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

39303results about How to "Improve applicability" patented technology

Multi-source twin data fusion tunnel structure health monitoring and early warning method and system

The invention discloses a tunnel structure health monitoring and early warning system based on multi-source twin data fusion, which relates to the technical field of tunnel health monitoring and comprises a data acquisition module, a multi-source data fusion module, a health state evaluation module and an early warning module. The data acquisition module can collect various monitoring data of equipment state data, tunnel environment data, traffic condition data and tunnel structure of a monitored tunnel, the multi-source data fusion module integrates and processes the collected multi-source data by using an advanced data fusion technology, and the health state evaluation module evaluates the health state of the monitored tunnel according to the fused data. And the early warning module is used for evaluating the health condition of the tunnel structure by adopting an intelligent evaluation algorithm and identifying potential structure problems, and the early warning module is used for forming an early warning result by applying an early warning algorithm on the basis of an evaluation result. The advanced data fusion method and early warning algorithm are adopted, and the accuracy and reliability of tunnel health monitoring and early warning are improved.
Owner:SHENZHEN UNIV

Method and system for generating enhanced knowledge questions and answers for mixed retrieval of heterogeneous database

The invention discloses a method and system for generating enhanced knowledge questions and answers for heterogeneous database mixed retrieval, and the method comprises the following steps: (1) processing text data, and constructing a heterogeneous database; (2) receiving user query, and performing mixed retrieval in the heterogeneous database; the retrieval results are merged, and documents and triple information related to user query are obtained; (3) reordering the retrieved documents and triple information by using a reordering model, and selecting K retrieval results with the highest relevancy with user query; and (4) combining user query and retrieval results, constructing cue words, inputting the cue words into the large language model, and generating coherent and accurate natural language answers. By means of the method and device, more comprehensive and accurate information retrieval can be achieved, and the accuracy and reliability of knowledge question and answer results are improved.
Owner:ZHEJIANG UNIV

Commercial credit evaluation and supervision method based on multi-modal coevolution algorithm

The invention relates to a business credit evaluation and supervision method based on a multi-modal coevolution algorithm, and the method comprises the steps: collecting and cleaning multi-modal data, guaranteeing the data quality, uniformly storing the multi-modal data as a structural database, extracting features through cross-modal comparative learning, and enabling the features to be embedded into a dynamic multi-modal knowledge graph, a modal sharing network is used to extract and fuse modal sharing and specific features, consistency of modal feature distribution is optimized through adversarial learning, a global co-evolution credit scoring model is initialized, a dynamic multi-objective algorithm is used to optimize accuracy, robustness and interpretability of the model, and a credit scoring result is obtained. A causal reasoning model is introduced to analyze driving factors of enterprise credit changes, a causal relationship is established, a closed-loop feedback system is designed, a credit evaluation model is updated, adaptability is enhanced, an interpretable credit evaluation report is generated based on a generative adversarial network, and transparency and credibility are enhanced. According to the method, the accuracy and adaptability of credit assessment are improved, and the robustness and interpretability of an assessment result are ensured.
Owner:SOUTHEAST UNIV

Dynamic energy efficiency optimization method and system based on artificial intelligence and machine learning

The invention belongs to the field of energy efficiency optimization in industrial production, and discloses a dynamic energy efficiency optimization method and system based on artificial intelligence and machine learning, and the method comprises the steps: collecting energy consumption data and environment parameters in real time through a sensor network and Internet of Things equipment; cleaning, screening and standardizing the collected data; constructing an energy efficiency prediction model and an optimization model based on a machine learning algorithm; training the prediction model and the optimization model by using historical data, and adjusting model parameters; and the trained model is utilized to analyze and predict data acquired in real time, and production parameters are dynamically adjusted to realize optimization of energy efficiency. According to the method, key parameter changes in the production process can be accurately predicted and optimized, dynamic adjustment of production parameters is achieved, an optimal production adjustment strategy can be automatically made, scientificity and efficiency of production decision making are improved, and the method has remarkable technical and economic effects in the field of industrial production.
Owner:CHONGQING IND BIG DATA INNOVATION CENT CO LTD

Virtual power plant optimal scheduling method and system based on adjustable capability aggregation of source-load-storage heterogeneous resources

