Patents
Literature
1475 results about "Raw data" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Raw data, also known as primary data, is data (e.g., numbers, instrument readings, figures, etc.) collected from a source. If a scientist sets up a computerized thermometer which records the temperature of a chemical mixture in a test tube every minute, the list of temperature readings for every minute, as printed out on a spreadsheet or viewed on a computer screen is "raw data". Raw data has not been subjected to processing, "cleaning" by researchers to remove outliers, obvious instrument reading errors or data entry errors, or any analysis (e.g., determining central tendency aspects such as the average or median result). As well, raw data has not been subject to any other manipulation by a software program or a human researcher, analyst or technician. It is also referred to as primary data. Raw data is a relative term (see data), because even once raw data has been "cleaned" and processed by one team of researchers, another team may consider this processed data to be "raw data" for another stage of research. Raw data can be inputted to a computer program or used in manual procedures such as analyzing statistics from a survey. The term "raw data" can refer to the binary data on electronic storage devices, such as hard disk drives (also referred to as "low-level data").
System and method for orchestration of multi-agent operations using language models
In a described embodiment, a multi-agent system for processing information is provided including a data processing agent configured to ingest and normalize raw data inputs to produce standardized data and a standards integration agent configured to apply reporting standards into the standardized data thereby generating integrated reporting standards. The system further includes a performance alignment agent configured to align performance indicators based on the standardized data and the integrated reporting standards and an information synthesis agent configured to process narrative information from the standardized data and the integrated reporting standards. An orchestration framework configured to manage operations of the data processing agent, the standards integration agent, and the performance alignment agent to produce a regulatory repot compliant with regulatory requirements is further provided. The orchestration framework is further executable by a large language model.
Owner:STANDARD CHARTERED BANK SINGAPORE BRANCH
Data analysis pipeline engine in a data intelligence system
PendingUS20260004135A1Data processing applicationsPlatform integrity maintainanceData sourceEngineering
Methods, systems, and computer storage media for providing a data analysis pipeline using a data analysis pipeline engine in a data intelligence system are described. A data analysis pipeline refers to a structured sequence of data processing steps that support transforming raw data into meaningful insights or actionable outcomes. The data analysis pipeline engine is an unsupervised learning pipeline based on clustering, topic modeling, and Large Language Models (LLMs). For example, the data analysis pipeline can use advanced machine learning techniques to automatically categorize emails into semantically similar clusters, enabling the data intelligence system to quickly identify and prioritize potentially high-risk emails for further investigation. The data analysis pipeline employs AI agents for context-aware graph induction relevance assessment. The AI agents employ induction and deduction loops to build and refine a data feature hypergraph (e.g., vulnerability hypergraph) that encompasses identified relevant data providing a holistic view of a contextual landscape.
Owner:MICROSOFT TECHNOLOGY LICENSING LLC
Multi-parameter intelligent sensing and state monitoring system for power transformation equipment
The invention relates to the technical field of power transformation equipment state monitoring, in particular to a power transformation equipment multi-parameter intelligent sensing and state monitoring system which comprises a sensing unit, a data processing unit, a state analysis and diagnosis unit and an upper computer monitoring and management unit. Through data collection preprocessing, lightweight AI model anomaly preliminary screening and hierarchical edge cloud cooperative transmission strategies, efficient cleaning of original data, rapid edge end anomaly identification and optimal utilization of network resources are realized, data transmission bandwidth occupation is greatly reduced, monitoring real-time performance is improved, and the method is suitable for large-scale popularization and application. The method integrates technologies such as digital twinning, federated learning and a time-space attention network, realizes equipment cross-time-space fault accurate positioning, fault type reliable identification and residual life dynamic prediction in combination with a quantification algorithm, triggers hierarchical early warning through hierarchical health assessment, and provides scientific and accurate decision support for refined operation and maintenance of power transformation equipment.
