Patents
Literature
187results about How to "Fast iteration" 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
Multi-scale adaptive time sequence InSAR (Interferometric Synthetic Aperture Radar) earth surface deformation extraction method
PendingCN119415833AEasy to implementgood effectRadio wave reradiation/reflectionInterferometric synthetic aperture radarData set
The invention discloses a multi-scale adaptive time sequence InSAR (Interferometric Synthetic Aperture Radar) earth surface deformation extraction method, and relates to the field of remote sensing signal processing. The method comprises the following steps of: constructing an unwrapped time sequence InSAR phase simulation data set, performing pixel-by-pixel time sequence empirical mode decomposition on a time domain, obtaining a time sequence intrinsic mode function at the same pixel position, and constructing a FastICA wavelet enhancement denoising algorithm on a space domain: performing two-dimensional multi-stage wavelet decomposition on the intrinsic mode function of each dimension by taking a power spectrum entropy as a cut-off condition, the method comprises the following steps of: acquiring a wavelet approximation coefficient of each level and a detail coefficient, performing noise reduction and dimension reduction processing and reconstruction on the wavelet approximation coefficient and the detail coefficient of each level, and finally performing wavelet reconstruction on the updated wavelet coefficient to obtain a processed time sequence InSAR phase simulation data set. The method provided by the invention is based on data self-driving, self-adaptive decomposition is carried out on phase signals from two scales of time and space, earth surface deformation related signals can be effectively captured, other phase signals can be separated and explained, and a scientific theoretical basis is provided for high-precision earth surface deformation monitoring.
Owner:CHINA UNIV OF MINING & TECH
Hysteresis loop prediction model establishment method and device, electronic equipment and storage medium
ActiveCN119670538Aimprove accuracyImprove generalization abilityGeometric CADBiological modelsOverfittingk-nearest neighbors algorithm
The invention provides a hysteresis loop prediction model establishment method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out a hysteresis characteristic experiment on a ferromagnetic material, setting hysteresis loop related characteristics comprising magnetic induction intensity waveform data and magnetic field waveform data under the excitation of preset sine waves under different magnetic induction intensity peak values, establishing a hysteresis loop prediction model comprising an improved convolutional neural network model and a K-nearest neighbor algorithm model, training the improved convolutional neural network model, inputting a magnetic induction intensity peak value and magnetic induction intensity waveform data, and outputting magnetic field waveform data; and adjusting the magnetic field waveform data based on the feature data by using a K-nearest neighbor algorithm model to obtain a final magnetic field waveform result. According to the method, the hysteresis loop prediction model is established based on the improved convolutional neural network in combination with the K nearest neighbor algorithm, the overfitting phenomenon can be prevented, the training speed is higher, the resource consumption is lower, and rapid iteration and adjustment in practical application are facilitated.
Owner:GUANGDONG POWER GRID CO LTD DONGGUAN POWER SUPPLY BUREAU +2
Cloud platform rapid deployment system and method
InactiveCN119396422ASimplify the deployment processReduce deployment timeVersion controlSoftware deploymentManagement toolAttack
The invention discloses a cloud platform rapid deployment system and method. The system comprises a Docker, a container arrangement tool, a database service, a cloud platform application, a monitoring and management tool, a network infrastructure and a security component. The Docker is responsible for managing the complete life cycle of the container and providing a virtualization environment; the container arrangement tool is used for managing and arranging a Docker container; the database service is used for storing information in the cloud platform and carrying out communication through the network service of the Docker; the cloud platform application is a set of service logic and functions deployed in a Docker container; the monitoring and management tool is used for monitoring a running state and managing log information; the network infrastructure is used for ensuring network interconnection between the servers and supporting network communication of the Docker container; and the security component is used for protecting the cloud platform from external attacks and internal threats. According to the method, the deployment process can be simplified, the deployment time is shortened, manual intervention is reduced, the deployment efficiency is improved, resources are optimized, and the operation cost is reduced.
