100results about How to "Improve division efficiency" patented technology

The invention discloses an equipment cluster health state evaluation method based on industrial big data and solves a health state evaluation problem of equipment clusters. Based on the full service period big data environment of the equipment clusters, an entity state information slicing management model is constructed to pre-process data; a condition similarity clustering method is utilized to carry out equipment cluster division, a regression method is utilized to establish a mirror image model, the mirror image model and an entity health state are mutually mapped, and a health state quantification model of the different equipment clusters is acquired; fusion and reconstruction of the health state data of the different equipment clusters are carried out to acquire health degrees of the different pieces of equipment, fitting of an equipment health state degeneration curve is carried out, and residual service life of the equipment is predicted. The method is advantaged in that a big data clustering modeling method is utilized, not only can effectively different evaluation on the equipment cluster health state be carried out, but also model redundancy of cluster modeling can be reduced, the internal structure of the model is simplified, normal operation of the cluster equipment is guaranteed, and use values of the equipment can be mined to a maximum degree.

The invention relates to the technical field of video play and discloses a plug-in subtitle automatic synchronization method and device. The method comprises the following steps: extracting an audio portion of a video file, and carrying out decoding on the audio portion to obtain pulse coding modulation data; dividing the pulse coding modulation data into audio clips, and classifying the audio clips; dividing the audio clips, which are classified as speeches, into short sentences, and determining the start time and end time of each short sentence; searching a match item in a plug-in subtitle file according to the determined start time and end time of the short sentences; changing the start time of the match item into presentation time stamp PTS of a current video, and updating the starting time of each item, the start time of which is larger than that of the match item, in the plug-in subtitle file according to the presentation time stamp. The display time of the subtitle file is allowed to be consistent with the play time of audio / video, thereby realizing automatic synchronization of plug-in subtitles and improving watch experience of a user.

The embodiment of the invention provides a user group partitioning method and device. The method comprises the steps of capturing user identifiers; building a joint behavior relation between the user identifiers; obtaining a user node diagram; in the user node diagram, recognizing one or more core user groups according to the joint behavior relation; in the one or more core user groups, dividing one or more target user groups according to the joint behavior relation. On one hand, manual rule setting is avoided, different user groups have different distribution features, although individual behavior differences are large, the potential relation of the user is more stable, and the user group partitioning accuracy of the user groups is improved in a diagram-based user group partitioning mode; on the other hand, by roughly partitioning the core user groups, the data size is greatly decreased, the partitioning efficiency is improved, and the user group partitioning accuracy is improved.

The invention provides a three-dimensional building model construction method and device, electronic equipment and a storage medium, and relates to the field of three-dimensional modeling. The specific implementation method comprises the following steps: extracting point cloud data of at least one building from filtered point cloud data; performing surface vector extraction on the point cloud data of the at least one building to obtain a building vector model; extracting angular points of the building vector model; obtaining a control point corresponding to the angular points of the building vector model on a pre-processed oblique photography image, and obtaining a corresponding relation based on the control point; and based on the corresponding relationship, performing texture mapping on the building vector model by using the pre-processed oblique photography image to obtain a three-dimensional building model. By adopting the embodiment of the invention, the three-dimensional model of a building can be simply, quickly and precisely established, and the real scene of an urban building is highly restored.

The invention relates to a fault diagnosis method based on multi-feature information weighted fusion under spectral clustering analysis. The method comprises the steps of firstly, carrying out spectral clustering analysis on fault equipment; secondly, obtaining the reliability of a local diagnosis evidence of each SVM to each fault mode; thirdly, constructing basic probability distribution througha local diagnosis hard output judgment matrix of each SVM; fourthly, performing weighted processing on the basic probability distribution; fifthly, obtaining the credibility and the uncertainty; andfinally, through a set diagnosis rule, and in combination with the credibility and the uncertainty, performing diagnosis. Compared with the prior art, the method has the advantages that the situationthat evidences of different sources have different reliability for identification of propositions in an identification framework is considered, the conflict between the local diagnosis of the SVMs isreduced, effective combination of the SVMs and an improved evidence theory is realized, and the shortcoming that a synthetic result cannot reflect an objective fact due to the unreliability of identification is overcome.

