566 results about "Data allocation" patented technology

The invention relates to a matrix storage operator optimization method and device, computer equipment and a readable storage medium. The method comprises the following steps: allocating memory resources for target data; determining a logic structure corresponding to the source data, establishing a first mapping relation between each logic index in the logic structure and a register address of the source data, and establishing a second mapping relation between each logic index in the logic structure and a memory address of the target data; and determining a corresponding relationship between a memory address of the target data and a register address of the source data based on the first mapping relationship and the second mapping relationship, and storing the source data stored in a register into a memory resource corresponding to the target data based on the corresponding relationship. By adopting the method, the generalization ability of the matrix storage operator can be improved.

The invention relates to the technical field of inventory management, in particular to an intelligent warehouse management system for main shaft parts based on digital twinning, which comprises a main shaft mapping module for acquiring a main shaft shipping space and an allocation path, comparing shipping space nodes with a moving timestamp difference to generate an identification result, the fluctuation identification module extracts channel nodes to analyze fluctuation threshold deviation and generate a report, the allocation control module generates a scheduling list based on queuing duration and shipping space data distribution weight, and the abnormity positioning module extracts unallocated parts to analyze interface difference and mark abnormal links. And the state early warning module integrates four indexes of abnormal data identification risk paths to generate a management scheme. According to the method, bin position nodes and main shaft movement timestamps are synchronously compared to identify difference nodes, a dynamic weight distribution mechanism is established in combination with channel node fluctuation analysis, an interruption link difference trend and interface delay data are integrated to construct a risk assessment model, and abnormal conduction path positioning and early warning are realized. And scheduling conflicts and resource waste caused by data deviation are reduced.

The invention relates to the technical field of direct current motor servo drive control, and provides a Zynq AXI4 bus-based real-time transmission system and a control method thereof, and the system comprises a Zynq system on chip, a DDR3 common data area, an AXI4 bus interconnection architecture and a dynamic partition manager. By implementing the scheme, firstly, the multi-channel characteristic of the AXI4 bus is utilized, special transmission channels are distributed for different types of data, and parallel data interaction is realized; secondly, a dynamic partition manager is deployed at a PL end, a DDR3 public data area is dynamically divided into a plurality of partitions in real time according to needs, multi-task parallel is achieved, and the real-time performance of the system is improved; thirdly, the read-write operation on the DDR3 in each control period is split into a plurality of state machines through precise scheduling of the state machines, the state machines are combined into a state loop to be mounted to an AXI4 bus, it is ensured that key tasks are completed on time, and the certainty of the system is improved; fourthly, the size and the number of partitions in the DDR3 are dynamically adjusted through a dynamic partition manager, and the storage efficiency is improved.

The invention provides a wireless sensing platform with an edge computing function, and relates to the technical field of wireless sensors. In order to solve the problems that in the prior art, a wireless sensor is large in data redundancy, high in transmission delay and weak in emergency recognition and response capacity in a complex scene, a joint processing mechanism combining a causal event recognition assembly, a causal data fusion assembly, an event separation assembly and a data distribution assembly is adopted. By detecting a causal change event in an environment and dynamically extracting and fusing sensor data with a strong causal relationship, redundant data transmission caused by emergencies is reduced, transmission delay is reduced, the robustness of causal data fusion in a concurrent event scene is improved, and the reliability of data fusion is improved. According to the invention, efficient and accurate sensing and response to emergencies in a complex environment are realized.

The invention relates to the technical field of digital communication, in particular to an intelligent electric meter dual-mode communication self-adaptive switching method, and provides the following scheme that operation data are divided into a plurality of data units at a sending end, priority evaluation is carried out, and the time delay, the packet loss, the energy consumption and the bandwidth state of two links are combined, so that the intelligent electric meter dual-mode communication self-adaptive switching method is realized. Positioning clearing cut points in the priority sequence so as to realize differentiated distribution of data between a high-speed link and a low-speed link; when the data is distributed to a low-speed link, sparse coding is carried out by adopting a preset dictionary, and differential updating is triggered when an error exceeds a limit, so that decoding consistency is ensured; a receiving end realizes unified analysis of the original message and the sparse message through cross-link association, dictionary updating and residual synthesis; according to the invention, the timeliness of the key message can be guaranteed, the communication energy consumption and cost can be reduced, and the cooperation efficiency of the dual-mode link can be improved.