Owner:STATE GRID ZHEJIANG ELECTRIC POWER CO LTD NINGBO POWER SUPPLY CO +1
Large language model federated fine-tuning method and apparatus based on gradient compression
PCT designated stageWO2025260391A1Biological modelsComputer security arrangementsData privacy protectionData set
Disclosed in the present invention are a large language model federated fine-tuning method and apparatus based on gradient compression. The method comprises the following steps: constructing, on the basis of a gradient tensor generated during fine tuning of a large language model, a raw data set having a time series relationship, performing inference by means of an autoencoder to obtain a reconstructed gradient data set, and constructing a reconstruction loss function to optimize the autoencoder; and initializing a base model of the large language model as a global model at a server end, the server end updating the global model to a client, using a pre-trained encoder to obtain a compressed gradient at the client, and at the server end, using a pre-trained decoder to decode and aggregate the compressed gradient, and then updating the global model. The present invention can improve the fine-tuning efficiency of the large language model and reduce computing resource requirements while ensuring data privacy protection, and is suitable for application scenarios such as communication optimization improvement and privacy protection enhancement in the process of scientific computing-oriented large model fine-tuning and training.
Owner:ZHEJIANG LAB
System and method for data-driven decision optimization for autonomous driving
PendingUS20260084708A1Internal combustion piston enginesExternal condition input parametersEngineeringData-driven
A data-driven autonomous driving decision optimization system and method, comprising: a data production module, a data screening module, a model encapsulation module, and a parameter tuning module; the data production module takes human driving data as input, annotation, preprocessing, format conversion, extracts key features and performs standardization, normalization, and encoding to generate raw data for the data-driven process that meets the algorithm input requirements; the data screening module screens corresponding data from training data by the decision-making module and performs effective classification; the model encapsulation module encapsulates C++ decision code and constructs a trajectory-pair evaluation cost map required for training using a ground-truth evaluation method based on trajectory pairs; the parameter tuning module, based on screened data under different scenarios, the encapsulated decision algorithm model, and the trajectory-pair evaluation cost map, employs black-box optimization to obtain decision parameters for the corresponding scenarios under the current decision algorithm.
Owner:SHANGHAI JIAOTONG UNIV
Robot track online compensation and precision control method and system based on multiple sensors
The invention discloses a robot track online compensation and precision control method and system based on multiple sensors, and belongs to the technical field of robot track control. The method comprises the following steps: acquiring target trajectory data and original data of each sensor, and performing time alignment on the original data of each sensor to generate a fusion data set comprising vibration acceleration data and pose data; performing time-frequency feature extraction based on the vibration acceleration data to obtain a dominant vibration frequency spectrum; constructing a mechanical transfer function model of the joint driving torque and the tail end vibration displacement; based on the mechanical transfer function model, constructing an adaptive notch filter to generate a feed-forward compensation sequence; according to the target trajectory data, the pose data and weight parameters, updated in real time, of the sensors, a trajectory fusion deviation value is obtained through calculation; and after the feedforward compensation sequence and the trajectory fusion deviation value are integrated, a joint driving current instruction is generated based on a fuzzy PID control algorithm. The robot trajectory control precision can be improved.
Owner:SUZHOU UNIV
Federated Codebook Optimization and Neural Upsampler Training for Distributed Device Networks
A federated system and method for data compression optimization in distributed device networks. The system comprises multiple edge devices that analyze local data patterns to generate device characteristic profiles while performing local compression optimization and maintaining data privacy. Edge devices contribute to collaborative learning by generating privacy-preserved updates without transmitting raw data. A central coordination system aggregates encrypted contributions using secure multi-party computation protocols, identifies device groups based on data pattern similarities, and generates optimized compression parameters for each group. The system coordinates collaborative training of data reconstruction models across device groups and deploys group-optimized reconstruction capabilities. Device grouping is performed by calculating similarity scores between device characteristic profiles and clustering devices with scores above predetermined thresholds. The system dynamically adapts compression and reconstruction parameters through federated learning while preserving individual device data privacy, enabling efficient data compression and near-lossless recovery across heterogeneous Internet-of-Things networks.