The invention discloses an urban traffic control sub-area division method based on fast global K-means spectral clustering, and relates the field of the urban traffic sub-area division. The basic thought thereof is taking traffic correlation between adjacent signal intersections in a road network as a division evidence, the method comprises the following steps: 1) selecting a to-be-researched roadnetwork to acquire a road topology structure and traffic flow data of the road network, and establishing a road network undirected graph; 2) computing a road network density Laplace matrix L and feature vectors corresponding to front k minimum feature values to construct a clustering feature matrix; 3) implementing the global K-means clustering algorithm on the clustering feature matrix X, and dividing the intersections in the same cluster into the same sub-area; and 4) judging whether the amount (absolute value of Vi) of the intersection in the sub-area after the clustering satisfies a constraint condition, performing boundary adjusting on the sub-area dissatisfying the constraint condition , and performing traffic control sub-area division on the road network G by adopting the fast global K-means spectral clustering. The time required by the clustering is shortened, and the traffic sub-area division efficiency can be improved.

A compressing encode method for a three-dimensional model texture based on Quardtree comprises the following steps: (1) dividing texture images according to Quardtree, resolving the texture images into a series of base textures with different resolutions, and coding the base textures; (2) arranging texture coordinates, and coding the texture coordinates successively. A decoding method corresponding to the coding method comprises the following steps: synchronously transferring texture data stream and geometric grid data, and decoding and recovering the texture data stream and the geometric grid data; and matching the texture data stream and the geometric grid data with geometrical model of the corresponding resolutions according to the textures with different resolutions. The compressing encode / decode method f of the invention has the advantages of high division efficiency, fast convergence rate as well as short client terminal responding and waiting time, and can control precision of model texture decode.

The invention relates to the technical field of computers, in particular to a WBS automatic construction method and system based on highway engineering design information. The method comprises the following steps: S1, acquiring highway engineering design information data; s2, building a WBS template; s3, extracting corresponding data from the highway engineering design information data according to the WBS element meta-information of each level in the WBS template, and generating WBS elements, wherein the WBS elements and the WBS template jointly form WBS data of the highway engineering design information. According to the method, the WBS division efficiency is improved, the engineering design information is included, the association between the WBS hierarchy and the engineering design information and the BIM model is established, the data format is unified, butt joint with an informatization application system is facilitated, and the butt joint efficiency is high.

The invention discloses a network community division method based on core nodes and a community fusion strategy, and mainly solves the resolution problem of an objective function itself in an existing community fusion algorithm. The network community division method comprises the steps of 1), reading in a network diagram S, and generating an adjacency matrix A corresponding to the network diagram S; 2), calculating the node degree of each node in a network according to the adjacency matrix A, and searching a network core node set C; 3), updating a network node label set f' according to a similarity function value F; 4), obtaining a network community label set f according to the network node label set f'; and 5), performing iterative community fusion on communities in the current network based on an improved module density increment delta D, and outputting a final network node label set fz. The network community division method based on the core nodes and the community fusion strategy provided by the invention has the advantages of being high in node information utilization ratio and high in community resolution ratio, and can be applied to community detection.

The invention relates to a method for dynamically and fairly partitioning shared cache based on a chip multiprocessor, and belongs to the field of computer architectures. Fairness is a key optimization problem; when a system lacks of the fairness, problems such as starvation of a thread and priority inversion may take place. But the current design of CMP mainly aims to improve throughput of the system, so the fairness of the system can be sacrificed generally. The invention provides a new fairness parameter and a division method. The partitioning method comprises three steps: initializing, roll-backing and repartitioning, wherein at the initializing stage, the cache is equally partitioned before the running of the thread; at the roll-backing stage, the partition which is repartitioned forthe previous partition but has reduced fairness is canceled; and at the repartitioning stage, the cache is newly partitioned for all the threads if no roll-backing occurs and the fairness parameter is greater than a repartition threshold. The partitioning method for the shared cache provided by the invention remarkably improves the fairness of the system; and the throughput of the system is also improved slightly.

