for the current time; dividing multiple nodes in the adjacency list into multiple regions, selecting the bus node of each region as the traversal start node for each region, creating a corresponding worker thread for each region, and performing a breadth-first search traversal operation on each region starting from the traversal start node to obtain a connection tree for each region; merging the connection trees of each region to generate the connection tree of the power distribution network; converting the connection tree of the power distribution network into a topology matrix of the power distribution network; inputting the topology matrix of the power distribution network into a drawing tool; and having the drawing tool generate the topology diagram of the power distribution network at the current time. This application is beneficial for improving the efficiency of topology diagram generation.

This solution provides an anomaly detection method and system for streaming traceability graphs based on iterative prediction correction, including the steps: S1: Collect streaming data; S2: Sample and encode the encoding information of each node in the streaming traceability graph based on the streaming data; S3: Predict abnormal nodes based on the encoding information. This solution uses an adjacency list to cache streaming events and a sliding time window mechanism to update the adjacency list. It uses a Gaussian mixture model to predict the predicted node state of a node and then corrects it with the current node state. Based on the event frequency as the weight of neighbor nodes, weighted aggregation is used to obtain the encoding information of the current node. Based on the encoding information, abnormal points are judged, and real-time detection is implemented in a streaming manner through a streaming graph neural network, solving the problems of system scalability and detection accuracy.

The invention discloses an adjacency list-based model calculation sequence and algebraic ring judgment method and system. The method comprises the following steps of: initializing an adjacency list, and calculating in-degrees of all model vertexes; searching model vertexes which need to be deleted at this time and have in-degree of 0; deleting model vertexes and updating the adjacency list; repeating the steps until the model vertex with the in-degree of 0 does not exist; judging whether undeleted model vertexes exist or not, and if yes, indicating that the remaining model vertexes form an algebraic ring; determining a calculation sequence of the model according to the deletion sequence of the model vertexes; the system comprises an initialization module, an algebraic ring module and a sequence calculation module. Model information is stored in an adjacency list mode, and more space is saved; the calculation speed for a large model is high; when the model sequence is calculated, whether an algebraic ring exists or not is detected, and a model which finally forms the algebraic ring is prompted to a user.

The application provides a kind of oiling path planning method and device, and a kind of computer readable storage medium.The planning method includes the following steps: according to the process flow diagram of crude oil tank farm, directed graph model is established, and the attribute data and state data of process equipment and / or pipeline in the crude oil tank farm are saved to corresponding node and / or corresponding link;Adjacency list is established to store the attribute data, state data, index information of the pointing node of each node and link information connected to the pointing node;Obtain at least one subtask starting node information, termination node information and ideal pump flow data;According to multiple optimization objectives, the directed graph model and the adjacency list, the objective function of the multi-objective optimization problem is established, and the feasibility constraint conditions of path search and path combination are set;And based on the feasibility constraint condition, multi-objective evolution operation is carried out to determine at least one non-dominated path planning scheme.

The invention relates to the technical field of power distribution networks and the technical field of deep learning, and discloses a power distribution network topological graph generation method, device and equipment based on a time clear network, and the method comprises the steps: a power distribution network automation system generates an adjacency list at a current moment; dividing a plurality of nodes in the adjacency list into a plurality of areas, selecting a bus node of each area as a traversal starting node of each area, creating a working thread corresponding to each area, and starting from the traversal starting node in each area through the working thread corresponding to each area; performing a traversal operation of breadth-first search on each region to obtain a connection tree of each region, merging the connection trees of each region to generate a connection tree of the power distribution network, converting the connection tree of the power distribution network into a topological matrix of the power distribution network, inputting the topological matrix of the power distribution network into a drawing tool, and drawing the topological matrix of the power distribution network. The drawing tool generates a topological graph of the power distribution network at the current moment. The generation efficiency of the topological graph can be improved.

A wireless communication resource allocation method based on a depth-unrolled distributed graph neural network provided by the present invention obtains a channel gain matrix of the current time slot from a D2D network and converts it into a graph data structure stored by an adjacency list; any transmitter determines its own user channel gain and the status information of other transmitters in the graph data structure; taking the user channel gain of the current time slot, the transmission power of the previous time slot, and the status information of other transmitters as inputs, and aiming at maximizing the sum of communication rates, through the operation mode of a multi-layer graph neural network simulated and trained in multiple time slots, the finally allocated transmission power is obtained to transmit signals. The present invention can enhance the interpretability and generalization of the graph neural network, reduce the dependence of the graph neural network on training samples, use distributed parallel processing for iterative processing of transmission power, reduce the task calculation delay, and meet the actual requirements of future D2D network communication.

The invention discloses a workflow automatic layout method based on topological sorting and dynamic adjustment, which comprises the following steps of: analyzing input data, constructing a workflow directed graph of modules and a connection relationship thereof, constructing an adjacency list and an in-degree list based on the workflow directed graph, and then hierarchically dividing the modules by using an improved topological sorting algorithm, and dynamically calculating the coordinate position of each module in the canvas in combination with canvas boundary constraint and minimum spacing limitation, finally outputting JSON data containing complete layout information of each module, and realizing flow chart drawing in combination with a front-end visualization technology. According to the scheme, the problems of low manual layout efficiency, node omission, layout chaos and the like caused by loop dependence in a traditional layout algorithm are effectively solved, and the method has a good visualization effect and high adaptability and can be applied to automatic layout and scene generation of the flow chart of modular systems such as a software system, a micro-service architecture and a low-code platform.

The invention discloses a graph rearrangement algorithm for optimizing GNN training, relates to artificial intelligence, and provides a scheme for solving the problem of memory access bottleneck in the prior art. Comprising the following steps: S1, adjacency list extraction and node classification; s2, performing initialization and random selection; s3, selecting a next node to be sorted by using a maximum public neighbor strategy and a random selection strategy; s4, iteration: repeatedly executing the step S3 until all non-isolated nodes are sorted; s5, a processing step of isolated nodes; and S6, updating node attributes. The method has the advantages that GNN training is optimized specially, millions of node maps can be sorted quickly, data locality is improved to a great extent, the hit rate of a memory access place is increased, and memory access delay is reduced. The method not only can be used for acceleration of GNN training, but also can be used for acceleration of SpMV; this is natural, because the locality of the data is increased, the hit rate of memory access can be improved when the data is accessed, and then the calculation of the SpMV is accelerated.

The invention relates to a feature extraction method, device and equipment based on an industrial triangular mesh model, and the whole scheme. According to the invention, connectivity analysis is carried out on the industrial triangular mesh model, a connected region can be rapidly divided, a clear data framework is provided for subsequent processing, unnecessary calculation is reduced, and the overall processing efficiency is improved; an angle limitation triangular surface adjacency list is generated based on connectivity analysis and preset normal included angle range limitation, invalid adjacency relations are accurately filtered, the data complexity is reduced, and the processing speed is further increased; boundary and vertex information is marked, so that boundary features are clear and definite, boundary recognition errors are avoided, and the accuracy of feature extraction is guaranteed; a boundary edge adjacency list is generated, continuous boundary edges are obtained, the topological structure of the model boundary is completely reserved, and a reliable basis is provided for shape feature extraction; finally, shape features extracted based on continuous boundary edges can highly restore geometric features of the model.