The invention discloses a method for aggregating ocean buoy data, which comprises the following steps of: establishing a space-time correlation model according to the spatial distribution density of buoy clusters and the dynamic characteristics of an ocean environment, and adaptively dividing buoy groups with a data compression cooperative relationship; performing a data-related compression operation within the buoy group; constructing a transmission priority dynamic sorting engine, and calculating a dynamic weight coefficient; performing hierarchical packaging processing on the differential coding data, distributing enhanced forward error correction coding resources for high-priority data, and reserving redundant transmission time slots; the method comprises the following steps: receiving a hierarchical data packet through a low-orbit satellite and executing on-satellite preprocessing, reconstructing an original data stream by using a distributed decoder, and meanwhile, fusing multi-dimensional space-time related data to generate an aggregated data product; and finally, the aggregated data product is transmitted to a ground data center through a satellite downlink. According to the invention, uplink data volume is reduced, disaster monitoring timeliness is improved, ground station processing load is reduced, and system environment adaptability is enhanced.

The invention relates to a map visualization rendering method and system based on a WebGPU engine and computer equipment, and the method comprises the steps: obtaining map data in a unified format; inputting the map data into a rendering layer, and performing layer division on the map data according to spatial levels and element types to obtain multiple pieces of layer data; distributing an independent rendering thread for each layer data; performing block processing on the map data to obtain picture blocks with pyramid structures; and calling a corresponding rendering thread in the current view range and the scaling to load the corresponding picture block for rendering display. By means of the method and device, the problem that the map visualization rendering performance is poor is solved.

Disclosed herein are methods and systems that use artificial intelligence techniques for determining an allocation of an asset based, at least in part, on analyzing electronic images of the asset. In an embodiment, an AI model can be executed on a received data packet comprising a set of electronic images to generate a set of attributes. In an embodiment, the AI model can identify a segment of the asset associated with the electronic image, generate an attribute comprising a description of quality of the segment of the asset, and determine whether the electronic image is associated a subset of the electronic images. In an embodiment, the set of attributes and an indication of the subset with user input about the asset can be aggregated to generate evaluation data. In an embodiment, a computer model can be executed to determine an allocation of the asset to a class.

The invention relates to the technical field of data encryption transmission, in particular to an electric power communication resource data encryption method, system and equipment based on multi-algorithm fusion, and the method comprises the steps: dividing data into a core control class, an equipment state class and a resource information class according to the purpose of electric power communication resources, a security label containing data sensitivity, transmission timeliness and storage cycle information is distributed to each type of data; an initial quantum key is generated by adopting a quantum key distribution protocol, an asymmetric key pair is generated in combination with an SM2 algorithm, and the quantum key is signed by using an SM2 private key and then transmitted; the dynamic secret key update is triggered based on the timestamp or the data volume threshold value; encrypting core control type data, dynamically generating confusion layer parameters, and meanwhile, generating a message digest in combination with an SM3 hash algorithm; and the receiver decrypts the confusion layer parameter, decrypts to obtain the plaintext, and verifies the SM3 hash value to confirm the data integrity. And the limitations of a single algorithm on efficiency, strength and aggressiveness resistance are made up.

The invention discloses an energy storage power station operation and maintenance management system based on big data, and relates to the technical field of power station operation and maintenance management, and the system comprises an operation and maintenance platform which is in communication connection with a multi-source data processing module, a health state evaluation module, a fault analysis module and an optimization module. The multi-source data processing module is used for collecting energy storage associated data of a plurality of dimension types in the energy storage power station, allocating a time sequence operation window for the energy storage associated data of each dimension type for time sequence allocation association, performing feature quantization on the energy storage associated data under each time sequence to obtain a time sequence feature coefficient under the corresponding time sequence, and outputting the time sequence feature coefficient to the energy storage power station; summarizing into a time sequence feature set; the health state evaluation module is used for evaluating the energy storage state of the energy storage power station according to the time sequence feature set; the fault analysis module is used for carrying out fault analysis on the energy storage power stations of which the energy storage states do not reach the standard to obtain corresponding energy storage fault detail information; and the optimization module is used for establishing a multi-target constraint optimization model and completing energy storage optimization of the energy storage power station.