The invention provides a social network partitioning method and a system based on cloud computing, and the method comprises the steps: working out maximum groups in a social network; merging the maximum groups according to the proportion of common points among the maximum groups in corresponding maximum groups; and after merging the maximum groups and obtaining a community as a new point, connecting adjacent points by edges, thus obtaining a new social network. The social network is partitioned according to the maximum groups, the new social network is obtained by taking the merged maximum groups as the new point, and the new social network can reflect analysis needs accurately and improve the utilization ratio of the social network; and simultaneously, the maximum groups are worked out in the manner of the cloud computing, which can improve the social network partitioning efficiency.

The invention provides a fracture surface pre-processing method for a variable cross-section beam finite element model, and solves a problem cross section matching in variable cross-section beam modeling. The method comprises two steps of fracture surface auxiliary mesh generation and automatic cross section matching. The main process comprises: first, segmenting in advance to form cross section blocks and automatically generating nodes and units, segmentation of the cross section blocks being completed by manpower, when the fracture surface of a variable cross-section beam is processed, region segmentation of each pair of cross sections having a similar structure; adjusting size of the cross sections and matching the units and the nodes in the cross section according to a minimum distance principle; after node matching is completed, according to the unit and node number of a reference cross section, numbering the units and nodes of a target cross section again, and finally obtaining a pair of matched cross sections. A method for determining whether segment is intersected is also provided, and line segment intersection point coordinates can be calculated. The method can be suitable to be used for various types of cross section forms in practical engineering, and can simplify modeling flow and prevent modeling mistakes.

The invention discloses a multi-dimensional end-to-end network slice generation device and method. The problems that in the prior art, a user plane only sends a slice request, slices are not created,resource division is not thorough enough, and the generated general abstract topology cannot accurately reflect the actual connection situation between network function units are solved. The method comprises the implementation steps of: (1) creating an application container, (2) generating a user plane network slice, (3) dividing a virtual network according to a flow space rule, (4) generating a control plane network slice, and (5) generating a multi-dimensional end-to-end network slice. In the process of designing the multi-dimensional end-to-end network slice generation device and method, network slices are generated from two dimensions of a user plane and a control plane, and functionally and logically isolated end-to-end network slices are generated.

The embodiment of the invention discloses a region division method and device, a storage medium and an electronic device. The method comprises the steps that latitudes and longitudes in a first data set are encoded to obtain a second data set, N groups of latitudes and longitudes with corresponding relations are recorded in the first data set, N region numbers are recorded in the second data set,and each region number is used for representing a geographic region; a third data set is obtained according to the second data set, the third data set records M groups of region numbers having a corresponding relationship, central point latitude in a geographic area represented by the region numbers, and central point longitude in the geographic area represented by the region numbers; and according to the third data set, the geographic area represented by the M area numbers is divided into P areas, P is a natural number, and each area comprises the geographic area represented by at least one area number. According to the invention, the technical problem of relatively low region division efficiency in related technologies is solved.