Owner:TONGLING VOCATIONAL & TECH COLLEGE
Tool calling semantic analysis method and system for low-resource data environment
ActiveCN119623478AImprove analytical abilityImprove performanceSemantic analysisMachine learningLinguistic modelData set
The invention discloses a tool calling semantic analysis method and system for a low-resource data environment, and relates to the technical field of semantic analysis, and the method comprises the following steps: constructing a logic expression of each tool based on a context-independent grammar, and generating initial training data in combination with a cue word project and a large language model; performing iterative training on a pre-configured semantic analysis model according to the obtained training data in combination with a self-training method to obtain an optimized semantic analysis model; and performing similarity calculation on the tool candidate data and the training data through a recall model to obtain vector scores of the tool candidate data and the training data, and screening out tools which do not conform to user semantics to obtain a final semantic analysis model. According to the method, the initial analysis data set based on the context-independent grammar is constructed, and the analysis capability of the model is enhanced in combination with the self-training strategy, so that the performance of the model in a low-resource environment is improved, and the semantic analysis capability is effectively expanded.
Owner:UNIV OF SCI & TECH OF CHINA
Aero-engine vibration imbalance phase testing device and testing method
PendingCN119394660ASimple structureNot easy to interfereGas-turbine engine testingSustainable transportationConvertersRotor (electric)
The invention provides an aero-engine vibration phase testing device. The aero-engine vibration phase testing device comprises an optical fiber sensor, a photoelectric converter, a triaxial accelerometer, a vibration analyzer and a test computer. The vibration analyzer is responsible for collecting signals and performing related processing on the signals through a built-in data processing module; and the test computer is connected with the vibration analyzer through the Ethernet to realize data real-time transmission and program end data processing, and is accessed to a test run data monitoring interface. The invention discloses an aero-engine vibration imbalance phase test method. The method comprises the following steps: marking key phase blades; preparation before installation of the optical fiber sensor; installing an optical fiber sensor; performing photoelectric conversion; a triaxial accelerometer is installed; picking up a rotor key phase positioning signal; solving a vibration imbalance phase under multiple rotating speeds; and carrying out counter weight adding test run verification. The method has the advantage that the unbalanced phase is efficiently and accurately obtained. The reliability and accuracy of vibration phase test data are improved, and the follow-up engine balancing quality is further improved.
Owner:SHENYANG LIMING AERO-ENGINE GROUP CORPORATION
Dynamically updated byte code security protection method, system and program product
PendingCN119312344ARealize dynamic adjustmentquick responsePlatform integrity maintainanceSoftware engineeringLogisim
The invention provides a dynamically updated byte code security protection method, which comprises the following steps of: loading an agent program when an application program is started, acquiring and caching an instrumentation object, receiving security detection logic and instrumentation configuration information issued by a management end by the agent program through a communication mechanism with the management end, and storing the security detection logic and the instrumentation configuration information in the application program; and the agent program translates the instrumentation logic issued by the management end into byte codes by using the byte code operation library, generates a class file conversion object containing the instrumentation logic, registers the class file conversion object into an instrumentation object, and calls a class re-conversion method of the instrumentation object to perform byte code updating on a target class, thereby realizing dynamic insertion of the instrumentation logic. And the updated byte code takes effect immediately, and security vulnerabilities in the target method are detected and protected in real time. According to the method, the capability of quickly responding to the 0Day vulnerability and blocking the 0Day vulnerability in real time is realized, and the time for updating the protection logic is greatly shortened.
Owner:CTRIP TRAVEL NETWORK TECH SHANGHAI0
AGV obstacle avoidance path planning method in unmanned warehousing system
ActiveCN119533503Agood varietyrich diversityInstruments for road network navigationControl engineeringControl theory
The invention discloses an AGV obstacle avoidance path planning method in an unmanned warehousing system, and the method comprises the steps: employing nested Logistic mapping and Skew Tent mapping to generate a chaotic initialization population to replace a population generation mode of a conventional differential evolution algorithm, carrying out the Levy flight operation after the selection operation, and carrying out the global path search on an environment map, thereby achieving the optimal path planning. And obtaining an AGV global feasible path avoiding obstacles in the environment map, and when the AGV has an unknown obstacle on the feasible path, re-planning the local path by adopting a dynamic window method so as to enable the AGV to realize an obstacle avoidance process in an interval where the unknown obstacle is located. The improved differential evolution algorithm enhances the global search capability through chaos initialization and a Levy flight strategy, and the AGV can quickly respond to an unknown obstacle through a dynamic window rule. The adaptability and the accuracy of path planning are improved, and the obstacle avoidance capability and the safety of the AGV in a complex environment are enhanced at the same time.