The invention discloses a virtual power plant optimal scheduling method and system based on source load storage heterogeneous resource adjustable capability aggregation, and relates to the field of source load storage heterogeneous resource aggregation and optimal scheduling, and the method comprises the steps: collecting physical characteristic parameters of each resource of source load storage in an area under jurisdiction of a virtual power plant; predicting distributed wind power photovoltaic output and fixed load of the regional power grid; determining operation characteristics of internal resources of the virtual power plant according to the physical characteristic parameters, the distributed wind power photovoltaic output and the fixed load; according to operation characteristics of internal resources of the virtual power plant, considering network operation constraints, determining a regulation and control power feasible region and regulation and control cost of the virtual power plant, and constructing a bidding decision optimization model of the virtual power plant participating in the energy-standby joint market by taking the lowest total cost of the virtual power plant as an optimization target; and performing optimization scheduling on the virtual power plant according to the bidding decision optimization model. According to the method, aggregation of massive source-load-storage heterogeneous resources can be realized, and a basis is provided for optimization decision-making of virtual power plant operators.
Owner:NORTH CHINA ELECTRIC POWER UNIV +2

Riverbed evolution prediction method based on multi-source data fusion

The invention provides a riverbed evolution prediction method based on multi-source data fusion, and belongs to the technical field of hydrological resource monitoring. The method comprises the following steps: acquiring and fusing multi-source data by using different devices, performing data preprocessing to form unified three-dimensional grid data, and then generating a multi-scale feature map based on a designed feature extraction network; based on the designed multi-channel convolutional neural network, learning and extracting features on multiple channels at the same time, and performing hierarchical expression and fusion on different information sources; and finally, based on the designed time sequence regression network, extracting a complex dependency relationship in the time sequence through a double-layer LSTM structure, learning residual error information through an adaptive residual error correction module to improve prediction precision, and obtaining a final prediction result. The method shows relatively high applicability and robustness in a complex river channel environment, and can cope with various river bed form changes caused by natural factors and human activities through multi-source data fusion and an innovative feature extraction and regression method.
Owner:YELLOW RIVER INST OF HYDRAULIC RES YELLOW RIVER CONSERVANCY COMMISSION

Agricultural machinery scheduling method and system based on deep learning

The invention provides an agricultural machinery scheduling method and system based on deep learning. Belongs to the technical field of deep learning and agriculture, and the method comprises the steps: collecting multi-source data, carrying out timestamp synchronization on data from different sources through a data synchronization mechanism, integrating the multi-source data through a data fusion algorithm, and constructing a multi-dimensional data set; through an adaptive network structure, a feature extraction strategy is automatically adjusted according to different crops and environmental conditions, a farmland environment model is constructed based on extracted features through a machine learning algorithm, and a prediction result is output. Through synchronization and fusion of multi-source data, a more comprehensive farmland environment model can be constructed, so that agricultural operation is more accurate, resource waste is reduced, and crop yield and quality are improved.
Owner:ZHEDA ZHENGCHENG TECH CO LTD

Tumor area double-view detection method and system based on image modality and view

The invention provides a tumor area double-view detection method and system based on image modalities and views, and relates to the field of medical image analys.The method comprises the steps that multi-modal image data are collected from the body of a patient, and image data of different view angles at the same position are obtained according to time frames; extracting multi-modal features from the image data of different modalities and different view angles, and performing weighted fusion on the extracted multi-modal features to form a fused feature map; inputting the fused feature map to train and predict the 3D-CNN network so as to identify and position a tumor region, and marking the tumor region by using a mask output by the segmentation model; calculating the pixel number of the tumor area in each layer of image according to the marked mask, and multiplying the pixel number by the volume of each pixel to obtain the volume of the tumor area; and the segmentation results of each layer of image are superposed, and a three-dimensional model of a tumor region is generated by using 3D reconstruction, so that the limitation of single-mode and single-view image data in tumor detection is overcome.
Owner:SICHUAN CANCER HOSPITAL

3D model multi-scale lightweight method and system based on topological optimization

The invention provides a 3D model multi-scale lightweight method and system based on topological optimization, and relates to the technical field of model lightweight, and the method comprises the steps: carrying out the mesh generation of a target 3D model, and generating a finite element model; a multi-scale topological optimization model is established, density distribution is adjusted through a density filtering function, and a macroscopic lightweight structure is obtained in combination with minimum size constraint and gray scale gradient projection processing; based on a macroscopic lightweight structure, a skeleton structure is obtained, based on an adaptive polygon segmentation rule, a porous filling unit is constructed, a multi-parameterized model is fused, a parameterized representation method of three variables of density, size and shape is established, and an intermediate scale structure is obtained through a multi-target genetic algorithm in combination with a sensitivity analysis technology; and embedding the intermediate-scale structure into the macroscopic lightweight structure, generating a multi-scale structure in combination with cascade optimization, calculating equivalent material attributes by adopting a homogenization method, iteratively adjusting model parameters, and finally obtaining a multi-scale lightweight result.
Owner:HANGZHOU ELECTRIC EQUIP MFG

