Product
Patsnap Eureka
For R&D, Patsnap Eureka makes reading and utilizing patents & technical documents easy.
Patsnap Eureka AIR
Designed for self-driven R&D workflows. Generate viable solutions, solve complex R&D challenges, empower your innovation with AI.
Patsnap Eureka Materials
Designed for material experts only. Revolutionize your material R&D, from search, analyze, to developing new materials.
Feature
TechResearch
Generate reliable direction feasibility study reports for your R&D in just a few steps.
TechSeek
Discover and master advanced knowledge NOW. Basics, ideas, possibilities, all at once.
TechMind
As an expert in R&D Theories, TechMind can generates customized viable solutions instantly.
TechRisk
Analyze your overall solution with one click, know your potential R&D risks in advance.
TechMonitor
Get weekly tech updates, stay abreast of the latest tech innovations and key insights.
Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
37results about How to "With parallelism" 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
Mobile robot path planning method based on variable-dimension particle swarm membrane algorithm
InactiveCN104050390AReduce computational complexityOvercome the weakness that it is very complicated to build a topology mapSpecial data processing applicationsPosition/course control in two dimensionsAlgorithmState parameter
The invention belongs to the technical field of intelligent robots and control and particularly relates to a mobile robot path planning method based on a variable-dimension particle swarm membrane algorithm. The mobile robot path planning method based on the variable-dimension particle swarm membrane algorithm includes the main steps of performing environment modeling, inputting path parameters, setting a fitness function, initializing parent particles and child particles of an effective path, classifying and re-classifying the path particles, evaluating the fitness function, updating the positions of d-dimension particles, judging whether the current particles have invalid dimensions or not, correcting the particles with invalid dimensions, removing redundant dimension information and the like. According to the mobile robot path planning method based on the variable-dimension particle swarm membrane algorithm, the optimum path suitable for a robot to walk along can be fast planned through global known or local position information of a complex environment and state parameters of an undetermined mobile robot.
Owner:SOUTHWEST JIAOTONG UNIV
An ant colony optimization processing method for large-scale multi-objective intelligent mobile path selection
InactiveCN102278996AQuality improvementImprove convenienceInstruments for road network navigationBiological modelsRobustificationSimulation
The invention discloses an ant colony optimization processing method of large-scale multi-target intelligent moving route selection; after data of NTSP target logistics delivery addresses, distances between every two addresses, and M price of cost for passing through each route are obtained, a route planning unit is solved by ant colony optimization technology so as to obtain a specific walking route for intelligent mobile-agent delivery, and the route is outputted to an executive mechanism for realization. When the method is used to solve the problem of large-scale multi-target intelligent moving route selection, the invention has good optimization performance, and has the advantages of parallelism, self-organization, strong robustness, and the like, and the obtained solutions are large in quantity, high in quality, and have strong approximation capability to the real Pareto solution set; the obtained solution set has uniform distribution; the calculation speed is high. The inventioncan be used in intelligent processing units of route planning systems in fields of logistics distribution, intelligent traffic, internet, robots, etc.
Owner:SOUTHWEST JIAOTONG UNIV
Aircraft route planning method and device on the basis of pigeon colony algorithm
ActiveCN107504972AGood path stabilityPracticalNavigational calculation instrumentsForecastingSimulationPlanning method
The invention provides aircraft route planning method and device on the basis of a pigeon colony algorithm and belongs to the technical field of unmanned aerial vehicle route planning and multi-attribute decision making. The aircraft route planning method includes the steps of: 1) firstly establishing a trace prediction model including uncertainty, and determining a to-be-optimized route in a specified region; 2) with the pigeon colony algorithm, performing iteration to obtain an optimized route through operations with a map and a compass and landmark operation, and finally outputting parameters of the acquired optimized route. Correspondingly, the invention provides an aircraft route planning device on the basis of pigeon colony algorithm, which includes: an acquirement module, a construction module, a determination module, an optimizing module and a memory module. The method and device can deduct and calculate the trace prediction model, so that a route acquired by the model has good stability and has robustness and availability; in addition, through the pigeon colony intelligent optimization method, a problem of complex continuous optimization is solved. The calculation and searching process has the characters of parallelism, availability and high robustness.
