1130 results about "Data combination" patented technology

The present invention discloses an aero engine bearing fault diagnosis method based on STFT-IncepNext, which includes the following steps: step 1: collect acceleration signals under various working conditions and fault conditions of the bearing as sample data; step 2: determine the number of sampling points of each sensor first, and perform data merging by permuting and combining data collected by sensors at different positions in the same time series, and then use short-time Fourier transform to convert the fused vibration signal into a time-frequency diagram; step 3: data preprocessing: divide into training set, test set and validation set according to fault type, and normalize the converted time-frequency diagram using average value and standard difference; step 4: build an IncepNext model, train and validate it; and step 5: use the trained IncepNext model to diagnose aero engine bearing faults and obtain a diagnosis result. The present invention can effectively identify aero engine bearing faults, can guarantee the fault identification rate, and has good accuracy.

The invention provides a fragment data cross-modal label generation system and method based on deep transfer learning. The method comprises the following steps: extracting first high-dimensional feature vectors in different modes; mapping the first high-dimensional feature vectors of different modals into the same semantic space through a cross-modal comparison loss function to realize multi-modal alignment and fusion to obtain second high-dimensional feature vectors; labeling semantic tags corresponding to the second high-dimensional feature vectors based on the fragmented data components by adopting a small sample transfer learning algorithm; a multi-channel Hash encoder is adopted, a self-adaptive encoding strategy is called according to different modal data combinations, and the second high-dimensional feature vector is encoded into a multi-channel binary Hash code; in combination with an incremental graph neural network, the binary hash codes and the corresponding semantic tags are dynamically expanded into the historical knowledge graph; matched fine-grained tags are established for semantic differentiation features of different entity combinations in the target knowledge graph, and a cross-modal tag tree is obtained by combining three-matrix hierarchical construction.

Systems and techniques for generating and animating non-player characters (NPCs) within virtual digital environments are provided. Multimodal input data is received that comprises a plurality of input modalities for interaction with an NPC having a set of body features and a set of facial features. The multimodal input data is processed through one or more neural networks to generate animation sequences for both the body features and facial features of the NPC. Generating such animation sequences includes disentangling the multimodal input data to generate substantially disentangled latent representations, combining these representations with the multimodal input data, and using a large-language model (LLM) to generate speech data for the NPC. Further processing using reverse diffusion generates face vertex displacement data and joint trajectory data based on the combined representation and generated speech data. The face vertex displacement data, joint trajectory data, and speech data are used to produce an animated representation of the NPC, which is then provided to environment-specific adapters to animate the NPC within a virtual digital environment.

The invention provides an urban green land coverage rate dynamic monitoring method and device. The method comprises the following steps: acquiring multispectral satellite remote sensing data, and preprocessing the multispectral satellite remote sensing data; carrying out cloud detection and shadow detection on the preprocessed remote sensing image, and calculating an effective pixel proportion index, a spectrum quality index, a time phase matching degree index and a spatial integrity index; based on the four indexes and the corresponding weights, constructing a fuzzy comprehensive evaluation matrix, and selecting an optimal data combination; calculating a normalized difference vegetation index (NDVI) based on the near infrared band and the red light band in the optimal data combination, and performing illumination angle correction and time sequence smoothing processing on the NDVI to obtain an optimized NDVI; performing green space identification on the optimized NDVI by adopting a first deep learning model, and outputting a green space distribution map; and calculating a green land area change rate, an NDVI difference value and connectivity index change based on the green land distribution map. According to the invention, high-precision dynamic monitoring of the green land coverage rate is realized, and the monitoring efficiency and accuracy are improved.

Systems and techniques for generating and animating non-player characters (NPCs) within virtual digital environments are provided. Multimodal input data is received that comprises a plurality of input modalities for interaction with an NPC having a set of body features and a set of facial features. The multimodal input data is processed through one or more neural networks to generate animation sequences for both the body features and facial features of the NPC. Generating such animation sequences includes disentangling the multimodal input data to generate substantially disentangled latent representations, combining these representations with the multimodal input data, and using a large-language model (LLM) to generate speech data for the NPC. Further processing using reverse diffusion generates face vertex displacement data and joint trajectory data based on the combined representation and generated speech data. The face vertex displacement data, joint trajectory data, and speech data are used to produce an animated representation of the NPC, which is then provided to environment-specific adapters to animate the NPC within a virtual digital environment.