The invention provides an efficient image dividing method and system for larger image calculation. The method includes the steps: dividing image data into a plurality of vertexes, and randomly sortingthe vertexes into a queue; performing partition distribution on the first vertex according to a queue sequence, namely, distributing the first vertex to a processing unit, taking partition information of the first vertex as a value, taking adjacent vertexes of the first vertex as keys, and storing the information into a DRAM (dynamic random access memory) or NVM (non-volatile memory) in a dictionary entry mode after distribution; judging whether entries taking the subsequent vertexes as keys are present in the DRAM or NVM or not, and directly adding the partition information of the vertexes to the corresponding entries in the DRAM or NVM if the entries taking the subsequent vertexes as the keys are present in the DRAM or NVM; distributing the vertexes to the processing unit with minimum load if not. Partition information of each distributed vertex serves as a value, the adjacent vertexes of the vertex serve as keys, and the vertex is stored into a corresponding buffer memory in the dictionary entry mode. According to the method, the corresponding entry taking a vertex as the key can be directly searched according to the current vertex every time, and efficiency is improved.

The invention discloses a hybrid tree parallel construction method based on a GPU. A coordinate axis with the maximum patch distribution variance is selected from three coordinate axes of model space needing to be rendered, the coordinates of a location point, perpendicular to the coordinate axis, of a cutting plane are calculated, the numbers of patches on the two sides of a cut face based on the location point are made to be equal, KD tree division is carried out on scene data in the model space step by step, and then octree division is carried out on divided leaf nodes in sequence. In the process of dividing the model space, the GPU technology is used for accelerating processing, and the scene division speed is greatly improved. Due to the fact that selection division based on the KD tree is firstly carried out on three dimensions, divided scenes have the advantage that patches on the three dimensions are evenly distributed, quality guarantees are provided for the subsequent fast octree division, invalid traversal and intersection operation are greatly reduced, and the method is especially suitable for calculating visibility of nonuniform complex scenes.

The invention relates to infrastructure operation and maintenance, and provides a service area division method, an electronic device and a readable storage medium, and the method comprises the steps:after receiving a division instruction for a service area, carrying out calculation according to preset rules and the divided service areas to obtain the number of clients in each service area; Analyzing whether the number of clients in each service area conforms to a preset distribution rule or not; If the number of the clients in each service area is analyzed to accord with a preset distributionrule, automatically distributing insurance service personnel according to each currently divided service area; And if it is analyzed that the number of the clients in each service area does not accord with the preset distribution rule, sending preset prompt information to the user to prompt that the current service area is unreasonably divided and needs to be re-divided. According to the method,the dividing efficiency and the reasonability of resource distribution are improved, and the condition of unreasonable resource distribution caused by non-uniform distribution of the client quantity is effectively avoided.

A point cloud registration method based on an extended Gaussian image relates to a point cloud registration method of an image. The purpose of the invention is to solve the problem of the coarse registration algorithm SAC commonly used in the prior point cloud registration field. The overhead of IA algorithm is too large. It takes a long time to divide the Gaussian sphere into polyhedron or longitude and latitude based on EGI algorithm, and it is difficult to balance the efficiency and precision requirements when searching the optimal solution in feasible domain directly based on EGI algorithm. The procedure is as follows: the coordinate set of the normal in the respective ontology coordinate system is recorded as NT and NS; Gaussian extended image is gridded. Sorting the preliminarily matched grids in descending order according to the number of normal lines in the grids; The modified mesh is obtained; Calculating a rotation matrix; Output iter, Corr and M. The present invention is used in the intersecting field of computer graphics and machining technology.

The invention discloses a business district division method and a device, electronic equipment and a storage medium, and the method comprises the steps: obtaining historical order information, and obtaining merchant position information and order distribution information according to the historical order information; calculating the distance similarity between every two merchants according to themerchant position information and the order distribution information; clustering the merchants according to the distance similarity and a preset distance similarity to generate a plurality of class clusters; and according to the generated multiple class clusters, dividing the corresponding merchants into corresponding business districts. Compared with the prior art, according to the method of theinvention, when merchants at different positions in a specific area are automatically divided into the same business district, the merchant distance and the order intersection amount are used as measurement factors of the similarity degree between the merchants, so that the division efficiency is improved, the effective merchant coverage rate is improved, and logistics scheduling is conveniently provided in the business district.