Intelligent temperature compensation method for pressure sensor

The invention discloses an intelligent temperature compensation method for a pressure sensor, and relates to the technical field of sensor equipment. Comprising the following steps: constructing a temperature response characteristic model, collecting temperature data in real time, loading a temperature compensation algorithm on the temperature data collected in real time, updating the temperature response characteristic model in real time, performing real-time temperature compensation correction in output of a pressure sensor, and establishing an error prediction and early warning mechanism. And carrying out adaptive optimization and long-term calibration on the temperature compensation method. According to the intelligent temperature compensation method for the pressure sensor, through temperature response characteristic modeling, a temperature compensation algorithm, real-time model updating and other methods, the measurement precision and long-term stability of the sensor in a complex environment are remarkably improved, and rapid temperature difference changes and nonlinear errors can be adaptively coped with. Meanwhile, it is ensured that the temperature compensation algorithm can be continuously optimized along with environment changes and sensor aging, high applicability and expandability are achieved, and technical guarantee is provided for application of the sensor under severe working conditions.
Owner:XIAN SIWEI SENSOR TECH CO LTD

Bridge structure fatigue crack propagation prediction method and system

The invention provides a bridge structure fatigue crack propagation prediction method and system, and relates to the technical field of artificial intelligence. According to the method, structural parameter information of a bridge is collected, a three-dimensional model of the bridge structure is established through finite element analysis, and a fatigue crack propagation prediction model is established based on the fracture mechanics theory and a fatigue damage accumulation model; performing expansion simulation on preset cracks in the three-dimensional model of the bridge structure, calculating a crack expansion rate and predicting a crack expansion path, correcting a crack expansion simulation result in combination with an actual working environment and a load condition of the bridge structure, and calculating a crack expansion simulation result according to the corrected crack expansion simulation result. And predicting crack propagation information of the bridge structure in the preset time period. The method can accurately predict the expansion path, rate and future state of the crack under a complex working condition, thereby providing a scientific basis for maintenance and repair of the bridge, realizing real-time feedback and dynamic adjustment, ensuring high efficiency and accuracy of a prediction result, and providing a scientific basis for maintenance management of the bridge.
Owner:ZHEJIANG YIQIAO ENG TECH RES CO LTD

Supervised time sequence water level data generation method and system and storage medium

The invention relates to the technical field of machine learning, in particular to a supervised time sequence water level data generation method and system and a storage medium. The method comprises the following steps: acquiring multi-source hydrological monitoring data; performing abnormal value identification and data preprocessing according to the hydrological monitoring data to generate standardized hydrological basic data; constructing a data generation model of the conditional generative adversarial network based on the standardized hydrological basic data; hydrological variable driven condition feature regulation and control are carried out on the data generation model, parameter optimization of the generation network is carried out, and hydrological adaptability enhanced condition input feature data are generated; and injecting the condition input characteristic data into the data generation model to generate initial water level time sequence data. According to the method, the rare extreme hydrological events are oversampled by using a hierarchical importance sampling and self-service method (Bootstrap), so that the performance of the generative model in rare event prediction is remarkably enhanced.
Owner:YANGTZE RIVER WATER CONSERVANCY COMMISSION HYDROLOGY MIDDLE YANGTZE RIVER HYDROLOGY & WATER RESOURCES SURVEY BUREAU (YANGTZE RIVER WATER CONSERVANCY COMMISSION HYDROLOGY MIDDLE YANGTZE RIVER WATER ENVIRONMENT MONITORING CENT) +2

Data asset multi-dimensional evaluation method and system based on block chain and machine learning