Owner:MASCH TECH DEV CO LTD
Rescue unmanned aerial vehicle path planning method based on support vector machine and genetic algorithm
ActiveCN119249914Aimprove securityImprove stabilityInternal combustion piston enginesKernel methodsAlgorithmCollision analysis
According to the rescue unmanned aerial vehicle path planning method based on the support vector machine and the genetic algorithm provided by the invention, the support vector machine and the genetic algorithm are combined, and the collision risk of the flight path of the unmanned aerial vehicle is quantitatively evaluated and graded by using the multi-classification support vector machine; the support vector machine model is iteratively optimized by using the genetic algorithm, collision analysis and evaluation can be performed on the simulated rescue path, the result is scientific and reasonable, the safety and stability of unmanned aerial vehicle rescue are effectively improved, and the precision and quality of path planning are improved; besides, a traditional genetic algorithm is improved, the path is planned by using the improved genetic algorithm, and risk identification is performed on the planned path in combination with the trained support vector machine model, so that the path quality is further improved, and high-precision and high-efficiency path planning is realized. Compared with a traditional rescue line constructed by a single weight and a model, the method has more advantages.
Owner:GUANGDONG URBAN & RURAL PLANNING & DESIGN INST
Low-code platform development method and low-code platform system
PendingCN119556903AImprove development efficiencyshorten the timeVisual/graphical programmingSoftware engineeringSoftware development
The invention discloses a low-code platform development method and a low-code platform system, and belongs to the technical field of software development, the overall architecture design of a low-code platform is carried out based on Vue2 and TypeScript, and the method comprises the combination of modularization, data management and visual editing function modules; the system comprises a component management module which comprises a component library and user-defined component management; the visual editor is used for realizing real-time layout adjustment and attribute configuration; the user previews the modification effect in real time by dragging and adjusting the position, the size and the hierarchical relationship of the component; the data management module is used for realizing data binding, API integration and dynamic data management; and the performance optimization module comprises a virtual DOM (Document Object Model) optimization and on-demand loading mechanism. The invention provides a development environment which is high in expandability and easy to use. A complex coding process is simplified into visual operation, and cooperation between developers and non-technical users is promoted.
Owner:INSPUR SOFTWARE TECH CO LTD
Multi-dimensional joint closed-loop optimization positioning method based on channel multi-domain characteristic digital twinning
PendingCN119596232Afast iterationPosition fixationComplex mathematical operationsMathematical modelMesh grid
The invention discloses a multi-dimensional joint closed-loop optimization positioning method based on channel multi-domain characteristic digital twinning, and aims to improve the positioning precision in multi-path propagation and non-line-of-sight propagation environments. A multi-path received signal calculation method is given by analyzing the components of a multi-path signal and considering the time delay, attenuation and phase change of each path. Multi-domain decomposition is carried out on a received signal by adopting wavelet transform, features are extracted from a time domain, a frequency domain and a Doppler domain, and a waveform difference degree calculation method based on multi-domain features is defined. A multi-base-station signal fusion positioning mathematical model based on channel multi-domain characteristic digital twinning is constructed, and compared with an existing researched single-base-station closed-loop positioning method, the multi-dimensional joint closed-loop positioning is realized by arranging a plurality of base stations in the physical world. A posterior probability is deduced by combining the comprehensive difference degree of a plurality of base stations with Bayesian inference, in posterior probability optimization, a dynamic optimization mechanism is designed, Newton method rapid iterative optimization is proposed according to a hardware resource selection optimization algorithm and a low-computing-performance environment, and optimization is realized by adopting a grid search method in a high-computing-performance environment. And accurate estimation of the position of the to-be-positioned node is realized.
Owner:BEIJING TECH & BUSINESS UNIV
Drug effect simulation system
PendingCN119132645Aimprove accuracyIncreased sensitivityMedical simulationDrug and medicationsAnalogue computationStatistical analysis
The invention relates to the field of drug effect simulation, in particular to a drug effect simulation system, which comprises a data input module for receiving drug information and patient information and carrying out structured storage on the information; the data preprocessing module is used for performing cleaning, standardization and feature extraction on the input data; the intelligent model selection module automatically selects the most suitable PK and PD models through a machine learning algorithm in combination with historical data; the personalized parameter adjusting module is used for dynamically adjusting simulation parameters according to the specific data of the patient; the drug effect simulation module is used for simulating the action process of the drug in the virtual environment and adjusting the simulation in real time through the real-time feedback module; the calculation module is used for executing simulation calculation and generating a drug effect prediction result; and the result analysis module is used for carrying out statistical analysis on the simulation result and generating a drug effect prediction curve and a detailed report.