The invention discloses a database data classification method and apparatus, and relates to the field of data processing. The apparatus comprises a parameter configuration module for setting configuration parameters including polymerization parameters and classification parameters, a general framework module for generating a data classification module based on a general calculating template according to the configuration parameters, and the data classification module for calling the data in a database to classifying data according to configuration parameters, wherein the general framework module has variables corresponding to the configuration parameters. Data classification efficiency is improved. Since a general calculating template is used, only simple configuration is needed. The expendability and universality of the apparatus are substantially improved.

The invention provides a cold and hot data analysis method and system. The cold and hot data analysis method comprises the following steps: a file data information acquisition step: analyzing a metadata node NameNode and an audit log AudiLog of a Hadoop distributed file system HDFS to obtain data information of a file in the HDFS; and a cold and hot data analysis step for analyzing the data information of the file in the HDFS by using a neural network algorithm model LSTM, and dividing the file of the HDFS into cold data and hot data by taking the data information as a reference. According tothe scheme, the file data information is analyzed through the neural network algorithm model LSTM to obtain the cold and hot data; and meanwhile, a cache elimination algorithm LRU is utilized to improve the accuracy of analyzing the cold and hot data, data support for data cleaning is provided for a service group and operation and maintenance personnel, the efficiency and accuracy of data cleaningare improved, and the personnel cost is saved.

The embodiment of the invention provides a game data processing method based on artificial intelligence and cloud computing and a game cloud center, and the method comprises the steps: representing anoperation behavior according to the dynamic and static operation data of the operation behavior; according to the operation relationship between the dynamic and static operation data and the operation behavior labels, carrying out key element matching on the operation behaviors according to the operation behavior label categories where the operation behaviors are located, and then dividing the operation behavior label categories obtained after key element matching into multiple different operation behavior objects, so that the operation behavior objects are automatically divided; then carrying out corresponding data statistics after the target operation statistical element of each target operation behavior object is determined, so that invalid operation behavior data is prevented from being processed, in addition, the division standards of the operation behavior objects are unified, updating can be carried out in time, the calculated amount is small, and if new operation behaviors areadded, the new operation behaviors can be directly added to the operation behavior objects where the new operation behaviors are located, and therefore the operation behavior object division efficiency is improved.

The invention discloses a method for designing a modular complex product based on spectral features. The method comprises a step of taking a structure, a function and a physical association as three factors that affect the correlation strength between components, quantifying the correlation strength from the three aspects, setting a weight ratio, calculating a comprehensive correlation strength between the components, and constructing a design structure matrix, a step of combining the graph theory and applying singular value decomposition to extract first k largest spectral features of the design structure matrix and corresponding feature vectors to form a streamlined spectral mapping space, a step of using a k-means algorithm to perform clustering operation in the space to obtain a program set of module division, and a step of selecting an optimal division scheme by using an extended weighting modularity EWQ function and combining and modeling modules in the optimal scheme in a software environment to obtain a three-dimensional model of a loader working device. According to the method, products with performance and specifications in response to client requirements can be quickly achieved according to the client requirements, and the modular design of a wheel loader working device is achieved.

The invention discloses a graph division method and device and a computer readable storage medium. The method comprises the following steps: determining a graph division rule for proportionally distributing graph data to each computing node in advance according to the computing performance of each computing node of the parallel computing system and a preset measurement mechanism; in a graph division process, determining a data reading range of each loading thread according to a graph vertex number and a loading thread number, simultaneously calling a plurality of loading threads to read graphdata to be divided in parallel by taking a vertex as a unit, and transmitting the graph vertex data read by each loading thread to a corresponding division thread. Each division thread determines a sub-graph to which each piece of graph vertex data transmitted into the corresponding thread belongs according to a graph division rule based on the graph data divided at the current moment, and finallydivides each piece of graph vertex data to the corresponding calculation node based on a corresponding relationship between the sub-graph and the calculation node. The large graph can be efficientlyand accurately divided into a series of sub-graphs with the vertex number or the connecting edge number meeting any proportional relation.