Owner:ATOMBEAM TECH INC
Energy storage cabinet multi-parameter intelligent monitoring device based on MEMS sensor array
The invention relates to the technical field of energy storage cabinet safety monitoring, in particular to an energy storage cabinet multi-parameter intelligent monitoring device based on an MEMS sensor array, a sensor array unit in the energy storage cabinet multi-parameter intelligent monitoring device comprises distributed MEMS temperature and vibration sensors, the temperature sensors are installed in a non-vertical mode, detection deviation caused by direct blowing of airflow is avoided, and the energy storage cabinet multi-parameter intelligent monitoring device based on the MEMS sensor array is obtained. The airflow disturbance correction unit corrects temperature original data in real time and eliminates interference of airflow on module body temperature detection by means of a heat dissipation airflow vector field dynamic model and a compensation algorithm, and the multi-parameter coupling analysis unit processes vibration and temperature signals in a double-channel mode, extracts tab loosening high-frequency micro-vibration characteristics and correlates the temperature rising rate. Early fault logic is triggered through two-stage judgment, the fault decision and output unit fuses the correction temperature, the coupling strength coefficient and the speed-up ratio, graded early warning is generated through a fault tree model and transmitted to an external monitoring system, and accurate monitoring of early faults of the energy storage cabinet is achieved.
Owner:WEICHU NEW MATERIALS CO LTD
Bid invitation file generation method and system and storage medium
The invention provides a bid invitation file generation method and system and a storage medium, and the method comprises the steps: obtaining original data from a plurality of modalities, the original data comprising a project demand text, a design model file, a law and regulation text and historical bid invitation case data; performing multi-module semantic deep analysis and structured processing on the original data, and respectively generating a project demand sub-graph, a spatial logic sub-graph, a structured regulation knowledge base and a historical case strategy feature library; performing cross-modal dynamic fusion on the project demand sub-graph and the spatial logic sub-graph to construct a unified knowledge graph, and performing dynamic knowledge tracing and strategy injection in the fusion process based on a structured regulation knowledge base and a historical case strategy feature library; and generating a first bid invitation file draft based on the unified knowledge graph, performing compliance verification on the first bid invitation file draft, and determining the first bid invitation file draft passing the verification as a final bid invitation file. The bid invitation file generation accuracy can be improved.
Owner:HANGZHOU GOLDEN SOFTWARE SYST INC
Photovoltaic grid-connected box electric energy quality monitoring system based on Internet of Things data acquisition
The invention relates to the technical field of intelligent monitoring and fault diagnosis of a photovoltaic power generation system, in particular to a photovoltaic grid-connected box electric energy quality monitoring system based on data acquisition of the Internet of Things. Comprising an environment perception and time domain synchronization unit which is used for triggering sampling by using a unified clock source and generating a full-dimensional synchronous original data stream; the ideal reference physical reconstruction unit is used for calculating a theoretical output amplitude envelope and generating an ideal state reference vector; the fault mechanism parameterization evolution unit is used for converting the fault mechanism into a fault operator, selecting a high-risk fault mode according to the current operation condition, and superposing the corresponding fault operator onto the ideal state reference vector to generate a theoretical damaged state waveform; the double-rail-difference manifold extraction unit is used for obtaining a real residual vector and a theoretical fault residual vector; and the isomorphic feature coupling judgment unit is used for calculating the similarity between the real residual vector and the theoretical fault residual vector. According to the invention, the robustness of power quality monitoring under complex meteorological conditions is significantly improved.
Owner:SICHUAN ABA JINCHUAN HUADIAN NEW ENERGY CO LTD
Virtual power plant optimization control system based on source network load storage cooperation
PendingCN121308079AAc network load balancingAc network voltage adjustmentData acquisitionMulti source data
The invention relates to the technical field of virtual power plants, and discloses a virtual power plant optimization control system based on source-grid-load-storage collaboration, which comprehensively acquires and preprocesses wind and light output, load, weather and power grid operation data through a multi-source data acquisition module, provides accurate original data for error analysis, and improves the accuracy of the system. The prediction error space-time correlation analysis module mines space correlation and time self-correlation characteristics of errors and generates an error scene set containing correlation characteristics, the source load scene library module combines with the error scene set to update and form a dynamic scene library containing risks, and the robust optimization scheduling module carries out robust optimization scheduling on the basis of the dynamic scene library. A scheduling scheme is solved by taking economy and robustness as targets, and a real-time scheduling execution module is linked with a system monitoring feedback module, so that error perceptibility, scene dynamic updating and scheduling adaptive adjustment are integrally realized, the ability of a virtual power plant to deal with uncertainty is improved, the economy and power grid stability are guaranteed, and the method is suitable for large-scale popularization and application. And the reliability and competitiveness of system operation are enhanced.