The invention relates to the technical field of data processing, and discloses a multi-source heterogeneous big data processing system, which comprises the steps of dividing data into structural sub-data and non-structural sub-data, performing feature extraction on the non-structural sub-data to obtain multi-modal features, so that the data can be compared in a unified feature space, and then, performing feature extraction on the non-structural sub-data to obtain multi-modal features. The similarity and relevance between the data are identified through pattern matching, it is ensured that the data of the same entity are identified to be directly fused, redundancy is reduced, accuracy is improved, then the most similar data combination is selected through an optimization algorithm, and the actual relation type between the data is determined through relation deduction; and finally, the most suitable fusion mode is selected according to the relationships. According to the method, the accuracy and effectiveness of the data fusion process are ensured, and a reliable basis is provided for deeper analysis and decision making.

The invention relates to a vehicle suspension control method and device, computer equipment and a readable storage medium. The method comprises the following steps: combining obtained vehicle body dynamics time series data, vehicle control time series data and environment perception time series data in a vehicle driving process into time series data of multiple channels, determining a self-adaptive modal number of the time series data under each channel according to the time series data under each channel, and determining the self-adaptive modal number of the time series data under each channel according to the self-adaptive modal number of the time series data under each channel; and performing modal decomposition on the time sequence data based on the adaptive modal number, extracting frequency domain characteristics of the time sequence data, integrating the frequency domain characteristics of the time sequence data of the multiple channels, inputting the frequency domain characteristics into a trained suspension control parameter prediction model to obtain target time sequence characteristics for predicting suspension control parameters, and performing prediction on the suspension control parameters based on the target time sequence characteristics. Vehicle suspension control parameters are obtained through prediction, and a suspension system of the vehicle is controlled based on the vehicle suspension control parameters. By adopting the method, the risk of rollover of the vehicle can be obviously reduced.

The invention relates to the technical field of CIM (common information model) platforms, in particular to a multi-element and multi-scale data management method and system based on a CIM basic platform, and relates to data acquisition, access and classification in a service interface, incremental acquisition and offline copying mode. Data needs to be translated to conform to spatial, elevation and temporal references of a CIM platform, unify geocoding and spatial expression, and semantic-based place name address association. And the BIM and the CIM are fused through a knowledge graph. And performing three-dimensional data grading processing, and constructing CIM1 to CIM5 grade data. Government affair data are subjected to graph falling through a unified address library. Overall data governance covers data aggregation, classified governance, two-dimensional and three-dimensional fusion, BIM and CIM fusion technology and government affair data association, classification and level-to-level management of data is realized, different data combination application is supported, and CIM business application and government affair process auxiliary approval and decision making are assisted.

The invention provides a data leakage prevention system building method, which comprises the steps of building a data leakage risk prediction model according to principal component characteristics, training the model by adopting a support vector machine algorithm to obtain a data leakage risk prediction model, inputting to-be-predicted multi-source heterogeneous data into the risk prediction model, and obtaining a data leakage risk prediction result. Calculating a leakage risk score of the data combination; a sensitivity threshold value is preset, whether the leakage risk score exceeds the preset sensitivity threshold value or not is judged, if the leakage risk score exceeds the preset sensitivity threshold value, the data combination is marked as highly sensitive data, an early warning mechanism is triggered, and desensitization and encryption are conducted on the data combination according to a preset desensitization rule and an encryption method; and continuously monitoring data updating conditions of different data sources, when newly added data is detected, automatically triggering a data fusion and leakage risk prediction process, obtaining sensitivity characteristic representation of the newly added data, updating a data leakage risk prediction model, and dynamically adjusting the sensitivity degree and protection measures of data combination.