Owner:BEIJING FAZERUI TECHNOLOGY INFORMATION CO LTD
Generative model reasoning service deployment method and device based on AI questions and answers
PendingCN119293154AImprove efficiencyImprove convenienceDigital data information retrievalInference methodsModel sampleTheoretical computer science
The invention discloses a generative model reasoning deployment method and device based on AI questions and answers, and the method comprises the steps: constructing the basic configuration of an AI questions and answers system, and carrying out the model deployment; designing a database and storing model metadata, a model example code, a GPU (Graphics Processing Unit) configuration guide and a Dockerfile script template; receiving a user reasoning service request, generating a deployment task ID and recording task information; performing model existence check, and if the model is missing, automatically downloading model metadata according to a stored Git warehouse address; extracting corresponding configuration data and example code details from the database, and dynamically filling a script template with the configuration data and example code details to generate a Dockerfile script; executing a mirror image starting script constructed in the container, running a mirror image and returning a service address; according to the invention, the efficiency and convenience of model service deployment are improved, the threshold of AI technology application is reduced, the resource configuration is optimized, and the user experience is enhanced.
Owner:DMAI (GUANGZHOU) CO LTD +1
TBM tunnel support parameter optimization method based on mechanical response prediction
ActiveCN119323071Aimprove rationalityImprove targetingGeometric CADDesign optimisation/simulationClassical mechanicsFinite element computation
The invention discloses a TBM (Tunnel Boring Machine) tunnel support parameter optimization method based on mechanical response prediction, and relates to the technical field of tunnel construction, and the method is characterized in that a large number of mechanical response samples are generated by establishing finite element calculation models under different surrounding rock states and support parameter combinations; meanwhile, field measured data are collected for supplement and correction, a surrounding rock-support mapping relation model is established by utilizing a deep learning algorithm, and an optimal support parameter combination meeting safety requirements is intelligently selected according to the relation between specific surrounding rock conditions and support responses, so that accurate optimization of support design is realized.
Owner:SOUTHWEST JIAOTONG UNIV
DSP acceleration implementation method and device for slender matrix indirect singular value decomposition based on two-norm constraint
PendingCN119149888AImprove computing efficiencyreduce data volumeComplex mathematical operationsSingular value decompositionData class
The invention relates to a DSP acceleration implementation method and device for indirect singular value decomposition of a two-norm constrained slender matrix. The method comprises the following steps: obtaining a to-be-orthogonalized matrix of an initial data type according to an input matrix; the method comprises the following steps: performing orthogonalization processing on a to-be-orthogonalized matrix by utilizing Jacobi rotation transformation through a DSP (Digital Signal Processor), and when a second-order norm of a column vector in the processed to-be-orthogonalized matrix meets a preset condition, converting a data type of the processed to-be-orthogonalized matrix from an initial data type into a target data type, and performing orthogonalization processing again, performing zero setting on a relaxation factor corresponding to the to-be-orthogonalized matrix after the current orthogonalization processing, and continuing the orthogonalization processing until a convergence condition is met to obtain an iterated matrix; and obtaining a left singular matrix, a right singular matrix and a singular value of the input matrix according to the iterated matrix. According to the method, the to-be-orthogonalized matrix is used for processing, the calculation amount is reduced, the iteration speed is higher, and the calculation speed is increased.
Owner:XIDIAN UNIV
Rule engine and related product
PendingCN119539063Aefficient executioneasy maintenanceKnowledge based modelsCode compilationProgramming languageA domain
The invention provides a rule engine and a related product. The rule engine comprises a rule set configuration module, a rule set management module, a domain model design module, a rule execution report module, an engine execution result module, a rule factory module, a rule context module, a rule executor and a dynamic rule adding module. Compared with the prior art, the rule engine provided by the invention has the advantages that simple definition and efficient execution of the rule are realized by utilizing the advantages of the Spring framework. Compared with an existing rule engine based on XML or a specific rule language, the rule engine has the advantages that a developer is allowed to use the SpEL expression to directly define the rule, and the efficiency of the rule engine is improved.