Owner:INNER MONGOLIA ELECTRIC POWER (GRP) CO LTD TRAINING CENT
Multi-source data fusion method based on AGV cooperative positioning
The invention relates to the field of data processing, in particular to a multi-source data fusion method based on AGV cooperative positioning, and the method comprises the steps: carrying out the collection and preprocessing of AGV positioning original data through multi-source sensor fusion, and obtaining a standardized feature value and a prediction parameter for cooperative updating; carrying out noise factor optimization adjustment by fusing the physical propagation characteristics and the ranging fluctuation information to obtain a UWB measurement noise variance subjected to preliminary dynamic correction; deviation risk identification is carried out through risk factor adjustment based on innovation consistency inspection, and a final dynamically corrected UWB measurement noise variance is obtained; constructing a variance matrix through the dynamic variance and executing filtering updating to obtain corrected AGV state estimation and variance; by optimizing a state estimation result and recursively inputting a prediction process, dynamic updating of multi-AGV cooperative positioning is realized, so that the stability and the overall performance of AGV positioning in a dynamic environment are enhanced.
Owner:ZHEJIANG YIQIAO SOFTWARE DEV CO LTD +1
Power system data integrity verification system based on dynamic data block division mechanism and block chain
The invention discloses a power system data integrity verification system based on a dynamic data block division mechanism and a block chain. The system comprises a data preprocessing module, a hash generation module, a block chain storage module and a verification processing module. The data preprocessing module divides original data into data blocks with dynamically adjustable sizes; the Hash generation module adopts an iterative Hash algorithm to generate a chain Hash structure and an integrity Hash abstract; the block chain storage module uploads the hash data to a block chain network through a smart contract; and the verification processing module outputs a verification result by recalculating and comparing the hash values. The system realizes efficient and reliable data integrity verification, and has the advantages of high storage efficiency, good verification precision and low system load.
Owner:ELECTRIC POWER SCI & RES INST OF STATE GRID TIANJIN ELECTRIC POWER CO +2
Fire scene spreading time sequence reconstruction method and system based on Doppler weather radar mountain fire echo
The invention relates to the technical field of forest fire monitoring and power grid safety prevention and control, and discloses a fire scene spreading time sequence reconstruction method based on Doppler weather radar mountain fire echoes, which comprises the following steps: acquiring original data of the Doppler weather radar mountain fire echoes and carrying out quality control processing to obtain radar echo data; performing automatic segmentation of a smoke plume area on radar echo data by adopting an optimized Otsu-Unet network model, and extracting a centroid coordinate of a mountain fire echo area; a Kalman filtering algorithm is used and combined with external three-dimensional wind field data to predict a future motion trend of centroid coordinates of a forest fire echo region so as to reconstruct a continuous fire scene spreading time sequence voxel sequence; performing 3DTiles slicing processing on the fire scene spreading time sequence voxel sequence, and performing real-time rendering display on a power grid monitoring platform by using a WebGL technology; and obtaining an external expansion boundary of fire scene spreading based on the fire scene spreading time sequence voxel sequence, and calculating the distance between the external expansion boundary and the power transmission line GIS data in real time. The fire prevention and control capability of the power transmission line of the power grid is improved.