The invention belongs to the field of meteorological science and technology, and relates to a high-temporal-spatial-resolution three-dimensional water vapor assimilation fusion method based on multi-source data, which is realized based on an LAPS mode and comprises the following steps: acquiring multi-source data of a research area and processing the multi-source data; the multi-source data comprises reanalysis data, ground meteorological data, high-altitude meteorological data, foundation GNSS water vapor data and FY-4 satellite AGRI image data; performing assimilation fusion on the processed FY-4 satellite AGRI image data and water vapor products thereof, ground meteorological data, high-altitude meteorological data and foundation GNSS water vapor data by adopting a multi-grid variation assimilation method and taking reanalysis data as a background field through an LAPS mode; according to different data combinations of high-altitude meteorological data and FY-4 satellite AGRI image data, multiple schemes of three-dimensional water vapor assimilation fusion are formed based on assimilation fusion of multi-source data, an optimal scheme is established, and three-dimensional water vapor data with high temporal-spatial resolution are generated.

Provided are system, method, and device for automatically enhancing future prediction using a reservoir transformer in a machine learning model. According to example embodiments, the system may include: a memory storage storing computer-executable instructions; and at least one processor communicatively coupled to the memory storage, wherein the at least one processor may be configured to execute the instructions to: obtain current input data representing a current state of a complex system; determine a plurality of readout data based on previous input data representing a previous state of the complex system using a plurality of reservoirs; combine the plurality of readout data to form an ensemble reservoir data; and determine predicted output data representing a predicted state of the complex system based on the ensemble reservoir data and the current input data using a transformer.

The invention belongs to the technical field of underground engineering, and discloses an omnibearing intelligent monitoring system, method and device for an underground anchoring pull-out test. According to the method, electromagnetic signals generated in the stress process of the anchor rod are captured in real time through an electromagnetic sensor, and internal stress changes and energy distribution are reflected; the deformation and stress states of the anchor rod in the dynamic environment are evaluated through a strain gauge and accelerometer combined data combination method, the deformation condition of the anchor rod is monitored through the strain gauge, and the stress-strain relation is recorded; an accelerometer is used for detecting the dynamic response of surrounding rock mass so as to evaluate the overall stability of the rock mass; a coupling medium is injected, the computer terminal stores the collected signals locally or in a cloud end and analyzes the signals, stress and deformation parameters in an anchor rod pull-out test are extracted, and an anchor rod stress cloud picture is generated in real time. According to the method, the safety and reliability of underground engineering are remarkably improved, and a new technical support is provided for evaluation of anchoring quality.

The invention discloses a radiation source open set identification method and device based on a dynamic group isovariant network and a generative adversarial network. The method comprises the following steps: preprocessing a plurality of radiation source individual signals of known categories to obtain a data set for closed set identification; a dynamic group isovariant network is constructed, the network is trained on the obtained known category data set, and optimal model parameters are found; simulating virtual unknown category data on the basis of the known category data by using the generative adversarial network, and retraining the constructed recognition model by combining the unknown category data with the known category data to obtain a model capable of being used for open set recognition; and combining actually acquired unknown category data with known category data to serve as a test data set, extracting features through a retrained model, inputting the features into an OpenMax layer, judging whether the category is a known category or an unknown category according to threshold distribution, and detecting the recognition performance of the retrained network open set. The training efficiency is high in small sample recognition task, the classification recognition performance is good, and the robustness is high in open set recognition task.

A radar and camera sensor combined detection method, comprising: obtaining a plurality of image perception data groups and a plurality of raw radar perception data groups of a target object, the target object comprising a moving target object and a stationary target object (S11); according to a first data processing method, processing the plurality of raw radar perception data groups to obtain a plurality of radar sparse tensor information groups (S12); and forming a stereoscopic multi-dimensional depth perception data combination and outputting same, the stereoscopic multi-dimensional depth perception data combination being formed in a multi-dimensional matrix structure, and the stereoscopic multi-dimensional depth perception data combination comprising a plurality of radar sparse tensor information groups assigned to the multi-dimensional matrix structure and a corresponding plurality of image perception data groups (S13). The method solves the problems of how to effectively identify static objects, occluded objects, etc. and process a large amount of radar raw data.

