Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

22results about How to "Reduce data noise" patented technology

Power demand real-time sensing and response system and method based on Internet of Things driving, electronic equipment and computer readable storage medium

The invention relates to the technical field of power dispatching, in particular to a power demand real-time sensing and responding system and method based on Internet of Things driving, electronic equipment and a computer readable storage medium, and the method comprises the steps: collecting power utilization power and power utilization time data at a user side, and carrying out the preprocessing of the data; constructing an electricity utilization power curve, and extracting electricity utilization characteristics; using a K-means clustering algorithm to carry out clustering analysis on the user power consumption characteristics, and dividing the users into k power consumption categories; a power consumption demand prediction model is constructed by training a long short-term memory (LSTM) network model, and the future power consumption power curve trend is predicted; and establishing a power utilization scheduling response mechanism based on the prediction result and the clustering-divided power utilization behavior categories in combination with a charging and discharging scheduling strategy of the power energy storage equipment, thereby realizing flexible scheduling response to the power utilization behaviors in different time periods in the future. According to the method, the power consumption management precision and efficiency can be improved, personalized power consumption requirements are met, and energy utilization is optimized.
Owner:HUNAN XILAIKE ENERGY STORAGE TECH CO LTD

Enterprise science and technology evaluation service system based on big data and AI analysis

InactiveCN119005817APredict comprehensive innovation capability scoreImprove reliabilityMachine learningOriginal dataBusiness enterprise
The invention particularly relates to an enterprise science and technology evaluation service system based on big data and AI analysis, and the system comprises a data preprocessing unit which is used for collecting original data related to enterprise innovation capability from a plurality of data sources; the index construction unit is used for defining a set of evaluation system covering quantitative and non-quantitative indexes according to the characteristics of enterprise innovation activities; the non-quantitative conversion unit is used for converting a non-quantitative index into a quantifiable evaluation value by constructing an innovation level bag-of-word model; regularly updating the bag-of-words model; the multi-dimensional evaluation construction unit is used for constructing a multi-dimensional spatial vector evaluation model in combination with quantitative and non-quantitative indexes; the weight dynamic adjustment unit is used for automatically adjusting weights of different indexes according to historical data by utilizing a machine learning algorithm; and the enterprise innovation evaluation unit is used for inputting each index characteristic value of the enterprise into the multi-dimensional space vector evaluation model to generate a comprehensive innovation ability score.
Owner:SHENZHEN HEZONGTIANXIA INFORMATION TECH CO LTD

Comprehensive energy multi-element load short-term prediction method based on secondary decoupling

The invention provides a comprehensive energy multi-element load short-term prediction method based on secondary decoupling, and the method comprises the steps: collecting a multi-element load data set, reconstructing a winter data set and a summer data set in the multi-element load data set, and carrying out the primary decoupling; decomposing the reconstructed winter data set and summer data set; dividing the decomposed data into a periodic term and a trend term so as to perform secondary decoupling; screening the trend items, and predicting the screened trend items to obtain a trend item prediction result; and performing multivariate load prediction on the periodic term to obtain a periodic term prediction result, and adding the trend term prediction result and the periodic term prediction result to obtain a final prediction result. According to the method, the coupling of the multi-element load piece can be effectively captured, accurate decomposition of nonlinear and non-stationary signals can be realized, the anti-noise capability and the time frequency resolution are improved, and the balance is considered while the prediction accuracy is effectively improved.
Owner:EAST CHINA JIAOTONG UNIVERSITY

Rare variation driven Alzheimer disease new gene identification and function evaluation method

ActiveCN119049545Aimprove understandingExclude benign mutationsBiostatisticsProteomicsCandidate Gene Association StudyAssociation tests
The invention discloses a rare variation-driven Alzheimer's disease new gene identification and function evaluation method, relates to the field of molecular biology, and solves the problems that in AD genetic research in the prior art, rare variation race specificity is high, research in non-European people is limited, and the research time is short. And the statistical effect of rare variation is relatively low. The method comprises the following steps: acquiring DNA in a peripheral blood sample of a target subject, and performing whole genome sequencing on the DNA in the peripheral blood sample to obtain a queue-level VCF file; rare and harmful genome variation in the VCF file is obtained by designing variation priority pipeline screening; carrying out an optimal sequence kernel association test on genome variation by taking a gene as a unit, and carrying out repetition and meta analysis in combination with a replication queue; the biological rationality of candidate genes is evaluated through network analysis, pathway enrichment and endophenotype association, and the functional consequence of top-level candidate variation is explained through an in-vitro experiment, so that the problem is solved.
Owner:FUDAN UNIVERSITY