The invention discloses a data backup method and device, equipment, a medium and a computer program product, and relates to the technical field of data processing.The method comprises the steps that when a continuous protection snapshot is created for a source volume, the current backup generation of the source volume before the snapshot is created is set as a snapshot generation identifier; and synchronizing data by relying on mapping relation metadata including the current backup generation of the source volume, the last backup generation of the source volume, the storage backup data generation of the target volume and the water level field. Reading corresponding generation data of the source volume according to the water level mark in the first backup, writing into the target volume, and increasing the water level to the total number of source volume data blocks after completion; during incremental backup, a new snapshot is created, the water level is reset, only adjacent generations of difference data are synchronized, and a corresponding generation identifier is distributed for target volume data in each backup. In this way, no extra bitmap is needed to participate in difference marking, the synchronization efficiency is improved, the target volume keeps historical backup data of all generations in the synchronization period, and the problem that the target volume cannot effectively protect data in the traditional incremental backup synchronization stage is solved.

The invention provides a security optimization encryption method, system and device for data in a facility agriculture scene and a medium, and belongs to the technical field of agricultural informatization security, and the method comprises the steps: carrying out the dynamic division of security levels of data collected in a facility agriculture environment according to the data source device type, data content sensitivity and real-time requirements; distributing an encryption algorithm for the collected data according to a mapping relation between the security level and the encryption algorithm; secret keys are generated and managed based on an identity encryption mechanism, and then secret key fragmentation storage is carried out in combination with a trusted execution environment; when encryption is needed, whether the current trusted execution environment is safe or not is judged, and when the current trusted execution environment is safe, secret keys of the fragments are recombined, the multi-source heterogeneous data are encrypted through the distributed encryption algorithm, and encrypted data are generated; and carrying out secure transmission on an adaptive communication protocol after packaging the encrypted data. According to the method, the data security level is dynamically divided, the encryption algorithm is flexibly selected, the resource consumption is reduced, and the real-time performance and security are improved.

The invention relates to the technical field of storage system IO performance optimization, and provides a redirection data aggregation method and equipment, and the method comprises the steps: recognizing random small IO data from a written IO request, aggregating and writing back-end storage objects, and constructing a storage mapping relation in an index table; traversing the rear-end storage object in which the random small IO data is stored, and distributing data from the same logic address space to point to the same target rear-end storage object through a hash algorithm; writing the data distributed to point to the same target back-end storage object into the continuous physical space of the target back-end storage object in sequence, and updating the mapping relation in the index table; the step S2 and the step S3 are repeated before a preset termination triggering condition is met, and a hash algorithm different from all previous times is adopted when the step S2 is repeatedly executed each time; when the data are read, the storage mapping relation is obtained from the index table, and the read requests corresponding to the data located in the same back-end storage object are aggregated and then read at a time.

The invention provides a data transaction method and device, equipment, a medium and a program product, relates to the technical field of block chains, and aims to solve the problem of relatively poor efficiency of data transaction in related technologies. The method comprises the following steps: based on data demand information, creating a demand task pool by using an intelligent contract, and broadcasting the data demand information to each node in a block chain system; receiving at least one data abstract and guarantee fund submitted by the at least one data owner node through the smart contract, and writing the at least one data abstract into a demand task pool; under the condition that the first data demander node determines m first data owner nodes, price negotiation is carried out based on m price parameters submitted by the m first data owner nodes and the smart contract, and a transaction result is determined; and under the condition that the first data demander node pays the resource corresponding to the transaction price, distributing the first target data to the first data demander node. According to the invention, the data transaction efficiency can be improved.

The invention discloses a power time sequence prediction data set construction method and system, and the method comprises the steps: obtaining power load and weather data, forming a power data set, and carrying out the cleaning and alignment; meanwhile, multi-source public time series data such as energy and weather are obtained; uniformly formatting and normalizing the two types of data, extracting time-frequency features, calculating similarity between sequences, and screening out public data significantly related to the power load; a dynamic sampling weighting method is adopted to allocate a weight for each piece of data, a sampling frequency is set, and finally a heterogeneous power time sequence data set used for pre-training and fine tuning of a large model is constructed; according to the method, by integrating the multi-source heterogeneous data, constructing a unified standardized data format and fusing the multi-source data based on timestamp alignment, the consistency and availability of the data are remarkably improved, and the data fusion efficiency and the model generalization ability are improved.