The invention discloses a construction method of a medical auxiliary diagnosis system model based on a multi-modal breast cancer knowledge graph. The method comprises three parts of constructing a multi-modal breast cancer mapping knowledge domain, constructing a breast cancer mapping knowledge domain enhanced diagnosis system module and constructing a medical auxiliary diagnosis system model: 1) constructing the multi-modal breast cancer mapping knowledge domain, including data acquisition and processing, ontology modeling, entity and relation extraction model construction and image data combination; 2) constructing a breast cancer knowledge graph enhanced diagnosis system module, including knowledge query and knowledge screening and processing, and 3) constructing a medical auxiliary diagnosis system model, including data acquisition and processing, text data feature extraction, image data feature extraction, joint representation of learning texts and images, and constructing a breast cancer diagnosis module and a training model.

Certain aspects of the present disclosure provide techniques for wireless communication by a user equipment. The techniques include receiving, from a backscatter device, first backscatter data during one or more network entity downlink slots; receiving, from the backscatter device, during a network entity uplink slot immediately following the one or more network entity downlink slots, second backscatter data; and combining, at the user equipment, the first backscatter data and the second backscatter data to generate a backscatter data package.

The invention provides an object positioning method and device based on wireless devices, and the method comprises the steps: obtaining a plurality of pieces of initial channel state information of a plurality of wireless devices in response to an indoor positioning instruction, and enabling the initial channel state information to represent device data which are collected by the wireless devices and are related to a target object; for each piece of initial channel state information, performing deviation elimination processing on the initial channel state information based on a preset deviation elimination formula to obtain target channel state information; according to the multiple pieces of target channel state information, multiple data combination matrixes are generated, and each data combination matrix comprises multiple matrix row and column values; for each data combination matrix, processing a plurality of matrix row and column values and position coding data of the wireless device corresponding to each matrix row and column value by using a target encoder, and generating a coordinate data set of the target object; and performing weight fusion processing on the plurality of coordinate data sets to obtain target positioning information of the target object.

The invention provides a multi-mode PDF (Portable Document Format) document analysis method, device and equipment and a medium, and the method comprises the following steps: loading a PDF document and carrying out preprocessing, including page splitting and content cleaning, to generate standardized document data; dynamically extracting contents in the standardized document data, wherein the contents comprise paragraphs, pictures and table elements; processing the dynamically extracted pictures, including shielding meaningless pictures based on a preset rule, and analyzing picture contents by using a multi-modal model to generate readable picture information; processing the dynamically extracted table, including optimizing and merging the table structure into a single element format to generate structured table data; combining paragraphs, readable picture information and structured table data, converting the paragraphs, the readable picture information and the structured table data into a complete structured document format, and inserting in key positions to enhance coherence context description; and outputting the complete structured document format as a final analysis result. According to the method, the analysis precision and the knowledge extraction efficiency of the complex PDF document can be remarkably improved.

The present application provides a method and apparatus for batch generation of test cases, and a computing device. The method comprises: acquiring a function to be tested of a target software product; acquiring an action factor and a data factor corresponding to said function; the action factor is used to describe at least one of the following types of information in a test case: a pre-condition, a test step, and an expected result; the data factor comprises at least one test input parameter corresponding to said function, and multiple values of each test input parameter; combining the plurality of values of the at least one test input parameter, and obtaining multiple data combination results; on the basis of the action factor and the multiple data combination results, generating a plurality of corresponding test cases in batches; and displaying the generated multiple test cases to a user. The described method can not only increase the efficiency of generating test cases, but also avoid the problem of missed testing.

The invention relates to the technical field of flight management, in particular to a flight departure guarantee resource allocation scheduling optimization management method, which comprises the following steps: according to a runway, a running sequence and a time interval corresponding to a flight shift list, constructing a flight state diagram for the flight shift list in a preorder and postorder manner; determining the transfer cost of each state in the flight state transfer process by taking the maximum capacity and the minimum adjustment time interval of each node in the flight state diagram in a unit time period as constraint conditions, and sequencing the flights into a time-adjusted flight list according to the transfer cost; performing iterative solution by using the timing flight list, determining an optimal adjustment time interval in unit time, and determining an adjustment quantity in a unit time period according to the adjustment time interval in combination with time slot allocation; when the adjustment quantity accords with a preset constraint rule, performing data combination on the corresponding flights, and regarding the combined data as an output adjustment scheme; the efficiency and the performability of flight scheduling processing are realized.

