29 results about "Long term data" patented technology

The invention relates to a conveying belt tearing positioning and tracing system based on RFID and multi-sensor fusion, which effectively overcomes the influence of belt slipping and elastic deformation through multi-sensor fusion and motion compensation, and realizes accurate positioning of a tearing point. Meanwhile, through cross validation of vibration, vision and infrared data, the recognition accuracy is improved to 99.5% or above, false alarms caused by environmental factors (such as dust and moisture) are greatly reduced, stable operation in a severe industrial environment is ensured, and high reliability and low false alarm rate are achieved. And thirdly, the system supports long-term data tracing and intelligent operation and maintenance, and a user can analyze a tearing trend and identify a high-frequency fault area by storing historical data of at least 180 days and providing a visual tool, so that a maintenance plan is optimized, predictive maintenance is realized, and faults are reduced from the source.

The invention relates to the field of distributed storage, and discloses a distributed storage system, method and equipment for wind power plant data and a medium, which are used for realizing flexible adjustment of a storage strategy and meeting the storage requirement of dynamic change of a wind power plant data value. Comprising the following steps: collecting SCADA system data at edge nodes of a wind power plant, inputting an information entropy module and an access heat analysis module in parallel, fusing to generate a data value evaluation result, inputting the evaluation result into a cellular automaton engine, outputting a cellular state change instruction, summarizing collaborative strategy requests of at least two edge nodes to a regional center node, and carrying out collaborative strategy analysis on the regional center node. And a fair resource allocation scheme is output through the Sharpley value calculation module. And each edge node synchronously receives the scheme, the time sequence database controller generates a final storage instruction, and data distributed persistent storage is completed. According to the method, on the basis of the correlation between the long-term data value evaluation result and the actual use value, the entropy value and the popularity weight coefficient are optimized, and self-adaptive optimization of the system is achieved.

The application relates to a multi-unmanned aerial vehicle (UAV) assisted edge computing task offloading method and system, and belongs to the technical field of mobile edge computing task offloading. The method comprises the following steps: constructing a multi-UAV assisted edge computing task offloading system model and an optimization problem model; based on the above model, a Markov decision process under a multi-agent environment is constructed, a reward function taking queue stability and information age minimization as targets is defined, a deep deterministic policy gradient algorithm is adopted, the optimization problem is solved through interaction between the multi-agent and the environment, and an optimal offloading strategy of a ground communication device computing task is obtained. The system comprises a ground communication device and a UAV, the UAV is in communication connection with the ground communication device, and an edge computing server arranged on the UAV is used for realizing the above method. The application not only guarantees the stability of a long-term data queue, but also obtains an optimal information age in an online mode, and solves problems such as communication congestion and poor user experience quality.

This invention discloses a daily runoff prediction method based on the Long Short-Term Component Neural Network (LSTCNet) model. LSTCNet leverages the advantages of convolutional layers to extract features. An attention mechanism (AM) is introduced in the long-term component to automatically select relative times across all time steps of the long-term data. The short-term component utilizes a multi-layer residual structure to fuse multi-level features from the short-term data, effectively improving model performance. Furthermore, a traditional AR model is used as a linear neural network component to enhance the learning of linear dependencies between variables and modify the output prediction values. Experimental results show that the proposed method has good performance in daily runoff prediction; the long-term and short-term components can effectively learn from their respective data, demonstrating the strength of the ensemble model.

This invention relates to the fields of smart city and machine learning technology, specifically a method for continuous self-optimization of urban governance models based on data closed-loop feedback. The scheme involves: acquiring raw waveform data and operating condition reference signals from vibration sensors of urban lifeline facilities; generating a time-frequency energy spectrum matrix through time-frequency transformation; decomposing this matrix into slow-changing and fast-changing components using a machine learning model; removing slow-changing components correlated with long-term operating conditions; labeling the remaining slow-changing components as the first component and the fast-changing components as the second component; generating a frequency-related compensation curve based on the difference between the first component and the historical energy spectrum; performing multiplicative correction; and simultaneously calculating a coupling health index, triggering a calibration work order and updating the historical energy spectrum when the index falls below a threshold. This invention achieves closed-loop optimization of sensor coupling state self-sensing, self-compensation, and self-calibration, significantly improving the long-term data stability and accuracy of urban lifeline monitoring systems.