The application discloses a kind of electric power time sequence prediction dataset construction method and system, the method includes obtaining electric power load and weather data, constitutes electric power dataset and carries out cleaning alignment;While obtaining energy, meteorological and other multi-source public time sequence data;Two kinds of data are uniformly formatted and normalized, extract time-frequency characteristics and calculate similarity between sequences, filter out the public data significantly related to electric power load;Dynamic sampling weighting method is used to assign weight to each data, set sampling frequency, finally build heterogeneous electric power time sequence dataset for large model pre-training and fine-tuning;The application integrates multi-source heterogeneous data, constructs a unified standardized data format, and aligns and fuses multiple sources of data based on timestamp, significantly improves the consistency and availability of data, improves data fusion efficiency and model generalization ability.

The embodiment of the invention provides a distributed data deduplication method and product. The method comprises the steps of obtaining a data key of data to be subjected to deduplication; distributing the data to be subjected to duplicate removal to the ith sub-bucket according to the mixed hash value of the data key; mapping the mixed hash value according to a hash function attribute value configured for the ith sub-bucket to obtain k position numbers; reading numerical values of positions corresponding to the k position numbers from a bit array which is stored at the current moment and corresponds to the ith sub-bucket to obtain k numbers; if it is confirmed that the k numbers are all first numerical values, it is confirmed that the data to be subjected to duplicate removal belong to duplicate data; and if at least one of the k digits is confirmed to be a second numerical value, confirming that the data to be subjected to duplicate removal belongs to new data. According to the embodiment of the invention, the speed and accuracy of big data deduplication processing are remarkably improved, and the problem of single-node calculation overload (different buckets can be distributed to different calculation nodes) is effectively solved.

The invention discloses a multi-scale adversarial decoupling near infrared spectrum model transfer method. The method comprises the following steps: acquiring near infrared spectrum data sets of a source domain and a target domain, executing a spatio-temporal data distribution strategy in time and space dimensions, carrying out proportion division on the spectrum data sets, and carrying out near infrared spectrum model pre-training; respectively extracting instrument irrelevant features and instrument relevant features through a neural network, performing adversarial training to minimize domain difference loss, and outputting a domain invariant subset; inputting the related characteristics of the instrument, the near infrared spectrum data set of the target domain and the domain invariant subset into a generator to generate virtual spectrum data; and adopting a dynamic weighted sampling strategy, adaptively adjusting the sampling probability of the virtual spectrum data and the real spectrum data, performing model training fine tuning, deploying the fine-tuned model to a target instrument, and realizing cross-instrument transfer of the near infrared spectrum model. According to the method, the cross-instrument generalization ability of the model is greatly improved, and the robustness of the spectrum model in the space-time dimension is enhanced.

Various embodiments are provided for providing enhanced data allocation for machine learning operations in a computing environment by one or more processors in a computing system. One or more data sampling strategies may be determined based on a dataset. One or more enhanced training data allocations may be suggested for machine learning operations in a cloud computing environment based on the one or more data sampling strategies.

The invention discloses a deep learning-based satellite remote sensing collapsible loess foundation evaluation method, which comprises the following steps of: acquiring an optical remote sensing image, a radar remote sensing image and an infrared remote sensing image of a target area, and preprocessing; collapsibility feature weighted data are screened, and feature weights are distributed; feature extraction is carried out through a local space exhibition structure and a multi-layer Transform structure of the collapsibility foundation evaluation network; collapsibility area self-supervised collaborative learning is carried out, and self-supervised training is carried out based on spatial correlation, pseudo labels and spatial consistency regular terms; and carrying out risk grade division on the target area, and outputting a collapsibility risk result of each spatial position. According to the method, multi-mode remote sensing and deep learning are fused, high-precision intelligent partitioning of the collapsible loess foundation is achieved, and the method has the advantages of being self-adaptive, low in manpower and high in spatial resolution.