Owner:STATE GRID SICHUAN ELECTRIC POWER CORP ELECTRIC POWER RES INST
AI model intelligent training and reasoning integrated method and system
The invention provides an AI model intelligent training and reasoning integration method and system, and the method comprises the following steps: receiving a model training instruction, and carrying out the preprocessing of original data, and obtaining a training data set; executing model training based on the training data set, monitoring task priorities and resource requirements through a dynamic resource scheduling algorithm, and dynamically adjusting training resource allocation according to a real-time monitoring result; when the model reaches a preset performance index, performing model pruning and quantification to generate an optimization model, and performing parameter fine tuning on the optimization model to obtain a final deployment model; generating reasoning service configuration according to calculation characteristics of the final deployment model, migrating the model and the configuration to a reasoning environment, and starting reasoning service; and dynamically adjusting the number of reasoning nodes according to the real-time network flow of the reasoning service. By implementing the technical scheme provided by the invention, through bidirectional dynamic resource scheduling and model deep optimization, the computing resource utilization rate and the model deployment efficiency are improved, and the high availability of the reasoning service is guaranteed.
Owner:BEIJING HIZHI TECH CO LTD
Practical training evaluation method and system based on multi-modal data fusion
PendingCN121707797ACosmonautic condition simulationsData processing applicationsEngineeringMachine learning
The invention discloses a practical training evaluation method and system based on multi-modal data fusion. The method comprises the following steps: firstly, constructing a practical training task topology based on a knowledge graph, mapping elements such as knowledge points into nodes, and defining a relationship through directed edges; in response to task starting, dynamically deploying a multi-modal data fusion agent pre-loaded with a targeted model and a rule; in the operation process, the agent synchronously collects and processes original data from the machine vision and sensor network in real time, and a standardized feature flow is generated; then, on-line space-time correlation and reasoning are carried out on the multi-modal features based on a fusion strategy, and a task scene understanding model is dynamically constructed and maintained; after the task is completed, packaging a program code, a report and process abstract data exported by the model to form an enhanced practical training result package; and finally, multi-dimensional automatic comparison is carried out by calling the rule base and the case model, and an evaluation result is generated. According to the invention, intelligent perception and comprehensive evaluation of the whole practical training operation process are realized.
Owner:YAZHENG TECH GRP CO LTD
Industrial measurement data intelligent analysis report generation method and system based on large language model
PendingCN121902818AImprove utilization depthstable supportSemantic analysisBiological modelsLinguistic modelEngineering
The invention discloses an industrial measurement data intelligent analysis report generation method and system based on a large language model, and relates to the technical field of industrial measurement, and the method comprises a multi-source data and abstraction module which carries out the structural processing of original data DAT output by measurement analysis software; and the large language model service and interface module is used for submitting the PCT to a selected large language model LLM through a multi-model adaptation interface to execute semantic reasoning and output a structured text TXT. Structured analysis and semantic association modeling are carried out on original measurement data through multi-source data and an abstract module, when overall alignment deviation or local feature anomaly exists in the original data, a hierarchical dependency relationship between the data can be established through a semantic graph structure, a deviation transmission path is revealed, and the accuracy of the measurement data is improved. The large language model is analyzed in a unified data context, and the utilization depth of measurement data can be effectively improved, so that the accuracy and integrity of report analysis are improved.
Owner:NANJING YUNTONG TECH CO LTD
Cableway monitoring management method and system based on artificial intelligence and edge calculation
The invention relates to the technical field of cableway fault monitoring, solves the technical problems that in the prior art, the difference between a fault prediction result and the reality is large, nonlinear data is difficult to predict accurately, and the prediction result is inaccurate, and particularly relates to a cableway monitoring management method and system based on artificial intelligence and edge calculation. The method comprises the following steps: S1, obtaining original data of a cableway monitoring system, preprocessing the original data to obtain alignment data, capturing dynamic features through time sequence dependence, providing feature importance interpretation, organically combining time sequence prediction and feature contribution by a weighted fusion mechanism to generate health indexes, and obtaining health indexes of the cableway monitoring system; and the continuous health state can be mapped into a quantifiable fault probability, and finally real-time early warning is realized through threshold triggering, so that the interpretability of a model output result is improved, independent optimization and fault traceability are supported, and the prediction precision and the reliability of operation and maintenance decision are remarkably improved.