Owner:UNICLOUD TECH CO LTD
Block alignment multi-target trajectory optimization method based on IWOA
ActiveCN119398293Afast iterationHigh precisionInternal combustion piston enginesForecastingAlgorithmControl engineering
The invention provides a block alignment multi-target trajectory optimization method based on IWOA. The block alignment multi-target trajectory optimization method comprises the steps of 1, determining a ship block trajectory planning mode; 2, dynamic track parameters which need to be followed by the attitude adjusting trolley in different directions are obtained; 3, multi-target trajectory optimization of the block and the attitude adjusting trolley is achieved; 4, planning the attitude adjustment of the block, solving the optimal time within the constraint range, and solving the motion trail of the attitude adjustment trolley; 5, establishing a posture adjusting trolley block alignment model; and step 6, verifying the feasibility of block trajectory multi-objective optimization and the accuracy of block alignment trajectory planning. The method has the beneficial effects that efficient and accurate automatic alignment of the ship block is realized through automatic alignment system building, high-precision measurement system building, coordinate transformation model building, point set matching algorithm building, intelligent optimization algorithm building, trajectory planning algorithm building and ship block automatic alignment trajectory planning system building.
Owner:JIANGSU UNIV OF SCI & TECH
Framework model based on neural network learning and application method thereof
PendingCN119361026Aefficient analysisImprove generalization abilityChemical property predictionMolecular designData setModel testing
The invention relates to the technical field of medical science, in particular to a framework model based on neural network learning and an application method of the framework model. According to the model, a graph convolutional network (GCN, Transform architecture and long short-term memory network (LSTM) are fused, a multi-level analysis process is constructed, the spatial structure and sequence dynamics of drug molecules are deeply explored, and the model comprises the following specific steps: model training: performing model training by adopting data sets collected by people such as Cui, Q.Cui2020Improved and the like; the model test selects three independent test sets to comprehensively evaluate the performance of the model; according to the data preprocessing, an RDKit library is used, SMILES character strings are converted into molecular objects, a feature tensor is constructed and used for storing core features of each molecule in detail, data splitting is carried out, and the data splitting adopts versions, such as Ahmad, divided by people based on a Cui2020 data set to carry out training and testing of a machine learning model.
Owner:王晨旭
Permanent magnet synchronous wind driven generator parameter identification method based on improved grey wolf algorithm
PendingCN119298752ADiversity guaranteedfast iterationElectronic commutation motor controlVector control systemsLocal search (optimization)Local optimum
The invention discloses a permanent magnet synchronous wind driven generator parameter identification method based on an improved grey wolf algorithm, and the method comprises the steps: carrying out the optimization of an initial population through employing Cat chaotic mapping and a reverse learning strategy, enabling the traversal to be uniform, achieving a higher iteration speed, introducing a nonlinear convergence factor of cosine change, and strengthening the global search in an earlier stage, thereby achieving the recognition of the parameters of a permanent magnet synchronous wind driven generator. The local search is enhanced in the later stage, the diversity of populations is ensured, the search range is enlarged, the global search capability and the local search capability are balanced, the probability of falling into local optimum is reduced, the identification speed is accelerated, then the tube voltage drop nonlinearity of the voltage source type inverter is compensated, the voltage error between the output voltage and the reference voltage of the inverter is reduced, and the identification accuracy is improved. And finally, learning is carried out by combining memory self-learning and the individual experience of the grey wolf, so that the problems of low parameter identification precision, easiness in falling into local optimum and low speed of the parameter identification method of the permanent magnet synchronous motor are solved.