Owner:BEIHANG UNIV
Method for distributing thrust of multiple thrusters of ocean engineering equipment dynamic positioning system
InactiveCN103823372AOptimum Solving RealizationWiden the solution areaSteering by propulsive elementsVehicle position/course/altitude controlMathematical modelDynamic positioning
The invention discloses a method for distributing thrust of multiple thrusters of an ocean engineering equipment dynamic positioning system. The method comprises the steps of establishing a multi-thruster thrust distribution mathematical model according to the application requirements of dynamic positioning in an actual project, wherein the method for establishing the multi-thruster thrust distribution mathematical model comprises the steps of obtaining an objective function and obtaining the constraint conditions of the characteristics of the thrusters; conducting optimization solving on the established multi-thruster thrust distribution mathematical model according to a group area searching algorithm. According to the method for distributing the thrust of the multiple thrusters of the ocean engineering equipment dynamic positioning system, due to the adoption of the group area searching algorithm, the problem of limitation of the optimization solving of the nonconvexity problem of the thrust distribution mathematical model of the dynamic positioning system is solved, and the solving range of the optimized thrust distribution is effectively widened; due to the adoption of the regional limitation type searching strategy, the optimized solving of the thrust distribution is achieved; due to the fact that the method has the advantages of being good in robustness and parallelism, simple and effective, the method is extremely beneficial to solving of optimized thrust distribution of the dynamic positioning system and can be applied to actual projects.
Owner:中国船舶重工集团公司第七〇二研究所
Design method of built-in parallel two-dimensional discrete wavelet conversion VLSI structure
InactiveCN1448871AIncrease profitImprove processing speedSpecial data processing applicationsHardware structureShift register
The VLSI structure design method includes shift adding technology to converting the multiplication operation in the filter into shift register and adder operation to reduce the operation amount save hardware spending greatly; buffering technology to connect row and line filters for concurrent performing to save memory space; delay register technology for streamline processing to speed hardware processing and raise hardware utilization; and united symmetrical expanding technology to raise the versatility of the present invention. The present invention is one efficient 2D DWT hardware scheme with several advantages. Through the optimized design of row filter and line filter for 2D discrete wavelet conversion, the hardware utilization is made to reach 100 %, there are two outputs within one work clock, parallel processing is realized without increasing hardware spending, and the obtained hardware structure is simple and easy to realize with VLSI.
Owner:XI AN JIAOTONG UNIV
Method for optimizing allocation of emergency resources in chemical industrial park
ActiveCN108171413AImprove emergency performanceReduce wasteCharacter and pattern recognitionResourcesChemical industryConfiguration optimization
The invention relates to a method for optimizing allocation of emergency resources in a chemical industrial park. On the basis of the traditional hazard source grading model, a hazard source risk dynamic evaluation model is established by combining characteristics of the chemical industry park and considering a domino effect of the accident and risk factors of a process flow and a grading result is obtained based on a pedigree clustering method. On the basis of the hazard source risk grade, a hazard-source-risk-grade-based multi-target optimization emergency resource allocation model is establishing by considering the rescue timeliness and economic applicability of configuration decisions. With the genetic algorithm optimization method, the multi-target emergency resource allocation modelis optimized and then a configuration strategy meeting an expected target is obtained finally. According to the invention, in order to solve some problems in emergency resource configuration optimization, a configuration optimization method with high global optimization capability is put forward. The optimization method has characteristics of having high openness, robustness, parallelism, flexibility, and globally convergence and having no special requirement on the mathematical forms of questions.
Owner:HANGZHOU DIANZI UNIV
Production scheduling optimization method for beer saccharification process
The invention relates to a production scheduling optimization method for a beer saccharification process, wherein beer enterprises are not ideal in scheduling effect because of formulating production scheduling schemes by virtue of artificial experience presently. The production scheduling optimization method disclosed by the invention comprises the following steps of: establishing a production scheduling mathematical model on the basis of a technological process mechanism and first, and then converting to a discrete form capable of solving by virtue of an ant colony optimization algorithm, and finally determining a production scheduling method for the beer saccharification process. The invention discloses a saccharification production scheduling method with a high global optimization capacity aiming at some problems in optimized scheduling for beer saccharification production, and the optimization method has the characteristics of openness, robustness, parallelism, global convergence, and no special requirements on problematic mathematical forms, and the like.
Owner:HANGZHOU DIANZI UNIV
Design method of built-in parallel two-dimensional discrete wavelet conversion VLSI structure
InactiveCN1374692AReduce the numberNo overheadSemiconductor/solid-state device manufacturingSpecial data processing applicationsHardware structureShift register
The design method of built-in parallel two-dimensional discrete wavelet conversion VLSI structure includes: to adopt shift addition technology to change multiplication operation in filter into shift register and adder operation; to adopt sharer expression technology to reduce the number of shift register and adder for realizing the filter; to adopt filter merging technology to merge the common terms of both high-pass and low-pass filters and to realize parallel both high-pass and low-pass operation in one compact hardware structure; and to adopt filter combination merging technology to merge two 2-DDWT filters into one and to raise the hardware utilization. The present invention is one high-efficiency 2-DDWT scheme with very high hardware utilization and other advantages.