Owner:GAODE (TAIAN) IND SERVICES CO LTD
Universal Ambient AI Neural Field for Buildings (UANF)
A building-integrated artificial intelligence system forming a continuous ambient neural field is disclosed. The system includes a distributed multimodal sensor lattice, an on-premise symbolic cognition engine, and an adaptive environmental control kernel operating entirely at the building edge without reliance on external cloud services. Sensor data from optical, thermal, acoustic, airflow, pressure, structural, electrical, and chemical modalities are transformed into non-identifying occupancy vectors, behavioral glyphs, risk indicators, and environmental state descriptors. A privacy-governed policy graph determines sensor permissions, redaction thresholds, consent conditions, emergency overrides, and jurisdiction-specific compliance parameters. The neural field predicts occupancy loads, optimizes HVAC, ventilation, and lighting, detects accidents and structural anomalies, classifies emergent risks, and generates redacted event capsules for audit and emergency dispatch. A federated topology enables multiple buildings to exchange compressed symbolic templates to improve predictive accuracy without transmitting raw data. The system provides a universal, regulation-aligned AI nervous system for autonomous building operations.
Owner:ODEH SAMUEL
Multi-modal semantic communication method and device oriented to multiple tasks
The invention belongs to the technical field of wireless communication and artificial intelligence, and discloses a multi-mode semantic communication method and device oriented to multiple tasks, which avoid low-layer signal coding transmission of original data through extraction and transmission of semantic features, realize end-to-end optimization and channel adaptive robust transmission of semantic level communication, and improve the communication efficiency. The bandwidth occupation is obviously reduced; and the anti-noise performance is improved. Semantic transmission and task execution of multi-source data such as images, texts and voices under a unified framework are achieved through a shared neural network encoder and a channel interface, the robustness of tasks is kept under the complex channel condition, and cross-domain semantic expression among modes such as vision, texts and voices is achieved through the framework. Radio frequency data are sent and received in a real hardware platform, physical combination of an algorithm and a radio frequency link is achieved, and implementation of deep semantic communication in an actual wireless link shows that the method has engineering deployability and expansibility.
Owner:WUHAN EASTAR TECH CO LTD
System And Method For Operating A Device With A Reduced Data Set
A method and system for operating a device includes generating a first set of raw data, reducing the first set of data using a plurality of reductions to obtain reduced sets of data and a performance factor for each reduced data set to determine an inflection point relative to the performance factor, determining a second data set reduced from the set of raw data based on a reduction from the plurality of compressions at or below the inflection point, and controlling the device based on the second data set.
Owner:BOARD OF TRUSTEES OPERATING MICHIGAN STATE UNIV
Bridge health monitoring system and method and storage medium
The invention relates to the technical field of bridge health monitoring, and discloses a bridge health monitoring system and method and a storage medium. According to the method, a multi-modal monitoring original data flow of a bridge is collected through a distributed sensing network, intelligent data purification and heterogeneous data integration are performed on data, and a bridge health data set in a unified format is generated. And decoding a time sequence mark, a spatial position code and a monitoring parameter dimension identifier from the data set. And dividing the data into overlapped time slices according to the time sequence marks, and generating a time slice data set. And fusing spatial position codes, mapping spatial information in the time slice data set to a bridge global coordinate framework, and establishing a dynamic health state map. And based on the atlas and the time slice data set, performing multi-dimensional data fusion according to the monitoring parameter dimension identifier. And analyzing a state evolution path of the bridge member through a fusion result, and generating a health measurement and prediction sequence.