Child safety seat high-altitude falling collision test system and method

The invention discloses a high-altitude falling collision test system and method for a child safety seat. The system is characterized in that an experimental test object is fixed in a safety seat to be experimented, a data sensor and a collision marker are arranged in the experimental test object, and the data sensor is connected with a data acquisition and analysis module; the to-be-tested safety seat is installed in a preset seat position of the first test vehicle; the vehicle lifting module lifts the first test vehicle to a preset falling height; a trigger sensor is arranged in the second test vehicle and is connected with the data acquisition and analysis module; at least one in-vehicle high-speed camera is arranged in the first test vehicle, the in-vehicle high-speed camera is used for shooting an experimental test object, and the in-vehicle high-speed camera is connected with the data acquisition and analysis module; according to the invention, comprehensive safety testing can be carried out on the child safety seat in a high-altitude falling mode, and the accuracy of testing the child safety seat is improved by simulating the real motion state of vehicle collision.
Owner:亿科检测认证有限公司

Power distribution network data fusion method and device considering measurement time sequence and nonlinear error

The invention discloses a power distribution network data fusion method and device considering a measurement time sequence and a nonlinear error, and belongs to the field of power distribution network line multi-source measurement data fusion. The method specifically comprises the steps of collecting PMU data, RTU data and AMI data; a time window is set behind the moment when the master station of the dispatching control center receives the PMU data, the RTU data received in the set time window is consistent with the PMU data in time scale, and the RTU data with PMU time scale characteristics are obtained; adopting a least square method to realize time section alignment of measurement data; performing data filling on the measurement data after section alignment by adopting Lagrange interpolation; performing noise reduction on the measurement data after data filling, and constructing a power distribution network state and measurement model; and solving the state and measurement model by adopting an unscented Kalman filtering algorithm to realize the multi-source measurement data fusion of the power distribution network. According to the method, rapid data fusion of multi-source measurement data is realized, and the fusion result is more accurate by considering the condition of combination of actual communication delay and unscented Kalman filtering.
Owner:STATE GRID SHANGHAI MUNICIPAL ELECTRIC POWER CO

Safe computer tamper system

The invention relates to the technical field of secure computers, and discloses a secure computer anti-tamper system which comprises a multiple encryption module, a physical lock module, an intelligent sensor module, an alarm module, a user interface module and a cloud management platform module. The multi-encryption module is combined with a physical lock and a software encryption technology, and a computer system is protected through a multi-level encryption method. By combining a physical lock and various advanced encryption technologies, the security of the computer anti-disassembly system is remarkably improved, the physical lock provides basic hardware protection and prevents unauthorized disassembly and physical invasion, meanwhile, the system adopts multiple encryption algorithms to perform encryption protection on data and communication, the difficulty of cracking the system is increased, and the security of the system is improved. And the intelligent sensor monitors the state of hardware in the computer in real time, so that the system can provide higher protection capability when facing potential security threats, and the security of data and hardware is ensured.
Owner:广州新华学院

Entity alignment method for heterogeneous knowledge graphs based on multi-view and graph enhancement

The present invention discloses a heterogeneous knowledge graph entity alignment method based on multi-view and graph enhancement, which relates to the technical field of knowledge graphs, including: embedding entities, relationships, attributes and attribute values ​​of knowledge graphs by vectors to obtain initialization vectors; constructing entity neighbor adjacency matrix, entity relationship adjacency matrix, entity attribute adjacency matrix and entity attribute value adjacency matrix of knowledge graphs, and constructing entity neighbor view, relationship view, attribute view and attribute value view; fusing attribute view and attribute value view to form attribute-attribute value view; integrating various views to obtain knowledge graph entity fusion embedding representation; generating enhanced knowledge graphs, and obtaining entity fusion embedding representation of enhanced knowledge graphs; and obtaining heterogeneous knowledge graph entity alignment model. The present invention uses multi-views to better capture all information on knowledge graphs, and uses graph enhancement technology to reduce data noise and errors, thereby improving the accuracy of entity alignment.
Owner:HUAQIAO UNIVERSITY