The invention provides a data asset multi-dimensional assessment method and system based on a block chain and machine learning, an assessment index system covering dimensions such as data quality, value, risk and security is constructed, a block chain technology is adopted to carry out assessment process and result evidence storage, a machine learning model is introduced to dynamically optimize index weights, and the assessment efficiency is improved. And a data asset marketization pricing mechanism is explored and formed. According to the method, the problems that an existing data asset evaluation method is single in dimension, high in subjectivity and lack of credibility are solved, scientific evaluation, marketization pricing and credible publication of the data asset value are achieved, important reference is provided for data element circulation, and the method has great significance in promoting prosperous development of the data element market.
Owner:AACAT TECHNOLOGY LTD

Primary helium fan residual life prediction method fusing signal processing and intelligent algorithm

The invention discloses a signal processing and intelligent algorithm fused main helium fan residual life prediction method, and relates to the technical field of mechanical equipment state monitoring and maintenance management, and the method comprises the steps: collecting equipment operation data through multiple sensors, carrying out the feature fusion through principal component analysis, and achieving the fault mode recognition through a deep learning algorithm; and the residual life of the equipment is predicted by utilizing transfer learning and an XGBoost algorithm, and a diagnosis and prediction model is integrated into an online monitoring system, so that real-time perception and dynamic prediction of the health state of the equipment are realized, and decision support for maintenance and spare part management is provided. According to the method, the accuracy of fault diagnosis and life prediction of the primary helium fan is effectively improved, and a new thought is provided for safe operation and intelligent maintenance of equipment.
Owner:HUANENG NUCLEAR ENERGY TECH RES INST CO LTD +1

Intelligent financial statement automatic generation platform

The invention discloses an intelligent financial statement automatic generation platform, and relates to the technical field of financial statement generation, and the platform comprises a data integration module which is used for connecting and synchronizing a data source, the data source comprises a bank transaction record, an enterprise resource planning system and an electronic invoice system, and data is grabbed through connection configuration of a data interface. According to the intelligent financial statement automatic generation platform, classification rules of financial subjects are customized by a user, so that the platform can adapt to specific financial statement requirements and business characteristics of different enterprises, the flexibility and applicability of the system are improved, statement customization is supported through the platform, financial statements of different versions are provided, and the user experience is improved. According to the method, different requirements of different users on financial information are met, the flexibility and applicability of the system are improved, a temporary version exclusive to the current requirement is formulated based on a customizable mechanism of a classification rule, and more targeted financial support is provided for decision and management of enterprises.
Owner:JILIN ACAD OF TRADITIONAL CHINESE MEDICINE

Pumped storage power station operation trend early warning analysis method and system

The invention discloses a pumped storage power station operation trend early warning analysis method and system. The method comprises the following specific steps: S1, data collection and preprocessing; s2, feature selection and construction: screening key features of internal operation data by combining a principal component analysis method with a Lasso regression method; s3, constructing a prediction model: taking external environment factors as weight factors, adopting a deep learning model combining an attention mechanism and a bidirectional LSTM network, integrating an XGBoost model to perform training optimization to obtain the prediction model, and predicting various operation parameters of the power station; and S4, constructing a double-early-warning mechanism, setting an early-warning threshold value of each operation parameter of the power station based on an early-warning rule, comparing a prediction result of the prediction model constructed in the step S3 with the early-warning threshold value, and establishing an instant early-warning mechanism and a long-period early-warning mechanism at the same time. Accurate early warning of the operation trend of the pumped storage power station is realized, the operation and maintenance cost is saved, and the operation intelligence and reliability of an electrical system are improved.
Owner:CHINA THREE GORGES PROJECTS DEV CO LTD +1

Simulation method of geological mineral resource reserves

The invention relates to the technical field of mineral resource reserve simulation, in particular to a geological mineral resource reserve simulation method. The method comprises the following steps: carrying out multi-source heterogeneous data acquisition on a target geological area, and carrying out comprehensive feature extraction to obtain a comprehensive data set; constructing an initial static geologic model according to the comprehensive data set to obtain the initial static geologic model; building a multi-physics field coupling model according to the initial static geologic model to obtain a multi-physics field coupling model grid; performing metallogenic fluid migration and enrichment simulation according to the multi-physics field coupling model grid to obtain a metallogenic fluid migration simulation result; carrying out geologic body geometrical morphology updating on the initial static geologic model according to a metallogenic fluid migration simulation result to obtain a dynamic geologic model; and performing reserve prediction model training according to the dynamic geologic model to obtain a reserve prediction model. According to the method, the mineral resource reserves are evaluated more accurately and dynamically through a mineral resource reserve simulation technology.
Owner:XICHANG COLLEGE

Particle size distribution measuring method and system based on machine learning and light scattering method