The invention discloses a hierarchical processing system for energy storage data, and the system comprises an optimization feature extraction module which collects initial energy storage data, builds a unified time sequence on the time dimension of the initial energy storage data through a specified time interval # imgabs0 #, carries out the alignment sampling, introduces the initial energy storage data after the alignment sampling into an auto-encoder, and carries out the noise reduction processing, performing feature extraction on the data subjected to noise reduction processing to obtain optimized energy storage feature data; the dynamic grading adjustment module is used for standardizing the optimized energy storage characteristic data, calculating the information entropy of the standardized characteristic to obtain an initial weight, introducing a Q reinforcement learning algorithm to carry out online optimization on the weight, designing a third-order grading model based on the optimized initial weight, and carrying out dynamic grading adjustment on the optimal energy storage characteristic data; and performing data classification on different optimization energy storage characteristic data combinations and dynamic weights according to a third-order classification model, dividing the data into an emergency level, an important level and a common level based on the third-order classification model, and providing targeted processing and management strategies for different levels of data.

Proposed is a multi-sensor fusion device for performing extrinsic calibration in real time. The multi-sensor fusion device may include a camera unit including at least one camera and configured to produce image information of a surrounding environment by using the camera. The multi-sensor fusion device may also include a light detection and ranging (LiDAR) unit including at least one LiDAR and configured to produce a point cloud of the surrounding environment by using the LiDAR; a data combination unit configured to produce combined data in which the image information is combined with the point cloud. The multi-sensor fusion device may further include a data computation unit configured to derive an extrinsic calibration parameter by inputting the combined data into an extrinsic calibration model and perform extrinsic calibration of the camera and the LiDAR based on the derived extrinsic calibration parameter.

The invention provides a control method of storage equipment, the storage equipment and a computer storage medium. The method comprises the following steps: taking data read under initial threshold voltage as original data; symmetrically shifting the initial threshold voltage by taking the initial threshold voltage as a center, and reading data again as bias data; the mutually symmetrical bias data are logically processed, and judgment data are obtained; the original data and the judgment data are combined, a partition array is obtained, the credibility of the partition array is obtained through a lookup table, and the credibility and the accuracy of the judgment data are in positive correlation; the judgment data with high confidence reliability are preferentially judged, and the judgment data are decoded and processed; when decoding processing is started, the number of times of symmetrical offset initial threshold voltage is increased, and data are read again to serve as bias voltage data of the next round; and the step of obtaining the judgment data, the step of obtaining the credibility and the decoding process are cycled until decoding succeeds or the number of symmetric offset times reaches an offset time threshold value.

The invention discloses an electrical equipment defect detection method, system and equipment based on multi-modal alignment, and particularly relates to the technical field of electrical equipment detection. Obtaining a spatial feature difference and a time feature difference between every two pieces of modal data; performing logistic regression processing on the spatial feature difference and the time feature difference to obtain an alignable coefficient between each two modal data, and determining a to-be-aligned data combination for electrical equipment defect detection based on the alignable coefficients; and performing data alignment on the to-be-aligned data combination to obtain aligned modal data, and performing defect detection analysis on the power equipment by using the time sequence model according to the aligned modal data to obtain a defect detection result of the power equipment.

The invention relates to the technical field of hydrological data processing, in particular to a water conservancy intelligent internet-of-things sensing method and system, and the method comprises the following steps: combining statistical data into a feature set based on hydrological monitoring data including flow, water level and rainfall information, judging whether the data accords with normal distribution, and generating a data distribution feature vector set. According to the method, the change trend of the data is captured through multi-level data distribution characteristic analysis, and the monitoring precision of the data exception is enhanced. In the prediction process, new data and weight adjustment are gradually fused, so that the model can adapt to environmental changes in real time, and the hydrological state in a future short period is accurately predicted. Besides, abnormal data can be identified in real time based on analysis of quantiles, potential problems can be quickly found, and state division of different levels can be carried out according to abnormal degrees. In sampling scheduling, the priority of data acquisition is adjusted according to a probability model of state transition, so that monitoring resources are efficiently configured.