Aero-engine gas path parameter deviation value preprocessing method

ActiveCN115292648BEliminate the effects ofremove gross error
The present application relates to the technical field of aero-engine gas path parameter processing, in particular to a kind of aero-engine gas path parameter deviation value preprocessing method, first, using fingerprint map to correct the deviation of gas path parameter caused by non-component performance degradation in gas path parameter deviation value offset;Then, using Grubbs criterion to remove gross error in short-term gas path parameter deviation value that meets the requirement of independent and identically distributed by equal-length overlapping sliding window division;Then, aiming at the problems such as mode coupling and insufficient IMF signal based on correlation coefficient selection that EMD decomposition used by general noise reduction method may encounter, EEMD decomposition is introduced, and single-sample T test is used to distinguish high-frequency and low-frequency IMF signal, by only using SG filter to reduce noise for high-frequency IMF, the problem that long-term trend signal is damaged due to data filtering and noise reduction is avoided;Finally, the superiority of the data preprocessing method proposed in the present application in gas path parameter deviation value preprocessing is verified by comparison experiment.
Owner:SHANDONG TIANLAN INFORMATION TECH CO LTD

Methods for identification and functional assessment of novel rare variant-driven Alzheimer's disease genes

ActiveCN119049545Bimprove understandingReduce data noiseBiostatisticsProteomics
The present invention discloses a method for identifying and functionally evaluating new genes for Alzheimer's disease driven by rare variants, relates to the field of molecular biology, and solves the problems of strong racial specificity of rare variants and limited research in non-European populations in the genetic research of AD in the prior art; and low statistical power of rare variants; the present invention obtains DNA from peripheral blood samples of target subjects and performs whole genome sequencing on the DNA in the peripheral blood samples to obtain a cohort-level VCF file; obtains rare and harmful genomic variants in the VCF file by designing a variant priority pipeline for screening; performs optimal sequence-nuclear association tests on genomic variants on a gene basis, and performs repeated and meta-analyses in combination with replication cohorts; evaluates the biological plausibility of candidate genes through network analysis, pathway enrichment, and endophenotype association, and explains the functional consequences of top candidate variants through in vitro experiments, thereby solving the above problems.
Owner:FUDAN UNIVERSITY

Picture labeling and quality checking method and system based on multi-dimensional safety behavior

PendingCN122510899AReduce data noiseReal-time standardized operationsEngineeringData mining
The application discloses a picture labeling and quality checking method and system based on multi-dimensional safety behavior, relates to the technical field of image data labeling, and comprises the following steps: S1, collecting pictures to be labeled and performing pretreatment; S2, constructing a multi-dimensional labeling model and a dynamic behavior baseline model to obtain a picture labeling framework; S3, multi-dimensional collaborative labeling and dangerous behavior identification; S4, performing quality evaluation on the dangerous behavior labeling result through a hierarchical quality inspection model; and S5, calculating a comprehensive quality score, and optimizing and storing the labeling result. The picture labeling and quality checking method and system based on multi-dimensional safety behavior solve the problem of difficult control of the quality of dangerous behavior labeling data in the existing technology in the construction operation scene.
Owner:TIBET JUNZHICHONG INFORMATION TECHNOLOGY CO LTD

Cardiovascular age prediction method and device based on deep learning and metabolic profile analysis

The present invention discloses a cardiovascular age prediction method and device based on deep learning and metabolic spectrum analysis, which relates to the field of biological data processing, including: obtaining blood samples of healthy people and people with cardiovascular diseases and performing data processing and feature selection to determine relevant metabolites; constructing a cardiovascular age prediction model based on deep learning and training it to obtain a trained cardiovascular age prediction model; obtaining blood samples of people to be tested and processing it to obtain a metabolomics feature matrix composed of metabolic features corresponding to relevant metabolites, inputting the metabolomics feature matrix into the trained cardiovascular age prediction model, passing through a metabolic feature representation module to obtain a metabolic feature representation, inputting the metabolic feature representation into a feature representation processing module, predicting the cardiovascular age of the person to be tested, and calculating the cardiovascular age acceleration. The present invention solves the problems of high cost, low accuracy, lack of extensive verification, etc. of existing cardiovascular age prediction.
Owner:THE FIRST AFFILIATED HOSPITAL OF XIAMEN UNIV