Owner:XI AN JIAOTONG UNIV
Measured data-based load model and parameter identification optimization method
InactiveCN108564205AWith parallelismIncrease trustworthinessForecastingArtificial lifeLoad modelParallel computing
The invention relates to a measured data-based load model and parameter identification optimization method, and belongs to the technical field of electric power system load modeling. The method comprises the following steps of: establishing a static load model, a dynamic load model and a comprehensive load model, and determining a to-be-identified parameter of each load model according to three load model expressions; respectively identifying the to-be-identified parameters in the three load models by using a genetic algorithm on the basis of measured data; and carrying out error analysis on parameter identification results of the three load models and finally selecting the load model corresponding to a minimum load model identification error as an optimized load model. According to the method, three common load models are comprehensively considered, and load model optimization is carried out while parameter identification is carried out on the load models, so that benefit is brought to improve the load modeling precision and the effect is favorable.
Owner:KUNMING UNIV OF SCI & TECH
Micro fault diagnosis method and device based on data driving
ActiveCN112348124AImprove troubleshooting speedWith parallelismCharacter and pattern recognitionNeural architecturesTime informationData set
The invention provides a micro fault diagnosis method and device based on data driving, and the method comprises the steps: preprocessing an obtained to-be-detected sample of a to-be-detected aircraft, obtaining a feature data set of the to-be-detected aircraft, inputting the feature data set into a pre-trained fault diagnosis model so as to obtain fault probabilities of multiple various fault types of the to-be-detected aircraft, wherein the fault diagnosis model comprises an input layer, a time convolution network layer and a Softmax classification layer; selecting the maximum fault probability selected from the fault probabilities of the different fault types, and determining the fault type corresponding to the maximum fault probability as the target fault type of the to-be-tested aircraft. Therefore, long-time information capture is realized through introduction of the time convolution network, and the method has the characteristics of parallelism and low memory, and is beneficialto improvement of the fault diagnosis speed of the hypersonic aircraft.
Owner:BEIHANG UNIV
Identity authentication method and system based on electroencephalogram characteristics
ActiveCN107196809AEasy to use and maintainEasy to buildData switching networksPhysiological signal biometric patternsGreek letter alphaDeep belief network
The invention discloses an identity authentication method and system based on electroencephalogram characteristics. The method comprises the following steps: model training, password setting, alpha wave test, electroencephalogram characteristic extraction and classification, and model updating. The electroencephalogram characteristic extraction and classification method is composed of pre-training layer by layer, network fine tuning and classification. According to the identity authentication method and system disclosed by the invention, first level authentication is performed on a tester by means of the property that the alpha waves fluctuate greatly when the tester is lying, meanwhile the existing method is improved to establish a feature extraction method based on a depth belief network, meanwhile the feature extraction method is optimized by using an immune leapfrog algorithm, and when the tester observes password patterns, feature extraction and classification are performed on the brain waves so as to perform second level authentication. By means of the two levels of authentication, the security of the user can be guaranteed, and the improved depth belief network can also improve the existing identification precision, ensure less error rates, avoid a large number of false alarms and bring good user experience.
Owner:西安慧脑智能科技有限公司
Dynamic early warning method for major hazard source of chemical industrial park
ActiveCN108596364AStrong global optimization abilityStrong opennessForecastingCharacter and pattern recognitionKernel principal component analysisDimensionality reduction
The invention relates to a dynamic early warning method for a major hazard source of a chemical industrial park. The dynamic early warning method is characterized by adopting parallel coordinates to perform correlation analysis on correlated variables of key link parameters in the major hazard source, using screened variables with strong correlation as input variables, performing dimensionality reduction on the variables with strong correlation through a principal component analysis method before prediction; dynamically predicting a key production link variable through a recurrent neural network; and performing danger level evaluation by using dynamic prediction results of the key parameters. The dynamic early warning method of the invention is provided for solving some problems in dynamicearly warning of the hazard source of the chemical industrial park, and has the advantages of openness, robustness, parallelism, flexibility, no special requirement on mathematical form of the problem, and so on.