The invention discloses an intelligent monitoring, early warning and operation and maintenance management method for a photovoltaic system. The method comprises the following steps: S1, dynamic fusion and edge collaborative acquisition of multi-source heterogeneous data: constructing a full-dimensional data acquisition system covering a photovoltaic module, an inverter, a combiner box, a meteorological environment and a power grid side; a framework of'edge preprocessing + cloud fine fusion 'is adopted, edge nodes realize second-level data synchronization through a lightweight time sequence alignment algorithm, and sensor noise is eliminated by adopting an improved density-based abnormal value filtering algorithm; and the cloud end allocates dynamic weights to data of different frequencies and different types based on a multi-modal data fusion model of an attention mechanism, so that high-precision data fusion is realized. By constructing a full-dimensional data acquisition system and an innovative fusion framework, the problem of data fragmentation of traditional photovoltaic monitoring is solved, the data fusion precision is improved, and a high-quality data basis is provided for subsequent analysis and diagnosis.

The invention provides an AI-driven heterogeneous data optimization management method and system, and the method comprises the steps: extracting the production dynamics, equipment association and data demand dimensions in a flexible production scene through AI modeling, constructing a scene knowledge graph, carrying out the analysis of data from a heterogeneous terminal, and converting the data into a standard format matched with a current production scene; according to a multi-dimensional association relationship in the scene knowledge graph, performing exception recognition, deletion repair or redundancy elimination on the data in the standardized format to realize quality optimization, allocating storage resources for the data subjected to quality optimization, and pre-loading associated data before nodes are produced; according to the method, decision information for production scheduling, equipment predictive maintenance and product quality tracing is output in combination with the incidence relation among equipment, processes and data defined in the scene knowledge graph, and the timeliness and accuracy of core services such as production scheduling, equipment maintenance and quality tracing are improved by constructing and updating the scene knowledge graph in real time.

The invention belongs to the technical field of computers, and discloses a heterogeneous SoC task adaptive distribution system. Comprises: a migration determination module for identifying a load imbalance state between cores in a running process and screening migration candidate tasks; the migration preparation module is used for generating baseline snapshots for the running states of the migration candidate tasks and carrying out hierarchical division on the baseline snapshots; the semantic annotation module is used for carrying out semantic annotation on memory pages occupied by the migration candidate tasks, calculating regularized semantic scores and distributing levels to memory page data according to the regularized semantic scores; the compression migration module is used for carrying out mixed compression and self-adaptive reduction on the layered state data in the migration process; the preheating loading module is used for executing preheating loading of codes and data of a high-priority function on a high-speed cache of the target core according to the preheating priority sequence on the migrated target core; and through semantic annotation and hierarchical compression, the transmission data volume is effectively reduced, and the migration dwell time is shortened.

A system and method for robotic teleoperation enable a teleoperated robotic element to perform a sequence of actions based on intention estimation, eliminating the need for continuous human control. The system includes a robotic teleoperation input that receives motion inputs and gaze data from a human operator performing a robotic teleoperation task. A robotic teleoperation feature extractor analyzes and processes the motion inputs and the gaze data into sequential input data. A multi-window model assigns hierarchical prediction windows to the input data, generating windowed sequential data. A hierarchical neural network processes the windowed sequential data to determine low-level action intentions and high-level task intentions, and generates an intention estimation based on the low-level action intentions and the high-level task intentions. A hierarchical dependency model incorporates hierarchical dependent loss to refine the intention estimation.

The invention provides a data cold and hot hierarchical storage and data intelligent scheduling method and device. The method comprises the following steps: firstly, constructing a multi-dimensional cold and hot evaluation model, collecting and quantifying evaluation dimension data related to at least two types of services, dynamically allocating weights in combination with a scene, and calculating a comprehensive score to judge cold and hot attributes of the data; distributing the data to storage systems with corresponding characteristics according to cold and hot attributes and deletion and modification requirements; then, an intelligent scheduling center containing multiple modules is built, and automatic data circulation, real-time monitoring, resource regulation and control and full-life-cycle tracing are achieved; and finally, establishing an intelligent report, integrating data to form a subject data set, and automatically completing report query, calculation and the like. The method can achieve the full-life-cycle automatic management of the data, improves the management accuracy and efficiency, and is suitable for the fields with high data requirements, such as finance.