Target detection self-training method and system based on uncertainty

The invention discloses a target detection self-training method and system based on uncertainty, and the method comprises the steps: constructing a target data set which comprises a data sample with a label and a data sample without a label; calculating a first total error based on the prediction result of the labeled data sample and the true value of the labeled data sample, and training the teacher network through the first total error to obtain a first model; calculating a second total error based on the prediction result of the data sample without the label and the pseudo label of the data sample without the label, and inputting the second total error into a student network for training to obtain a second model; and performing integrated updating on the first model and the second model by using exponential moving average to obtain a final training detection model. According to the method, the problem of low model robustness caused by uncertainty of feature offset in the whole self-training process is solved, and data noise in a pseudo tag can be greatly reduced.
Owner:XI AN JIAOTONG UNIV

A self-balancing mounting device and method for large ground penetrating radar

The present invention discloses a self-balancing mounting device and method for a large ground-penetrating radar, comprising: a bracket (1), a plurality of mounting steel wires (2) and a tensioning device (3); two mounting steel wires (2) with ends at both ends are arranged on the bracket (1); the mounting steel wires (2) are connected to the tensioning device (3). The present invention arranges two mounting steel wires (2) with ends at both ends on the bracket (1), and uses the tensioning device (3) to laterally tension the mounting steel wires (2). As the mounting steel wires (2) are tightened, the radar is lifted and connected to the bracket (1); a rubber block (5) for seismic isolation is arranged on the mounting steel wires (2), so as to achieve the effect of protecting the radar and reducing interference with data. The mounting of the ground-penetrating radar can be completed by a single person, reducing labor intensity, and the mounting of four mounting ears can be completed by one tensioning. This saves time and effort.
Owner:GUANGZHOU CHENG AN LUQIAO DETECTION CO LTD

Structural finite element model correction method based on geometric deviation laser scanning

The invention discloses a structural finite element model correction method based on geometric deviation laser scanning, and the method comprises the steps: collecting the three-dimensional laser point cloud data of an engineering structure, determining the geometric deviation of the engineering structure based on the three-dimensional laser point cloud data, and correcting the geometric deviation of the engineering structure. The geometric deviation reflects the difference between a node coordinate of a finite element model of the engineering structure and a real measurement value; establishing a finite element model of the engineering structure, performing modal analysis based on the finite element model, and identifying modal parameters of the engineering structure based on the dynamic monitoring data of the engineering structure; and constructing a comprehensive target optimization function based on the geometric deviation and the modal parameters of the engineering structure, and correcting the finite element model based on the comprehensive target optimization function. According to the method, the finite element model is comprehensively corrected in combination with the geometric deviation and the dynamic monitoring data, geometric changes are considered, the dynamic response of the structure is reflected, and the accuracy of the finite element model of the engineering structure is improved.
Owner:HARBIN INSTITUTE OF TECHNOLOGY (SHENZHEN) (INSTITUTE OF SCIENCE AND TECHNOLOGY INNOVATION HARBIN INSTITUTE OF TECHNOLOGY SHENZHEN)

A new type of cantilever beam impact testing machine

The utility model relates to cantilever beam impact testing machine technical field discloses a new -type cantilever beam impact testing machine, including shell one, the fixed plate is fixedly connected with the shell one inner wall, the fixed plate outer wall is fixedly connected with the pivot, the pivot one end is fixedly connected with the swing bar, the buckle no.
Owner:HUBEI JULONG NEW MATERIALS TECHNOLOGY CO LTD

Live broadcast exception processing method, device, readable storage medium and equipment

The embodiment of the present application discloses a method, device and computer-readable storage medium for handling live broadcast exceptions; by obtaining the occurrence time of a preset specific interruption event and the live broadcast room identifier of the live broadcast room where the preset specific interruption event occurs from a real-time interruption database, the preset specific interruption event includes that the live broadcast audience terminal does not pull the live broadcast room video stream within a preset time after entering the live broadcast room; based on the occurrence time and the live broadcast room identifier, the preset specific interruption event is aggregated and processed to obtain the occurrence frequency information of the preset specific interruption event of at least one live broadcast room; according to the occurrence frequency information, an abnormal target live broadcast room is determined in the live broadcast room; abnormal prompt information for the preset specific interruption event is generated for the target live broadcast room and an alarm is issued, so as to handle the live broadcast exception for the target live broadcast room based on the abnormal prompt information of the alarm. In this way, real-time processing of abnormal situations in the live broadcast room can be achieved, thereby improving the efficiency of handling live broadcast exceptions.
Owner:SHUXING TECH (BEIJING) CO LTD