Owner:HANGZHOU DIANZI UNIV
An integrated optimization method for production planning and scheduling under uncertain requirements
The invention relates to an integrated optimization method for production planning and scheduling under uncertain requirements. At present, a production plan and scheduling problem solving mode is mainly a traditional optimization method, an optimization result is difficult to be an optimal solution, and the traditional optimization method cannot be realized technically, so that expected production targets cannot be realized, resource distribution cannot be issued, and the like. The method comprises the following steps: acquiring specific data of a production planning layer and a scheduling layer; respectively establishing a planning layer cost model and a scheduling layer cost model under the condition of uncertain demand according to the specific data; and utilizing an improved collaborative optimization method to optimize the plan and scheduling cost model under the condition of uncertain demand, and finally obtaining a production scheme with the lowest cost. Aiming at some problemsin production planning and scheduling optimization, the invention provides an optimization method with stronger global optimization capability, and the optimization method has the characteristics ofopenness, robustness, parallelism, global convergence, no special requirement on the mathematical form of the problem and the like.
Owner:HANGZHOU DIANZI UNIV
Method for optimizing beer production formula
The invention relates to a method for optimizing a beer production formula. A conventional production formula is determined by a method which combines human experience and tests. The method comprises the following steps of: obtaining main raw materials in a beer formula; establishing a comprehensive production performance index estimation model according to the parameters of the raw materials, wherein main considered comprehensive performance indexes comprise saccharifying power, wort total nitrogen, wort alpha amino nitrogen, wort beta glucan and the thickness of a distilled grain layer of a filter tank; and optimizing a beer formula model by using a variable-metric ant colony optimization method and finally obtaining a production formula with the lowest cost. The method of the invention has the advantages of openness, robustness, parallelism and global convergence.
Owner:杭州坤天自动化系统有限公司
Ant colony optimization processing method of large-scale multi-target intelligent moving route selection
InactiveCN102278996BQuality improvementImprove convenienceInstruments for road network navigationBiological modelsLogistics managementThe Internet
The invention discloses an ant colony optimization processing method of large-scale multi-target intelligent moving route selection; after data of NTSP target logistics delivery addresses, distances between every two addresses, and M price of cost for passing through each route are obtained, a route planning unit is solved by ant colony optimization technology so as to obtain a specific walking route for intelligent mobile-agent delivery, and the route is outputted to an executive mechanism for realization. When the method is used to solve the problem of large-scale multi-target intelligent moving route selection, the invention has good optimization performance, and has the advantages of parallelism, self-organization, strong robustness, and the like, and the obtained solutions are large in quantity, high in quality, and have strong approximation capability to the real Pareto solution set; the obtained solution set has uniform distribution; the calculation speed is high. The inventioncan be used in intelligent processing units of route planning systems in fields of logistics distribution, intelligent traffic, internet, robots, etc.
Owner:SOUTHWEST JIAOTONG UNIV
QIGA(quantum inspired genetic algorithm)-based discretized film system design method of broadband EUV (extreme ultraviolet) multilayer film
ActiveCN107678158AIncrease diversityWith parallelismGenetic algorithmsOptical elementsSystems designDiscretization
The present invention discloses a QIGA(quantum inspired genetic algorithm)-based discretized film system design method of a broadband EUV (extreme ultraviolet) multilayer film. The method comprises the following steps that: 1) the initial parameter values of the design method are inputted; and 2) quantum encoding is performed on the film layer deposition time of the multilayer film, so that an initial quantum group Q (t) can be generated; (3) observation states R (t) of quantum superposition states are constructed according to the quantum probability amplitude of the individuals in the initial quantum group Q (t); (4) whether a termination condition is satisfied is judged, if the termination condition is satisfied, the method stops and outputs an optimal solution, otherwise, the method continues; (5) the fitness of individuals in the R (t), which characterizes the deposition time of a discretized film system, is calculated, individuals corresponding the optimal deposition time of thefilm system are saved; (6) a quantum rotation gate is updated, and the updated quantum rotation gate is adopted to update the Q (t); (7) the Q (t) is observed, R (t) is generated and is evaluated, optimal individuals are retained; (8) the Q (t) is updated through a quantum NOT gate; and (9) an elite retention strategy is adopted, and the method shifts to step (4). With the method of the inventionadopted, problems such as low solution efficiency and high requirement for thickness control precision during film coating of an existing EUV multilayer film design method can be solved.