Owner:SICHUAN UNIV JINCHENG INST
Project risk dynamic prediction method and device, electronic equipment and storage medium
The invention provides a project risk dynamic prediction method and device, electronic equipment and a storage medium, and the method comprises the steps: collecting multi-dimensional original data in a whole life cycle of a target project, carrying out the preprocessing of the multi-dimensional original data to generate a standardized data set, building a case decision model based on the standardized data set, and carrying out the prediction of a project risk. And inputting real-time project data into the case decision model to obtain a prediction result, finally performing risk assessment according to the prediction result, and triggering early warning when the risk assessment result exceeds a risk threshold value, so that the problems of poor data quality and low risk caused by the fact that only single-dimensional data collection can be realized and data lacks standardized processing in the prior art can be solved. The problems of low project risk prediction precision and early warning lagging caused by a static evaluation mode are solved, and the technical effects of improving project risk prediction accuracy and realizing project full life cycle risk management real-time performance and effectiveness are achieved.
Owner:JINAN INSPUR DATA TECH CO LTD
Intelligent cutting production line control method and system
The invention provides an intelligent cutting production line control method and system, and the method comprises the steps: firstly obtaining cutting original data, constructing a cutting defect digital diagram based on the cutting original data, then calling a preset cutting technology knowledge base, constructing a cutting negative resource pool through employing a negative resource construction strategy, and then obtaining cutting production plan data, meanwhile, asynchronous clipping task data is obtained based on the clipping negative resource pool, a dynamic interpenetration agent is trained and constructed, the asynchronous clipping task data is imported into the dynamic interpenetration agent, an asynchronous task scheduling scheme is obtained, and finally a real-time clipping path is obtained based on the asynchronous task scheduling scheme and clipping process data. The cutting action is executed based on the real-time cutting path, the cutting method which can utilize negative resources and can perform cutting task on the small parts in the flaw area and the main parts of the main task in an asynchronous and interpenetrating mode is provided, and the overall production efficiency and the material utilization rate are improved.
Owner:SOUTH CHINA OUTDOORS FACTORY LTD
Multi-modal data acquisition and intelligent analysis application method based on AI glasses
PendingCN121785465AInput/output for user-computer interactionKnowledge representationData acquisitionEyewear
The invention provides a multi-modal data acquisition and intelligent analysis application method based on AI glasses. The method comprises the steps that the AI glasses collect text data through a camera and an OCR assembly, collect image video data through a binocular camera and an infrared sensor, collect audio data through an array microphone, and collect auxiliary data through an I MU and an eye movement tracking sensor; according to the AI glasses, a local lightweight AI model is used for basic analysis, and complex tasks are uploaded to a cloud large model for cooperative processing, including information summarization, deep mining and knowledge extraction; the AI glasses realize real-time pushing, scene assistance and content generation and output through an information application module according to an analysis result; according to the AI glasses, an end-cloud collaboration module and a federated learning framework are adopted, a local cloud processing strategy is dynamically adjusted, only feature parameters are uploaded, and it is ensured that original data are locally reserved.
Owner:HUPENG (CHENGDU) ARTIFICIAL INTELLIGENCE TECHNOLOGY CO LTD
Bridge maintenance decision-making method based on big data
PendingCN121882603AData processing applicationsDesign optimisation/simulationBridge engineeringKnowledge graph
The invention relates to the technical field of bridge engineering maintenance decision, and discloses a bridge maintenance decision method based on big data. The method comprises the steps of collecting bridge monitoring data through a sensor network, performing hierarchical processing on an original data stream, identifying and packaging abnormal fluctuation, trend change and a periodic mode, and forming a structured health event sequence. And dynamically constructing and updating the digital sign profile of the bridge in combination with the design parameters and the material degradation model. And when a decision is triggered, simulating a performance evolution path under multi-factor coupling based on the profile, generating a future maintenance scene set, and matching and assembling a preliminary plan from historical cases and the knowledge graph. And outputting a final maintenance instruction list sorted according to priorities by performing resource and time coupling analysis and multi-target tradeoff on all plans. According to the method, deep analysis and prospective simulation of the bridge state are realized, and the intelligence, accuracy and performability of maintenance decision making are improved.