Owner:HUNAN RAILWAY PROFESSIONAL TECH COLLEGE
Dynamic stiffness performance analysis method for vehicle all-in-one motor assembly
PendingCN119394652Aeffective assessmentReduce test costsVehicle testingEngine testingDynamic stiffnessElectric machine
The invention discloses a dynamic stiffness performance analysis method for an all-in-one motor assembly for a vehicle, sub-components of the all-in-one motor assembly comprise a controller upper cover plate, a controller lower cover plate and a rotary transformer cover plate, and the method comprises the following steps: 1) carrying out frequency domain response tests on the sub-components of the all-in-one motor assembly; 2) modal frequency identification of each sub-component of the all-in-one motor assembly; 3) performing frequency domain response test on each suspension mounting point of the all-in-one motor assembly; 4) calculating the dynamic stiffness of each suspension mounting point of the all-in-one motor assembly; 5) performing a frequency domain response test of the all-in-one motor assembly; 6) identifying the modal frequency of the all-in-one motor assembly; and 7) calculating and analyzing the rigidity performance of the all-in-one motor assembly. According to the invention, rapid iteration of the all-in-one motor assembly can be realized on the premise of no rack and whole vehicle resource support; the purposes of shortening the development period, recognizing risk problems in advance and saving the test cost of the whole vehicle are achieved.
Owner:CHONGQING TSINGSHAN IND
Cartoon video generation method and device and storage medium
PendingCN119603527AIncreased manufacturing flexibilityReduce labor burdenSelective content distributionComputer graphics (images)Imaging data
The invention relates to the field of video generation, and discloses a cartoon video generation method, cartoon video generation equipment and a storage medium. The method comprises the steps of receiving a text making instruction; judging whether the text making instruction is an instruction for making a specified neural network or not; if yes, inputting the theme content setting, the story background setting and the role information setting into a preset AI text generation model to generate a story text; judging whether the story text limits the image to maintain the original text; when the story text is not limited, performing scene division and retouching processing on the story text to generate scene description; when the story text is limited, performing division processing on the original story text to generate scene description; performing drawing processing on the scene description to generate image data; dubbing the story text to generate dubbing data; and combining the dubbing data and the image data to generate an audio cartoon video. In the embodiment of the invention, the neural network is utilized to draw images and make dubbles for the story text, and the picture book cartoon video is efficiently generated at low cost.
Owner:SHENZHEN MAIFENG TECH CO LTD
Decentralized micro-front-end construction method and device
PendingCN119088347AImprove collaboration efficiencyFacilitate division of labor and cooperationSoftware maintainance/managementProgram loading/initiatingIndustrial engineeringServer
Owner:CHINA CONSTRUCTION BANK +1
Capacitive micro-machined ultrasonic transducer based on 3D printing and preparation method of capacitive micro-machined ultrasonic transducer
ActiveCN119114405Areduce wastereduce consumptionAdditive manufacturing apparatusCeramic shaping apparatusVibrating membraneCapacitive micromachined ultrasonic transducers
The invention belongs to the field of material and electronic device manufacturing, and discloses a capacitive micromachined ultrasonic transducer based on 3D printing and a preparation method thereof.The CMUTs comprise an upper electrode, an upper insulating layer, a vibrating membrane, a supporting column layer, a silicon substrate and a lower electrode which are sequentially connected from top to bottom, and a cavity is formed among the vibrating membrane, the lower insulating layer and a supporting column of the supporting column layer; the method comprises the following steps: printing a pillar layer array on the front surface of a silicon substrate with a functionalized surface by utilizing a nano silicon dioxide composite material and adopting a photocuring 3D printing method; then curing and heat treatment are carried out to obtain a second device; bonding a vibrating diaphragm; printing an insulating layer on the front surface of the vibrating diaphragm; carrying out curing and heat treatment; printing an upper electrode; removing the silicon dioxide on the back surface of the silicon substrate through a wet etching process; printing a lower electrode on the back of the silicon substrate; and removing the photoresist. According to the invention, the consumption of required equipment and materials is reduced, and the production cost is reduced.
Owner:XI AN JIAOTONG UNIV
Driving behavior analysis and evaluation method and system
The invention discloses a driving behavior analysis and evaluation method. The method comprises the following steps: segmenting a vehicle travel according to a preset travel segmentation logic; obtaining multi-source data generated by the vehicle in the current split journey, and performing fusion processing; a driving behavior analysis model corresponding to the current vehicle model is obtained, the driving behavior analysis model comprises a multi-dimensional calculation formula for driving behavior evaluation, the fused multi-source data is calculated, and scores of all dimensions and evaluation results of driving behaviors are obtained; and outputting the score of each dimension and the evaluation result of the driving behavior. The invention further provides a corresponding system. By implementing the method, the efficiency, comprehensiveness and accuracy of driving behavior analysis and evaluation can be improved, the cost is reduced, and the iteration speed is increased.