An AoI optimization method based on a space-air-ground integrated network relates to the technical field of unmanned aerial vehicle communication, and comprises the following steps: 1, dividing a ground sensing area into a plurality of sub-areas configured with an unmanned aerial vehicle, dividing a long-term data collection period into a plurality of equilong basic time slots, and each basic time slot comprises a long time slot and a short time slot which are connected in time sequence; and 2, constructing an AoI optimization model of an air-ground link under a single time slot, and achieving the goal of minimizing the long-term average AoI of ground nodes by optimizing node clustering, energy supply and data transmission time distribution, cluster scheduling and unmanned aerial vehicle trajectory planning. And 3, constructing an AoI optimization model of an air-space link under a single time slot, and realizing the target of reducing the cache backlog of the unmanned aerial vehicle by optimizing an access strategy of the unmanned aerial vehicle and a low-orbit satellite and a transmission bandwidth allocation strategy. According to the invention, the technical problems of limited energy supply of the ground sensing network, high system AoI optimization complexity and difficult network resource scheduling in the prior art can be solved.

This invention discloses a postoperative rehabilitation training and health management system and method for breast cancer patients, belonging to the field of smart healthcare and rehabilitation training technology. The system includes a local smart terminal, a cloud server, and a medical staff terminal. The local terminal integrates a millimeter-wave radar sensor, which generates upper limb point cloud data by transmitting and processing echo signals. A three-dimensional human skeleton model is reconstructed using a deep learning network, and motion parameters such as joint angles and movement trajectories are extracted. The system is then compared with a standard template to calculate the motion completion score and provides real-time corrective feedback. The cloud server deploys a finely tuned large language model, which, combined with a rehabilitation knowledge base, performs enhanced retrieval and generation, providing personalized suggestions based on the patient's rehabilitation history and consultation questions. The rehabilitation plan generation module dynamically adjusts the training plan based on long-term data analysis. This invention achieves a deep integration of non-contact precise perception, real-time motion assessment, and intelligent health management.

The invention discloses a wind power plant extreme disaster risk pre-judgment method and device based on extreme value distribution, and the method comprises the steps: collecting the long-term extreme disaster data of more than or equal to 30 years, extracting an annual extreme value sequence, and carrying out the abnormal value processing through the combination of Grubbs detection and meteorological record verification; and selecting a Gunbel or generalized extreme value distribution model according to the disaster type, fitting parameters by using a maximum likelihood method, calculating an extreme value through distribution inversion, comparing the extreme value with an engineering design threshold, and outputting a risk conclusion and a protection suggestion. The method is based on long-term data and probability modeling, the accuracy and scientificity of extreme value prediction are improved, the problem of evaluation deviation caused by short data, rough model and separation from engineering in a traditional method is solved, and the method has good engineering application value.

Techniques of storing and retrieving medical data and information are provided. Medical result information can be stored on a long-term data repository in a shared network, e.g., the Internet. Pseudonymized identifiers of patients can be used to retrieve such data and information.

A threat management facility for an enterprise network provides visualization tools for threat analysis and investigation. While security events may generally be logged in a long term data repository such as a data lake, security events can be transmitted directly to a short term, higher performance data repository for faster visualization when fast response times might be necessary or helpful. In this context, the threat management facility may use time stamps associated with event reporting to select a time-indexed segment of the data lake as a target for supplemental, investigative queries.

The invention relates to a method (100) for providing long-term data collection, comprising the following, which is performed by a base station (10) of a communication network (1): - Sending (101) an information message (300) regarding the availability of a long-term data collection function, which provides long-term data collection via a dedicated interface to user equipment (20) in the communication network (1), wherein the information message (300) includes an indication of the availability of the long-term data collection function, - Receiving (102) from the user equipment (20) a user request based on the sent (101) information, wherein the user request includes the indication that references the availability of the long-term service function, - Providing (103) the long-term data collection of the long-term service function at the dedicated interface based on the received (102) user request.

This invention relates to the field of distributed storage, and discloses a distributed storage system, method, device, and medium for wind farm data. It enables flexible adjustment of storage strategies to meet the storage needs of dynamically changing wind farm data value. The system includes collecting SCADA system data at wind farm edge nodes, inputting the data into an information entropy module and an access heat analysis module in parallel, fusing the data to generate a data value assessment result, inputting the assessment result into a cellular automaton engine, outputting cellular state change instructions, aggregating collaborative strategy requests from at least two edge nodes to a regional central node, and outputting a fair resource allocation scheme via a Shapley value calculation module. Each edge node synchronously receives the scheme, and a time-series database controller generates the final storage instruction, completing the distributed persistent storage of the data. This invention optimizes the entropy value and heat weight coefficient based on the correlation between long-term data value assessment results and actual use value, achieving adaptive system optimization.