The invention relates to the field of particle size distribution measurement, in particular to a particle size distribution measurement method and system based on machine learning and a light scattering method.The method comprises the steps that a to-be-measured particle sample is irradiated through a light source, and meanwhile light scattering intensity data of different scattering angles and different wavelengths are collected through a multi-angle scattered light detector; constructing a multi-modal deep learning model, and forming an integrated learning architecture in combination with a support vector machine model and a random forest model; inputting the preprocessed multi-wavelength light scattering intensity data into the trained integrated learning architecture to obtain a preliminary particle size distribution result of the to-be-detected particle sample; based on the improved Mie scattering theory, a high-precision theoretical model considering the particle shape, the refractive index, the surface roughness and the internal structure is constructed; an improved Bayesian optimization algorithm is adopted, and the initial particle size distribution result is optimized and adjusted by minimizing the error between the theoretical light scattering intensity and the actually collected light scattering intensity, so that a corrected particle size distribution result is obtained.
Owner:CHANGCHUN UNIV OF SCI & TECH

Voice recognition method and device for audio and video progressive fusion training in noise environment

The invention discloses an audio and video identification method and device for multi-mode progressive fusion training in a noise environment, and relates to the technical field of audio and video multi-mode. The method comprises the following steps: firstly, constructing an audio and video joint encoder model by using multi-modal data, and carrying out iterative unsupervised pre-training; a multi-scale feature extraction module and a decoder are added to construct an audio and video recognition model; and finally, carrying out multi-stage progressive fusion training on the audio and video recognition model, specifically comprising two steps: a modal course learning stage and a noise course learning stage, and realizing guided learning between modals. According to the method, cross-modal feature joint mapping is realized by using audio and visual modal information, so that the model can still keep efficient voice instruction recognition performance in a complex noise environment.
Owner:SICHUAN UNIV

Risk assessment system for chronic diseases

The invention relates to the technical field of medical data processing, in particular to a risk assessment system for chronic diseases. The system is used for performing physiological index analysis according to a medical physiological monitoring signal to generate physiological index data; performing individual physiological index weight empowerment on the physiological index data based on genomics data to obtain individual physiological index data; performing quantitative analysis on pathological influence factors of chronic disease types according to the electronic health record data and the individual physiological index data, and generating individual chronic disease influence factor quantitative data; performing composite interaction risk factor analysis on the individual chronic disease influence factors based on the quantitative data of the individual chronic disease influence factors to generate individual chronic disease composite interaction risk factor data; and performing individual chronic disease risk assessment based on the individual chronic disease composite interaction risk factor data to generate individual chronic disease risk assessment data. According to the invention, chronic disease health management and accurate evaluation are realized through collaborative evaluation work of chronic disease risks.
Owner:SHANDONG VICTOR INFORMATION TECH CO LTD

Link test method, electronic device, storage medium, product and computing device

The invention discloses a link test method, electronic equipment, a storage medium, a product and computing equipment, and is applied to the technical field of data transmission. The method comprises the following steps: determining to-be-tested equipment according to a to-be-tested link and a link training trigger condition; when the to-be-tested link finishes executing the link training operation and the link is recovered, determining a target register storing running state parameter information of the to-be-tested link in the to-be-tested device based on a link training trigger condition, and reading the running state parameter information; and when the link to be tested is diagnosed to have no fault according to the running state parameter information, executing the corresponding link training operation again based on the link training triggering condition. According to the method, the problems that related technologies depend on special tools, the starting time is long and the low-probability hardware fault reproduction period is long can be solved, the applicability is high, and the link testing efficiency can be effectively improved.
Owner:INSPUR SUZHOU INTELLIGENT TECH CO LTD

Method and system for analyzing stability of mine slope under freezing and thawing action

The invention relates to the technical field of mine slope stability analysis, in particular to a mine slope stability analysis method and system under the freeze-thaw action, and the method comprises the following steps: carrying out a freeze-thaw cycle simulation experiment on a mine slope, and obtaining freeze-thaw cycle experiment information of the mine slope; establishing a freeze-thaw cycle-mine slope pore analysis model based on the freeze-thaw cycle experimental information, and analyzing the pore change trend of the mine slope rock under different freeze-thaw conditions through the freeze-thaw cycle-mine slope pore analysis model; according to the pore change trend and the freeze-thaw cycle experiment information, a mine slope gliding force prediction model is constructed, and the mine slope gliding force prediction model is utilized to predict the gliding conditions of the mine slope under different freeze-thaw conditions; and establishing a mine slope stability evaluation system by combining the pore change trend, the gliding condition and the freeze-thaw cycle experiment information, and analyzing the stability condition of the mine slope under the freeze-thaw condition according to the mine slope stability evaluation system.
Owner:KUNMING PROSPECTING DESIGN INSTITUTE OF CHINA NONFERROUS METALS INDUSTRY CO LTD +1