Owner:GUANGZHOU AUTOMOBILE GROUP CO LTD
Engineering potential safety hazard detection method based on multi-modal large model
PendingCN121921716AOvercome the problem of inaccurate positioningOvercome core flawsCharacter and pattern recognitionBiological modelsPattern recognitionData pack
The invention relates to an engineering potential safety hazard detection method based on a multi-modal large model. The method comprises the following steps: acquiring an on-site monitoring image of a target scene; identifying a key entity in the field monitoring image, and extracting positioning information of the key entity; generating a depth map based on the on-site monitoring image; based on the positioning information of the key entity, combining with the depth map, associating the positioning information and the depth information for the key entity, and generating a scene description text; based on the on-site monitoring image, the depth map corresponding to the on-site monitoring image and the scene description text, constructing a multi-modal input packet; inputting the multi-modal input packet into a learned engineering safety domain knowledge base and a visual language large model subjected to instruction fine tuning, and outputting a hidden danger judgment result; the instruction fine tuning of the visual language large model is based on a hidden danger inference chain data set, the data set has hidden danger images corresponding to hidden danger items in an engineering safety field knowledge base and annotation data of each hidden danger image, and the annotation data comprises step result data corresponding to each hidden danger detection step.
Owner:POWERCHINA HUADONG ENG CORP LTD +1
High-precision anti-vibration large-aperture mechanical phase-shifting interferometry method and device
ActiveCN119086019AContributes to high-precision calculationssuppress interferenceUsing optical meansTesting optical propertiesPrincipal component analysisEngineering
The invention discloses a high-precision anti-vibration large-aperture mechanical phase-shifting interference measurement method and device, and belongs to the technical field of optical precision measurement. The implementation method comprises the following steps: acquiring a phase shift initial value through a principal component analysis method; calculating background light intensity and modulation degree parameters of the interferogram to obtain an initial value of an iterative algorithm; by establishing an environment vibration error model, the capability of suppressing noise such as measured environment vibration and the like is improved; the iterative linear regression algorithm is combined with the initial estimation of the measured surface shape, and the space and time variables are iteratively solved respectively, so that the response speed and the measurement precision of the system are improved. The method has the advantages of being high in convergence speed and robustness, the initial value of the measured surface shape, the modulation degree and background noise are obtained, the noise model changing along with time is constructed, suppression of environment noise is achieved, and high-precision measurement of the surface shape parameters of the large-aperture optical element is achieved.
Owner:BEIJING INST OF TECH
Application release management system and application release management method based on credential environment
PendingCN119397550Ahigh speedquality improvementVisual/graphical programmingPlatform integrity maintainanceSoftware engineeringInteraction interface
The invention relates to an application release management system and an application release management method based on a credential environment, and the system comprises an infrastructure layer which is used for constructing the credential environment according to hardware resources; the publishing system platform layer is used for deploying at least one publishing platform, and the publishing platform is used for publishing an application; the presentation layer is used for providing a user interaction interface and acquiring configuration information of a user; the third-party tool layer is used for providing a release mirror image; the application layer is used for executing an application publishing task according to the publishing platform, the configuration information and the publishing mirror image, and the application layer comprises one or any combination of a login verification module, a publishing window management module, an application module management module, a publishing peripheral resource management module, an application service management module and a publishing configuration management module; and the logic layer is used for carrying out logic abstract packaging on the business of each module in the application layer. According to the invention, applications can be published and managed efficiently and safely in a unified manner.
Owner:太保科技有限公司
Water quality anomaly detection feature selection method and system based on improved pigeon inspired optimization algorithm
ActiveCN119669716AAvoid local suboptimal solutionsfast convergenceGeneral water supply conservationArtificial lifeAnomaly detectionSigmoid function
The invention relates to a water quality anomaly detection feature selection method and system based on an improved pigeon inspired optimization algorithm, and belongs to the field of machine learning. Executing an improved pigeon flock algorithm, and evaluating the pigeon flock; endowing the position and speed of the pigeon with the lowest fitness to the globally optimal pigeon; the speed of the pigeons is updated, the speed of the pigeons is converted through a Sigmoid function, and the positions of the pigeons are updated according to the output value of the Sigmoid function; when an iteration stopping condition is met, returning to a global optimal pigeon; otherwise, mutating the pigeon flock, and updating the position and the speed through the literanism; ranking the varied pigeons according to fitness values, calculating an ideal destination, and updating the positions of the pigeons; according to the method, the self-adaptive iterative map and the compass factor are set, and meanwhile, an Litalian mechanism is introduced, so that information interaction and cooperation can be carried out among pigeon individuals, the common evolution of a group is promoted, and the search capability of the group is further optimized.