The invention relates to the technical field of artificial intelligence hardware, and provides an artificial intelligence chip and an integrated device.The artificial intelligence chip comprises a computing bare chip, a first storage control unit, a second storage control unit and a first memory; the first storage control unit is arranged in the computing bare chip, and the first storage control unit is connected with the first memory; the second storage control unit is arranged in the computing bare chip or connected with the computing bare chip, the second storage control unit is used for being connected with a second memory, and the second memory is arranged outside the artificial intelligence chip; in the event that the computing die initiates a unified memory access, the data allocation of the first memory and the second memory is determined based on the bandwidth of the first memory and the bandwidth of the second memory. According to the chip and the device provided by the invention, the heterogeneous memory system design of the artificial intelligence chip is realized, the high bandwidth advantage of the first memory and the high capacity advantage of the second memory are both played, and dual optimization of bandwidth and capacity is realized for the heterogeneous memory system of the artificial intelligence chip.

The invention discloses a data processing method and device applied to edge computing equipment. The method comprises the following steps: accessing a plurality of different models and algorithms through model driving middleware; acquiring at least one path of data and caching the data to a pre-processing queue; performing standardization preprocessing on the multi-modal data by calling a standardization interface; according to the configuration information, distributing the preprocessed data to a corresponding model queue, obtaining a reasoning result through model reasoning, and sending the reasoning result to a post-processing queue for caching; or sending the preprocessed data to a post-processing queue for caching; reasoning results or data are taken out from the post-processing queue, and whether alarm needs to be triggered or not is determined through algorithm calculation, so that an alarm message is generated. According to the method, the difference in multiple aspects of each third-party model is shielded, the workload and the debugging period of developing adaptive codes are reduced, the problems of data transmission disorder and format mismatching during multi-model cooperation are solved through a standard data flow model architecture, and the stability of the system architecture is improved.

The invention relates to the technical field of petroleum and natural gas inspection, in particular to an unmanned aerial vehicle data processing method and system for petroleum and natural gas inspection key indexes. The method comprises the following steps: dividing petroleum and natural gas inspection subareas, and dynamically adjusting inspection priorities of the inspection subareas according to acquired data; dynamically adjusting the current inspection route based on the inspection priority, cooperatively scheduling multiple unmanned aerial vehicles, synchronously updating the inspection state, and uploading inspection data in real time; distributing corresponding priority queues for the inspection data according to the inspection priorities for data processing, judging the abnormal levels of the inspection sub-regions in combination with oil and gas key indexes, and performing graded early warning; the system comprises a regional inspection priority adjustment module, a task dynamic planning module and a data processing priority adaptation module. The inspection priority of the inspection sub-regions is dynamically adjusted, the multiple unmanned aerial vehicles are cooperatively scheduled to upload the inspection data in real time, and early warning is performed according to the abnormal levels of the inspection sub-regions, so that the risk troubleshooting efficiency is improved.

The invention discloses a multi-step time sequence prediction method and system based on double-segmentation conformal prediction. The method comprises the following steps: acquiring new input data; distributing the new input data to a corresponding clustering cluster to obtain a clustering result; extracting information from the recorder according to a clustering result to construct a prediction interval; adjusting the corresponding content according to the prediction interval; wherein a similar trend sequence is vertically classified and clustered, errors of adjacent time steps are horizontally and dynamically combined to optimize window division, over-estimation and under-estimation errors are asymmetrically processed to construct a precise confidence interval, the precise confidence interval is stored in the recorder, and an error set is dynamically updated. By implementing the method provided by the invention, the defects in the prior art can be overcome through a two-dimensional segmentation mechanism, more accurate uncertainty quantization is realized, and the adaptability and accuracy of multi-step time sequence prediction are improved.

The invention provides a Kubernetes-based large model reasoning optimization method, system and equipment, and the method comprises the steps: constructing a resource scheduling assembly for a large model, and monitoring a real-time resource state at a current moment; constructing a batch processing agent component for the large model, performing feature analysis on the received reasoning request queue, and determining request features corresponding to the reasoning requests in the reasoning request queue; the real-time resource state is evaluated according to the request features, and the batch processing size is adjusted based on the evaluation result to generate the optimal batch processing length; batching the reasoning requests in the reasoning request queue according to the optimal batch processing length, and packaging each batch of reasoning requests into batch data; the resource scheduling component is utilized to distribute batch data to Pod in the K8s cluster for reasoning, a large model reasoning result is obtained, the quantity of the reasoning Pod is automatically adjusted according to the flow and the resource load required by reasoning, a GPU with poor performance is prevented from becoming a system bottleneck, and the responsiveness and the stability of service are improved.