The invention provides a lake ecological flow processing method and a lake ecological flow processing device based on a remote sensing and water balance model, which are used for establishing a set of lake ecological flow standardization processing architecture based on a remote sensing and water balance equation, effectively solving the limitation that the traditional method seriously depends on actually measured hydrological data, and reducing the calculation cost. Meanwhile, compared with the characteristic that long-term data accumulation is needed in a traditional method, the timeliness is remarkably enhanced, so that the application requirements of low cost and rapid evaluation can be met, and good data support is provided for lake water resource development and utilization management and control and lake ecological management work.

A data hierarchical storage system of a vehicle remote monitoring system is provided with a hot data layer, adopts distributed storage, and is used for storing nearly real-time data, holding high-concurrency write-in and low-delay query, and ensuring real-time monitoring and quick response; a temperature data layer is arranged, column type storage is adopted, the temperature data layer is used for storing medium-term data, and query performance and storage cost are balanced through compression and partitioning technologies; a cold data layer is arranged, object storage or a distributed file system is adopted, and the cold data layer is used for storing long-term archived data and supports batch access and offline analysis. Through the arrangement of the hot data layer, the warm data layer and the cold data layer, the overall storage cost is remarkably reduced, the situation that all data occupy high-performance storage resources for a long time is avoided, the SSD consumption problem is solved, meanwhile, the data storage and analysis process is simplified through the mode, data redundant replication is avoided, and the data management efficiency is improved; and finally, data lake construction of vehicle enterprises can be supported, and unified data management and long-term data value mining are realized.

The application provides a lake ecological flow processing method and device based on remote sensing and water balance model, which is used for establishing a lake ecological flow standardized processing architecture based on remote sensing and water balance equation, effectively solving the limitation that the traditional method seriously depends on measured hydrological data, and reducing the calculation cost. At the same time, compared with the characteristics that the traditional method needs long-term data accumulation, the timeliness is also significantly enhanced, so as to meet the application demand of low-cost and rapid evaluation, and provide good data support for lake water resource development and utilization control and lake ecological management.

The invention belongs to the field of artificial intelligence, and relates to a risk trend data pushing method and device, equipment and a medium, and the method comprises the steps: obtaining crop multi-source data from a multi-source database, carrying out the coupling with an initial meteorological prediction model, and constructing a meteorological risk assessment model. Acquiring monitoring images and meteorological data of crops in the target plot, and determining the current growth period of the crops. And outputting risk degree indexes and risk driving factor analysis results through evaluation of the meteorological risk evaluation model by using the growth period data and the meteorological data. And correcting the full-period risk threshold according to the growth period to obtain a target risk threshold. And obtaining meteorological prediction data of a target time period, and combining the above results to perform prediction by using a time sequence prediction model to obtain risk trend data. And extracting user information of the target plot from the crop multi-source data, and pushing the risk trend data to a corresponding user terminal. The method can be applied to the business fields of financial insurance, agricultural management and the like, and the accuracy of crop risk trend data pushing can be improved.

The invention discloses a phytoplankton in-situ flow imaging device and an imaging method, and belongs to the technical field of marine ecology and water quality monitoring, and the phytoplankton in-situ flow imaging device comprises an optical imaging module, a fluid control module, a data processing module and a communication module which are fixedly arranged in a sealed cavity of an outer shell. According to the in-situ detection device, optical triggering is not needed, the structure is simple, full-fluid microscopic images can be continuously collected, automatic monitoring and high-throughput quantitative analysis of phytoplankton and suspended particles are achieved, the remote interaction operation capacity is achieved, and reliable long-term data support is provided for marine ecological monitoring.