Motion image acquisition rotary table, motion image acquisition device and control method

The invention provides a motion image acquisition rotary table, a motion image acquisition device and a control method, the motion image acquisition rotary table is provided with a plurality of position coding hole sites and a whole circle coding hole site, the plurality of position coding hole sites are uniformly arranged, the plurality of position coding hole sites form a circle, the circle is parallel to the plane of the rotary table, and the whole circle coding hole site is parallel to the plane of the rotary table. And the circle center of the circumference is positioned on a rotating shaft of the motion image acquisition turntable. By using the motion image acquisition rotary table, the image acquisition position can accurately correspond to the image data.
Owner:IRAY TECHNOLOGY CO LTD

Model training method and device for text open task

The invention provides a model training method and device for a text open task, on one hand, a reward model is trained through a training text pair marked with a relative quality advantage relation, original complex absolute quality scores of the reward model are converted into simpler and more stable relative advantage judgment, the model training cost and subjective influence are reduced, and the training efficiency is improved. The accuracy and confidence of the reward signal are improved, one first reply in the output replies of the existing model is set as a reference text reply, and it is ensured that the output quality of the trained text model is superior to the output quality of the existing model; on the other hand, in the training process of the to-be-trained model, a second reply output by the to-be-trained model is monitored, and the target reward model is adjusted according to the monitoring result; and / or, the reward value output by the target reward model for the second reply is monitored, and the reference text reply is adjusted according to the monitoring result, so that the optimization direction of the text model is ensured, and the training effect of the text model is improved.
Owner:SHANGHAI XIYU JIZHI TECH CO LTD

Virus transmission tracing method and system driven by spatio-temporal trajectory data

PendingCN119092145AHighly consistent with the real situationaccurate identificationMedical data miningEpidemiological alert systemsData setAlgorithm
The invention discloses a space-time trajectory data driven virus transmission tracing method and system. The method comprises the following steps: firstly, cleaning and preprocessing an obtained patient space-time trajectory data set; then, based on the risk area grading system and the infection theory, a set propagation chain conforming to the infection theory is constructed for the high-risk province; then coding the preprocessed patient space-time trajectory data into patient features by using a suitable coding mode; training a traceability model for high-risk provinces based on the constructed set propagation chain and the encoded patient features; and finally, applying the trained high-risk province traceability model to related high-risk provinces based on a knowledge migration mode, and performing a propagation chain construction task. The virus propagation chain constructed by the method disclosed by the invention has the characteristics of expandability in data volume, diversity in propagation mode and high conformity with the real situation.
Owner:XI AN JIAOTONG UNIV

Intelligent security vulnerability monitoring method and system

The invention relates to the technical field of network security, in particular to an intelligent security vulnerability monitoring method and system, and the method comprises the steps: deploying distributed monitoring nodes, collecting network asset data, traffic data and behavior log data in real time, and transmitting the data to a vulnerability analysis unit; the vulnerability analysis unit preprocesses the collected data and eliminates data noise and deviation; based on a mode of combining a deep learning model and a rule engine, performing vulnerability feature extraction and abnormal behavior identification; and performing risk level judgment on the identified vulnerabilities through the vulnerability risk assessment model, generating a vulnerability monitoring report, pushing the vulnerability monitoring report to a security management platform, and triggering a corresponding emergency response mechanism. According to the invention, through distributed monitoring, intelligent analysis and risk assessment, all-around, rapid and precise monitoring of network security vulnerabilities is realized, and the security protection capability of a network system is improved.
Owner:WUXI DAYUN EDUCATION TECH CO LTD

Method, apparatus and electronic device for determining quality model of transmission link

This invention discloses a method, apparatus, and electronic device for determining the quality model of a transmission link. The method includes: acquiring training index data corresponding to the transmission link of a power optical fiber, wherein the training index data is used to characterize the quality status of the transmission link; obtaining a classification result corresponding to the training index data based on a clustering algorithm; determining a quality level label corresponding to the training index data based on the classification result; and training a predetermined initial quality model using a deep residual network algorithm based on the training index data and the corresponding quality level label to obtain a target quality model for the transmission link. This invention solves the technical problems of low model accuracy, poor robustness, and low convergence caused by data noise interference in the quality model of the transmission link.
Owner:STATE GRID BEIJING ELECTRIC POWER CO +2