Owner:QILU UNIVERSITY OF TECHNOLOGY (SHANDONG ACADEMY OF SCIENCES) +1
Net size and shape collaborative manufacturing method for special-shaped curved surface inner runner of engine
ActiveCN119294003AGuaranteed accuracyfast iterationGeometric CADDesign optimisation/simulationMechanical engineeringMesh grid
The invention relates to a net size shape collaborative manufacturing method of an engine special-shaped curved surface inner runner, which comprises the following steps: carrying out parting design according to a theoretical design molded surface of an engine special-shaped curved surface runner structure to obtain an upper parting and a lower parting of the runner; the upper parting part and the lower parting part of the runner are unfolded to form a net-size flat plate with a plurality of plane grid units; carrying out hot press molding numerical analysis to obtain node displacement of unit nodes in the plane grid units under hot press molding; mapping a reconstructed molded surface of the engine special-shaped curved surface runner structure based on unit nodes in the plane grid units and node displacement inverse conversion, judging a molded surface error between the reconstructed molded surface and a theoretically designed molded surface, and outputting a net-size flat plate if the molded surface error meets a preset requirement; carrying out hot press molding on the flat plate test piece by adopting a net size molding mold to obtain a corresponding runner structure; and the upper flow channel structure and the lower flow channel structure are welded to complete net size shape collaborative manufacturing of the engine special-shaped curved surface flow channel structure.
Owner:NAT UNIV OF DEFENSE TECH
Dynamic reflection coefficient adjusting method for double intelligent reflecting surfaces based on deep reinforcement learning
ActiveCN119420384AadaptableImprove environmental adaptabilitySpatial transmit diversityBiological modelsCommunications systemSimulation
The invention discloses a method for adjusting dynamic reflection coefficients of double intelligent reflecting surfaces based on deep reinforcement learning. At present, the complexity and dynamic change of the environment are important factors influencing intelligent reflecting surface phase regulation and control, and long-term phase regulation and control are difficult in the dynamic and complex environment. According to the method, on the basis of deep reinforcement learning (DRL), continuous adjustment is carried out through feedback of environment information, an optimal strategy is achieved, and adjustment of the phase and the amplitude of the double-intelligent-reflection-surface unit is carried out. According to the method, a flexible, efficient and automatic method can be provided for a double-intelligent-reflection-surface wireless communication system to achieve reflection coefficient optimization control, especially in the face of constantly changing environments and complex system dynamic states. In this way, the energy efficiency ratio of the wireless network can be improved, the coverage range can be expanded, and the user experience can be improved. In addition, the convergence rate is improved by carrying out distributed training on the method provided by the invention. Simulation results show that the method of the invention is effective and feasible, and the method of the invention has a fast convergence rate.
Owner:ZHEJIANG UNIV
Rolling bearing weak fault diagnosis method based on improved sparrow algorithm
PendingCN119622300Afast iterationImprove search abilityMachine part testingArtificial lifeFeature extractionWavelet decomposition
The invention discloses a rolling bearing weak fault diagnosis method based on an improved sparrow algorithm. The method comprises the following steps: firstly, optimizing parameters alpha and k in the VMD by using a sparrow optimization algorithm (OCSSA) fusing eagle and Cauchy variation; decomposing the signal by using the optimized VMD, and screening components meeting conditions as optimal components by taking EWK as an evaluation index; secondly, multi-scale morphological energy entropy (WMMEE) based on wavelet decomposition serves as a fitness function to optimize MCKD so as to enhance fault features, and envelope spectrum analysis is adopted to extract actual fault feature frequency to achieve fault diagnosis; under the action of the two fitness functions, the OCSSA optimization algorithm is used for optimizing hyper-parameters in VMD and MCKD, the influence of the parameters on feature extraction is reduced, and meanwhile the problem that the parameters in the VMD and MCKD algorithms are difficult to determine is solved to a certain extent.
Owner:NANJING TECH UNIV