The invention discloses an intelligent supervision system for online data processing and transaction processing services, and belongs to the technical field of transaction processing supervision. Performing periodic monitoring statistics on all transaction processing businesses of the target, and performing periodic transaction state processing analysis and data combination on the target transaction type according to monitoring statistics data to obtain a local transaction processing business supervision sequence corresponding to the target transaction type in the monitoring period; according to the local transaction processing business supervision sequences corresponding to different monitoring periods, transaction eligibility analysis in different aspects is carried out on a target transaction type to obtain transaction eligibility identifiers corresponding to different aspects, and the transaction eligibility identifiers in different aspects are integrated and analyzed to obtain transaction eligibility identifiers corresponding to different aspects. The target transaction marketing scheme associated with the target transaction type is dynamically managed in a self-adaptive manner; the technical problems of poor autonomous supervision analysis processing effect and poor autonomous optimization management effect of the pre-stored transaction processing business in the existing scheme are solved.

The invention provides a foundation pit excavation deformation prediction method and device based on finite elements, and relates to the technical field of foundation pit construction.The method comprises the steps that firstly, a digital geologic model is built based on geological survey drilling data, and a foundation pit excavation finite element numerical simulation model is built in combination with a foundation pit supporting design drawing of a design institute; orthogonal design parameters are combined to obtain multiple groups of simulation data, and a twinborn database is established; constructing an initial prediction model and training by using a twin database to obtain a target prediction model; obtaining foundation pit monitoring data through a sensor, extracting data corresponding to the twin database to form an effective parameter combination, and inputting a target model to predict a foundation pit excavation deformation result in the next stage; and judging whether a foundation is excavated or not according to a deformation result, and if not, continuing to predict until the foundation is reached. According to the method, modeling is performed by combining geology and design data, the model is trained by using a twin database, deformation can be accurately predicted, dynamic monitoring and real-time prediction are realized by combining monitoring data, and construction safety is guaranteed.

The invention provides a method and a device for generating test cases in batches and computing equipment. The method comprises the following steps: acquiring a to-be-tested function of a target software product; an action factor and a data factor corresponding to the to-be-tested function are obtained, the action factor is used for describing at least one of the following information in a test case: a precondition, a test step and an expected result, and the data factor comprises at least one test input parameter corresponding to the to-be-tested function and a plurality of values of each test input parameter; combining the plurality of values of the at least one test input parameter to obtain a plurality of data combination results; generating a plurality of corresponding test cases in batches according to the action factors and the plurality of data combination results; and displaying the plurality of generated test cases to the user. The method not only can improve the generation efficiency of the test case, but also can avoid the problem of missing test.

The invention discloses an individual health risk dynamic assessment system based on multi-modal biological feature fusion, and relates to the technical field of medical health information processing. The method comprises the following steps: constructing a high / low-dimension data combination rule; dynamically adjusting the feature fusion weight according to the clinical feature importance weight and the medical efficiency value calculated by the detection time interval; a cost difference value formula is introduced to quantify the economical efficiency of combined evaluation and independent evaluation; calculating a health risk comprehensive coefficient of the monitoring scene by combining the health data credibility score and the index type / amplitude / duration weighted risk value; the risk level is rechecked through expert assessment or a secondary model, so that the assessment accuracy is ensured; updating the individual credibility score according to the evaluation result; and for the result exceeding the confidence interval, classifying and optimizing a feature fusion algorithm, a model parameter or a data combination strategy. The system realizes accurate, economic and real-time individual health risk dynamic assessment, and is suitable for scenes of chronic disease management, health intervention and the like.

The invention discloses an active and passive safety fusion evaluation weight coefficient determination method and system, and relates to the technical field of automobile safety evaluation, and the method comprises the steps: obtaining active safety data and passive safety data; combining the active and passive data to obtain total sample data; constructing an analytic hierarchy model and a judgment matrix of each hierarchy; calculating a total level sequence of all indexes in the scheme layer to obtain a weight value of each index; performing weighted scoring on each data sample to obtain a total score; constructing a multi-layer perceptron model; and extracting the final weight of each index. According to the method, the vehicle speed after braking of the vehicle active safety system is taken as the collision speed of human-vehicle collision simulation in the passive safety system, and an active safety index and a passive safety index are associated to carry out active and passive safety fusion evaluation; and the index weight coefficient value for fusion evaluation is extracted through an analytic hierarchy process and construction of a multi-layer perceptron model, so that the data accuracy is high, the objectivity is high, and the comprehensiveness is high.