The invention discloses a multi-theory fusion-based career planning large model construction method and system, and relates to the technical field of data mining and career planning, and the method comprises the specific steps: firstly integrating multiple theories to construct a career development heterogeneous causal graph, and defining an element relationship; constructing a dynamic heterogeneous graph neural network and training a large model by taking the dynamic heterogeneous graph as input; establishing a channel to collect control intervention variable data; reasoning and visually explaining the career path; and finally, providing a double-deployment mode, utilizing a dynamic incremental updating algorithm and combining with multi-party data to realize full-process linkage optimization of the model, and providing precise learning career planning for the user. According to the method, a structured heterogeneous causal graph is constructed, a career element association rule is defined, interference is weakened in combination with a related mechanism, accurate reasoning is guaranteed by means of long-term data training, and traditional problems are solved; and a multivariate data acquisition channel is also established, and a double deployment mode and a dynamic increment updating algorithm are adopted, so that static limitation is broken through.

The invention relates to the technical field of data processing, in particular to a fault detection method and system for an underwater granulator. The method comprises the following steps: acquiring preprocessed multi-dimensional equipment operation data; for any data point, constructing a working condition change index reflecting the change of the working condition at the moment of the data point; in response to the condition change index being greater than a preset change threshold, adjusting an initial distance between the current data point and other data points in the long-term data window based on the condition change index, and obtaining an optimal distance; and based on the optimal distance, carrying out anomaly detection on the data points by adopting a local outlier factor algorithm. The underwater granulator detection method has the effect of reducing the false alarm condition in the detection process of the underwater granulator.

Security events are reported to a threat management facility for an enterprise network as self-contained lineages that include data concerning related processes such as a parent or child process related to the source of the event. By transmitting these to a short term data store, threat timeline visualizations can be more quickly rendered for an analyst in a user interface, after which the visualization can be augmented with other data from other sources such as a data lake or other long term data repository for the enterprise network, third party reputation sources, and so forth.

The invention relates to the technical field of data processing, and particularly provides a data filling management method, system and device and a storage medium, and the method comprises the steps: obtaining a new version draft and a current effective version of a dynamic business index library in response to the version updating operation of the dynamic business index library; comparing the new version draft with the current effective version, and identifying a changed index; based on the changed indexes, the influence range of the changed indexes in the data filling management system is analyzed, and the influence range at least comprises associated filling tasks and early warning rules; and according to the influence range, performing linkage updating on a corresponding module in the data filling management system, the linkage updating at least including updating an index definition in an associated filling task and updating an associated early warning rule. According to the method, the problem of data aperture confusion caused by rule change is avoided, and the self-adaptive capability and the long-term data management level of the system are remarkably enhanced.

Methods, systems, and devices for data management are described. The initiation or forthcoming initiation of a data protection operation for a computing system may be identified. Short-term information of the computing system, including information stored in the volatile memory of the computing system, network traffic associated with the computing system, or both, may be obtained based on the initiation or forthcoming initiation of the data protection operation. Long-term information of the computing system, including information stored in the non-volatile memory of the computing system, may be obtained based on the data protection operation being initiated. Both the short-term information and the long-term information may be stored for further analysis.

The application relates to a long-term and short-term data backup and storage method and device based on a disk CDP, and belongs to the field of disaster recovery backup. The method comprises the following steps: creating and generating data mirror images, bitmaps, I / O logs and metadata under a current directory, then recording real-time change data by using the files, modifying the directory name of the current directory, deleting the I / O log, obtaining T0 directory and files, then repeating the above steps, and obtaining T1 to TN directories and files according to subsequent triggered generation of incremental data instructions, and rolling back long-term data through the data mirror images under the T0 to TN directories. The device comprises a short-term data directory generation module, a short-term data recording module, a long-term data generation module, a subsequent long-term data generation module and a real-time data storage module. The application realizes rolling back long-term and short-term data through a set of systems, the strategy configuration is unified, the storage cost is low, and the application provides one-stop and low-TCO data protection for high-compliance and high-continuity scenes.

The application discloses a traffic flow prediction method and system based on an STNR model, which comprises the following steps: sorting historical traffic flow data of a target road section in time sequence, dividing the historical traffic flow data into long-term data and short-term data, and obtaining a training set; constructing a traffic flow prediction model, taking the long-term data and the short-term data as inputs of the traffic flow prediction model, and training the traffic flow prediction model; the traffic flow prediction model comprises a full connection layer, a long short-term processing module, a space-time convolution module and a multilayer perceptron module; collecting traffic flow data of the target road section at a current time, inputting the traffic flow data into the trained traffic flow prediction model, and predicting traffic flow of a future period. The application effectively captures the spatial dependence and the time dependence of the traffic flow, and improves the prediction accuracy.