The invention discloses a space variable-radius spherical domain division method for orthodontic arch wire bending sequence planning, and relates to the technical field of orthodontic arch wire bending. The space variable-radius spherical domain division method for orthodontic arch wire bending sequence planning is established according to a personalized orthodontic arch wire curve of a patient, based on an orthodontic arch wire curve bending point information set and a bending point robot bending information set, and in combination with the movement characteristics of a robot bending orthodontic arch wire. The technical key points are as follows: variable-radius spherical domain division data is imported; the upper limit value of a spherical domain limiting parameter is set; the radius and the sphere center of a divided spherical domain are determined; a reasonable bending spherical domain is defined; whether spherical domain division is continued or not is judged; and with the density of spherical domain bending points as an index, all spherical domains are arranged in a descending order, the bending point sequence of the spherical domains is specified, and the final bending point bending sequences T1 and R1 are output. According to the method, the space arch wire curve is divided into regions by the variable-radius spherical domains, so that the accuracy of orthodontic archwire bending planning can be improved, and the problem of interference in the process of bending the orthodontic arch wire by the robot is avoided.

The invention provides an SDH (Synchronous Digital Hierarchy) based wide-area peer-to-peer communication system and method. An area protection and control device and an EOS board card of an SDH device are interacted through an FE interface due to the specific message switching capability of the EOS board card of the SDH device, the area protection and control device sends multicast messages to the EOS board card of the SDH device through the FE interface, the EOS board card receives and identifies a VLAN (Virtual Local Area Network) tag in the multicast messages and judges whether the VLAN multicast message exist in a VLAN configuration list or not; transmitting the VLAN multicast message if the VLAN multicast message exists and not transmitting if the VLAN multicast message does not exist. The integral system can greatly improve the network partition efficiency of an area protection and control system which is large in size and complex in communication relationship due to the fact that a one-send-multi-receive peer-to-peer communication mechanism is achieved on the SDH device and enables the area protection and control system to be provided with clear network architecture and later maintenance and management to be convenient.

The invention provides a wide-area protection peer-to-peer communication system and method based on transmission equipment packet transport network (PTN) planes. The wide-area protection peer-to-peer communication system is characterized by comprising more than one piece of wide-area protection equipment, wherein each piece of wide-area protection equipment corresponds to one PTN plane; the PTN planes of the wide-area protection equipment comprise access ports and cascade ports; the wide-area protection equipment is connected with corresponding PTN planes through the access ports; the PTN planes are connected together through the cascade ports; and the PTN planes perform static multicast table configuration and flow control configuration according to wide-area protection services. Through adoption of the wide-area protection peer-to-peer communication system, the defect of over dependence on a dedicated E1 channel in service division in an existing power system transmission network is overcome completely; the network division efficiency is increased greatly for a large-scale wide-area protection system with a complex communication relationship; the wide-area protection system has a clear network framework; and convenience is brought to later-stage maintenance, management and the like.

The invention provides a method and a system for dividing a scheduling period of a power distribution network. The method comprises the steps of S1, acquiring load data of n nodes in the power distribution network within a specified scheduling period; S2, calculating fluctuation degrees of all bisection solutions of the scheduling period by using the load data, and dividing the scheduling period into two time periods according to the scheduling period bisection solution with the fluctuation degree being the minimum; judging whether the number of the divided time periods is less than a preset number or not, executing a step S4 if the number of the divided time periods is less than the preset number, and otherwise, stopping division; S4, calculating fluctuation degrees of all bisection solutions of each time period by using the load data, and respectively determining a time period bisection solution with the fluctuation degree being the minimum corresponding to each time period; S5, determining a time period bisection solution with the fluctuation degree being relatively small; and S6, further dividing the corresponding time period into two time periods according to the time period bisection solution with the fluctuation degree being relative small, keeping the rest of the time period being unchanged, and returning back to the step S3.