Method for measuring conductivity and inner diameter of pipeline based on sweep frequency eddy current

The invention provides a method for measuring the conductivity and the inner diameter of a pipeline based on sweep frequency eddy current. The method comprises the following steps: establishing an eddy current detection device, designing an absolute eddy current detection coil, placing the coil in a metal pipeline, arranging the coil and the pipeline coaxially, introducing alternating current to excite an alternating magnetic field, and inducing eddy current on the surface and the near surface of the pipeline; establishing a metal pipeline eddy current analysis theoretical model, setting coil geometric parameters, and calculating an analysis expression of coil impedance changing along with excitation frequency; calculating the peak frequency and the corresponding phase value of the impedance phase of the coil in a frequency sweep mode, respectively changing the conductivity and the inner diameter of the pipeline, extracting the corresponding peak frequency and phase value, and constructing an inner diameter relation model based on quadratic curve fitting and a conductivity relation model based on power function fitting; and inverting the inner diameter and the conductivity of the metal pipeline by combining a coil impedance sweep frequency phase measurement result. The method can be used for simultaneously measuring the conductivity and the inner diameter of the pipeline, and is suitable for nondestructive testing and performance evaluation of the metal pipeline.
Owner:TSINGHUA UNIVERSITY

Method and system for optimizing management model of video image shot by unmanned aerial vehicle

The invention discloses an optimization method for a management model of a video image shot by an unmanned aerial vehicle, and the method comprises the steps: monitoring the flight state and environment conditions of the unmanned aerial vehicle in real time according to the video image and data, carrying out the preliminary analysis of the video image, and carrying out the dynamic range adjustment based on a preliminary analysis result; synthesizing a video image through a software algorithm, and analyzing the image through a noise recognition algorithm; unmanned aerial vehicle flight parameters are optimized according to the image quality feedback, flight path planning parameters are set according to the target shooting area and the expected image quality requirement, and the flight path is dynamically adjusted by using the image quality feedback; and performing secondary evaluation, automatically adjusting shooting parameters of the camera, and processing the image based on the content identified by the model. The method can solve the problems that a traditional static and non-adaptive method is large in image quality fluctuation and poor in environment adaptability.
Owner:STATE GRID JIANGSU ELECTRIC POWER CO XUZHOU POWER SUPPLY CO

Lithium ion battery capacity attenuation estimation method, electronic equipment and storage medium

The invention discloses a lithium ion battery capacity attenuation estimation method, electronic equipment and a storage medium, belongs to the technical field of lithium ion battery capacity attenuation estimation, and solves the problem of how to improve the estimation precision of lithium ion battery capacity attenuation. According to the method, a lithium ion battery experience degradation model is corrected by considering the uncertainty of factors such as temperature, discharge depth and cycle index, a lithium ion battery capacity attenuation disturbance model is constructed, the health characteristics of battery capacity attenuation are accurately extracted by a data driving method, a data driving error compensation model based on a convolutional neural network is established, and the accuracy of battery capacity attenuation is improved. Through fusion of a battery capacity attenuation disturbance model and a data driving error compensation model, evaluation is carried out in a model driving mode, evaluation errors are inhibited in a data driving mode, and capacity attenuation information of the lithium ion battery is accurately predicted. The method also has good robustness and applicability for the capacity fading conditions of different batteries of the same type, and provides reliable technical support for evaluation of different types of batteries.
Owner:NANJING UNIV OF SCI & TECH

Crop growth vigor intelligent sensing algorithm and system, computer equipment and medium