The application relates to a kind of deep learning-based internet of things perception data prediction method, belonging to the technical field of internet of things.The method comprises: obtaining internet of things perception long-term data and pre-processing, according to the characteristics of internet of things data with seasonality, using STL method to decompose time series data, and extracting periodicity, trend and residual characteristics.Improve the whale algorithm from the perspective of location update strategy and prevention from falling into local optimum, and use the improved whale optimization algorithm to optimize the hyperparameters of the bidirectional long short-term memory network model;Use the optimized BiLSTM model to predict internet of things perception data.The application can efficiently predict the future trend of internet of things perception data, and is suitable for traffic monitoring, intelligent environment perception and other scenes, and provides reliable decision support for internet of things applications.

The invention discloses a river eutrophication prediction method considering climate and landscape pattern changes, and relates to the technical field of river management, and the method comprises the steps: collecting and verifying long-term data in a watershed, carrying out the preprocessing and space matching, and forming a standardized sub-watershed annual data set; based on the sub-basin year data set, extracting landscape indexes of sub-basin scales; calculating a river eutrophication index according to the water quality monitoring data; screening key landscape indexes by using a recursive feature elimination algorithm; constructing a machine learning regression model by taking the key landscape index and the climate data as input variables and the river eutrophication index as an output variable, and obtaining an optimal prediction model; carrying out interpretability analysis, and quantifying the contribution degree and the influence direction of each input feature to the river eutrophication index; and inputting data in different future scenes into the optimal prediction model, and predicting and evaluating river eutrophication risks in different scenes.

The invention belongs to the technical field of port task load prediction, and particularly discloses a port task load long-short time fusion prediction method and terminal equipment, and the method comprises the steps: obtaining historical operation data of a port, the historical operation data comprising influence factor data and task load; performing sparse sampling and dense sampling on the historical operation data to correspondingly obtain a long-term data set and a short-term data set, training a long-term prediction model by using the long-term data set, and training a short-term prediction model based on the long-term prediction model and the short-term data set; respectively inputting the real-time influence factor data of the port into the trained long-term prediction model and short-term prediction model to obtain a long-term prediction task load and a short-term prediction task load; and determining the fusion prediction task load of the port according to the long-term prediction task load and the short-term prediction task load. According to the method, short-term prediction drifting can be avoided in combination with long-term trend constraint, the matching degree of the prediction result and the actual production requirement is improved by 15%-20%, and the error is reduced by 5%-8%.

The application provides a kind of based on satellite internet's marine ship data transmission scheduling method and device. The method assesses the initial weight of ship sensor data queue according to data delay and bandwidth requirement, and dynamically adjusts the weight based on abnormal data detection and abnormal target detection method;Design a multi-layer buffer mechanism, define the validity period of data packet, and design a packet loss decision mechanism based on relaxation decision diagram;According to satellite ephemeris data, predict the satellite service window time, define the potential output performance function of data, and model the ship multi-sensor data queue scheduling process as a hierarchical Markov decision process;Design a ship multi-sensor data queue scheduling algorithm based on improved D3QN, which prevents queue overflow, ensures queue transmission fairness, and maximizes long-term data output performance within the service window time. The application effectively solves the problem that the effectiveness and fairness of real-time data transmission cannot be guaranteed under the condition of frequent switching of satellite-sea link.

The invention relates to the technical field of health monitoring, in particular to a diet monitoring method and system based on multi-terminal cooperation, electronic equipment and a medium. The invention provides a diet monitoring method based on multi-terminal cooperation. A diet monitoring system based on multi-terminal cooperation is composed of an earphone, a terminal and a cloud. The earphone end collects vibration data, screens out noise and then sends the vibration data to the terminal in real time; the terminal extracts behavior characteristics, identifies a diet mode and analyzes the diet mode to generate a monitoring result, if a preset condition is met, an intervention instruction is generated and sent to the earphone for intervention, and meanwhile multi-node data is uploaded to the cloud; and the cloud performs tracking analysis based on the long-term data, generates a personalized guidance strategy and returns the personalized guidance strategy to the terminal for pushing and displaying. According to the architecture, organic combination of edge perception, real-time interaction and deep analysis is realized through hierarchical cooperation, and balance is achieved in three aspects of guaranteeing real-time performance, improving monitoring precision and effective behavior analysis.