Owner:JIANGSU LIANXU HIGHWAY +1
Multi-agent layered collaborative optimization method based on knowledge graph and graph neural network
The invention discloses a multi-agent layered collaborative optimization method based on a knowledge graph and a graph neural network, and relates to the technical field of machine learning. According to the method, power-on self-test is carried out after software integrity verification, then multi-modal data is preprocessed to convert heterogeneous original data into a unified structured event object, and whether the corresponding candidate entities are divided into the entity candidate index table or not is judged based on the obtained confidence coefficient of the candidate entities; the method comprises the following steps: constructing a causal knowledge graph according to a verification result of a causal relationship of a to-be-verified causal edge, then constructing a space-time heterogeneous graph according to an embedding result of the causal knowledge graph, carrying out hierarchical collaborative optimization processing according to an output node representation matrix, and finally carrying out data twinborn simulation evaluation based on an optimal strategy parameter set. And the optimal strategy parameter set is screened according to the simulation evaluation result, so that the effectiveness of multi-agent hierarchical collaborative optimization is improved, and the problem of low effectiveness of multi-agent hierarchical collaborative optimization in the prior art is solved.
Owner:湖南工商大学
Trip detection method for leakage protection switch of electric energy metering box
The invention discloses an electric energy metering box leakage protection switch trip detection method, which comprises the following steps of: when power failure is detected, generating event context data under the constraint of limited energy based on multi-mode original data, execution energy and buffer combined arrangement; executing short window early warning analysis to obtain an early warning trigger instruction and an early warning feature set; and responding to the early warning trigger instruction, constructing a transient fingerprint and executing parallel crosstalk decoupling to obtain a transient fingerprint and decoupling conclusion. The fingerprint construction adopts logarithmic time scattering fingerprint and electric arc physical fitting, and the decoupling adopts geometric constraint sparse chromatography decoupling to identify a responsibility loop. And fusing the transient fingerprint, the decoupling conclusion and the early warning feature set, applying a multi-modal time sequence consistency constraint, and forming a tripping conclusion and a leakage protection type suggestion. According to the method, the technical problems of loss of evidence at the moment of tripping power failure and strong crosstalk of parallel loops are solved, the responsibility loop can be positioned with high confidence, and interpretable tripping mechanism diagnosis and type suggestions are provided.
Owner:SHAANXI ZHONGHAO ELECTRIC GRP CO LTD
Knowledge distillation-based multivariable measurement sensor state lightweight evaluation method
The invention discloses a multi-variable measurement sensor state lightweight evaluation method based on knowledge distillation, and belongs to the technical field of electric digital data processing and multi-sensor data fusion. The method comprises the following steps: firstly, carrying out time synchronization and physical consistency constraint modeling on original data of multiple sensors, and extracting feature representation; a high-precision teacher model is trained at the cloud end, and the high-dimensional mapping relation of the sensor state is learned; intermediate features, soft output and uncertainty information of a teacher model are extracted to serve as distillation knowledge, lightweight student model training is guided, and effective migration of discrimination knowledge is achieved in combination with soft label constraint, feature alignment and an uncertainty guiding mechanism; and finally, compressing, quantifying and optimizing the student model, and deploying the student model to a vehicle-mounted end to realize real-time evaluation and dynamic updating of the states of multiple sensors. According to the method, the model complexity is greatly reduced while the evaluation precision is ensured through a knowledge distillation framework, and efficient and reliable state perception and fault-tolerant control support is provided for an intelligent driving system.
Owner:LIAONING UNIVERSITY
Method for optimizing data transmission in a multi-agent system
The invention relates to a method for optimizing data transmission in a multi-agent system, comprising: - obtaining raw data streams respectively from at least two agents; - classifying the received raw data streams based on predetermined characteristics; - obtaining requirements regarding end-to-end (E2E) latency and performance metrics from at least one node of the multi-agent system; - obtaining a set of interdependent QoS requests for the raw data streams suitable for data transmission via a a communication system; - computing a figure of merit for all the raw data streams based on the classified raw data streams, the node requirements, and the set of interdependent QoS requests; and - encoding the raw data streams and transmitting through the communication system the encoded data streams to the node according to the computed figure of merit. The invention ensures efficient and reliable transmission in the multi-agent communication system.
Owner:MITSUBISHI ELECTRIC R&D CENTRE EUROPE BV +1