The invention discloses an intelligent sensing algorithm and system for crop growth vigor, computer equipment and a medium, and the algorithm comprises the following steps: 1, carrying out the image collection of a farmland, and collecting the spectral data of different wavebands in the farmland; 2, preprocessing the collected images and spectral data, and extracting multi-dimensional feature information of the crops; 3, fusing the extracted multi-dimensional feature information of the crops to obtain comprehensive and accurate crop growth feature information; 4, performing data analysis on the crop growth feature information through deep learning, and constructing an intelligent decision model; through the implementation of the invention, the characteristics and beneficial effects of multi-information fusion, high-precision perception, cost reduction, quality improvement, wide application and the like are realized through the crop growth vigor intelligent perception algorithm, and the method plays an important role in improvement of agricultural production and agricultural product quality.
Owner:HOUJI SHUNONG (HANGZHOU) TECH CO LTD

Power equipment fault diagnosis method based on time sequence-spectrogram-discrete data multi-modal fusion

The invention relates to a power equipment fault diagnosis method based on time sequence-spectrogram-discrete data multi-modal fusion, and belongs to the field of power equipment fault diagnosis. A multi-modal feature fusion technology based on an attention mechanism is adopted, and a power equipment fault diagnosis model based on multi-modal information fusion is constructed. Compared with a traditional deep learning fault diagnosis method, the problem that a fault single-modal sample is insufficient in diagnosis model accuracy can be solved through multi-modal data fusion, and the universality and diagnosis precision of the model are improved.
Owner:STATE GRID FUJIAN ELECTRIC POWER RES INST +2

Pallet protection corner automatic putting machine

The invention discloses a pallet protection corner automatic putting machine. The pallet protection corner automatic putting machine comprises a top protection corner putting combination machine and aside face protection corner putting combination machine; the top protection corner putting combination machine comprises a bottom circulation base and a pair of top protection corner single machines,the side face protection corner putting combination machine comprises a bottom fixed base and four side face protection corner single machines, the top face protection corner single machines are provided with electrical control parts, top protection corner containing grooves, limiting mechanisms, and top protection corner mounting claws, the top protection corner mounting claws grab protection corners in the top protection corner containing grooves, and put the protection corners on four edges of the top of a packaging box, each side face protection corner single machine comprises a second electrical control part, a side protection corner containing groove, and a side protection corner mounting claw, and the side protection corner mounting claws grab the protection corners in the side protection corner mounting grooves and put the protection corners on the four sides of the side face of the packaging box. The pallet protection corner automatic putting machine is reasonable in layout,smooth in operation, high in adaptation, automatic in full process, free of artificial duty and capable of greatly improving protection corner putting and threading packaging capacity.
Owner:SUZHOU YUEHENG AUTOMATION TECH CO LTD

Fine-grained large model illusion processing method and system based on zero resource framework

The invention belongs to the technical field of natural language processing. According to the zero resource framework-based fine-grained large model illusion processing method and system, multi-dimensional anomaly detection is carried out on a preprocessed text through a large language model, and a multi-dimensional anomaly detection result is obtained; performing anomaly correction priority ranking according to a multi-dimensional anomaly detection result; dynamically adjusting a context window range according to a syntactic analysis result and the generated syntactic dependency tree; performing priority correction on high-risk vocabularies and sentences according to an abnormal correction priority ranking result in combination with the adjusted context window range; and evaluating the corrected content, and obtaining a corrected final text when a termination condition is met. According to the method, the excessive confidence, subtle semantics and grammar deviation illusion detection of the detection content is realized, and the illusion detection processing capability of a large language model, especially the detection processing capability of a fine-grained illusion problem, is improved.
Owner:SHANDONG COMP SCI CENTNAT SUPERCOMP CENT IN JINAN +3

Target person voice extraction method and device fusing voice and electroencephalogram signals

The invention belongs to the technical field of voice processing, and particularly relates to a target person voice extraction method, system and device fusing voice and electroencephalogram signals. According to the method, a voice separation model comprising an electroencephalogram encoder, a voice encoder, a separation network and a voice decoder is designed, and a voice extraction task of a target person is executed by using the trained voice separation model. Wherein the electroencephalogram encoder comprises a one-dimensional convolutional layer and an attention layer based on KAN. The voice encoder comprises a one-dimensional convolution layer and a bidirectional Mama layer. And the separation network adopts a cross attention module to perform feature fusion on the extracted voice features and electroencephalogram features, and analyzes the obtained fusion features to obtain a feature mask of the voice of the target person. And the voice decoder reconstructs the pure voice of the target person by using the feature mask and the voice features. According to the method, the problems that deep electroencephalogram features and global voice features are difficult to effectively capture and the precision of the reconstructed voice of the target person is poor in the existing scheme are solved.
Owner:ANHUI UNIV