Owner:CHANGCHUN UNIV OF SCI & TECH
Gamma adjusting and recalibrating method and system of OLED module based on multi-objective relational mathematic model
ActiveCN108597442AOvercoming problems such as poor consistencyIncrease production capacityStatic indicating devicesMathematical modelBiobjective optimization
The invention discloses a Gamma adjusting and recalibrating method and a Gamma adjusting and recalibrating device of an OLED module based on a multi-objective relational mathematic model. The method comprises the following steps: (1) establishing an objective function; and (2) establishing a constraint condition, wherein R, G and B register values of all the binding points of the module are in positive linear or negative linear dependence, the multi-objective relation optimization is converted into single-objective optimization, the double-objective optimization is converted into single-objective optimization, and finally, the R, G and B register values corresponding to F are solved and are output to R, G and B registers corresponding to the gray scale binding point m in a module chip according to setting. According to the technical scheme, the diversified mode of gamma adjustment is supported, the features of platformization and parallelism are realized, and the method and the systemhave the advantages of efficiency and stability in adjusting and good gradual change effect. The experimental result shows that with the provided mathematic model, the complexity can be effectively reduced, the conventional algorithm is converted into algorithm and calculation, the adjusting and recalibrating efficiency is very ideal, and the problems including that the module is poor in consistency are solved.
Owner:WUHAN JINGCE ELECTRONICS GRP CO LTD
Table Saw Fence With Adjustment Mechanism
PendingCN113001627AAdjustable positionPrecise cuttingMetal sawing devicesGuide fencesStructural engineeringKnife blades
A table saw fence with an adjustment mechanism is provided. A fence assembly configured to guide a workpiece with respect to a blade of a saw device includes a fence that is supported by rails for sliding movement relative to a work surface. Opposed ends of the fence are supported on the rails via rail engagement members. The fence assembly includes a fence angle adjustment mechanism associated with a housing of one of the rail engagement members. The fence angle adjustment mechanism includes biasing elements working in conjunction with an adjustment screw to adjust the position of the fence with respect to parallelism to the blade.
Owner:ROBERT BOSCH GMBH
Deductive reasoning method for separating contradiction based on extended triangle in first-order logic
InactiveCN108875945ASolve the cooperative logical relationshipHave diversityInference methodsFirst-order logicSatisfiability
The invention discloses a deductive reasoning method for separating a contradiction based on an extended triangle in first-order logic. The method sequentially constructs an extended triangle standardcontradiction for a clause set in the first-order logic, forms a standard contradiction separating formula, and judges the unsatisfiability of the clause set according to the standard contradiction separating formula. If it can be judged that the set of clauses is not satisfiable, stop operation is performed, and the conclusion can be drawn: the judged set of clauses S is not satisfiable; otherwise, the obtained standard contradiction is added into the clause set S to form a new clause set, and then the extended triangle standard contradiction is repeatedly constructed, and the extended triangle standard contradiction is separated until the judgment conclusion of the attribute of the clause set S is obtained or the set condition is satisfied and stop operation is performed. The inventionprovides an effective dynamic automatic deductive reasoning method based on the contradiction separation in the first-order logic, which can effectively characterize the cooperative logic relationshipamong multiple clauses, and can be applied to the fields of program verification, theorem machine proving and the like.
Owner:SOUTHWEST JIAOTONG UNIV
Discretized film system design method of broadband euv multilayer film based on qiga
ActiveCN107678158BIncrease diversityWith parallelismGenetic algorithmsOptical elementsMultilayer membraneFilm base
Owner:CHANGCHUN UNIV OF SCI & TECH
Method for optimizing beer production formula
InactiveCN101872374BIncrease opennessRobustSpecial data processing applicationsTotal nitrogenPerformance index
Owner:杭州坤天自动化系统有限公司
An Integrated Optimization Method for Production Planning and Scheduling
ActiveCN108805325BFast solutionReduce manufacturing costForecastingResourcesProgram planningGlobal optimization
The invention relates to an integrated optimization method for production planning and scheduling. At present, the solution to production planning and scheduling problems is mainly the traditional optimization method. The optimization result is difficult to be the optimal solution, and it is likely that it cannot be realized in the process, resulting in the failure of the expected production target and the failure of resource allocation. . The steps of the invention are: obtaining the specific data of the production planning layer and the dispatching layer; respectively establishing the planning layer expense model and the dispatching layer expense model according to the specific data; using the improved collaborative optimization method to optimize the planning and dispatching expense models, and finally obtaining the lowest cost production plan. The present invention aims at some difficult problems in production planning and scheduling optimization, and proposes an optimization method with strong global optimization ability. The optimization method has openness, robustness, parallelism, global convergence and a mathematical form for problems No special requirements and other characteristics.
Owner:HANGZHOU DIANZI UNIV
Intelligent Substation Equipment Status Early Warning Method Based on ICA Algorithm
ActiveCN105932774BReliable transmissionReduce power consumptionCircuit arrangementsInformation technology support systemInformation processingData set
The present invention provides an intelligent substation equipment state early warning method based on ICA algorithm, which includes the following steps: (1) information data collection; (2) information data transmission; (3) information data processing: the control center in the control room reads in real time Collect the data set sent by the collector, and use the ICA algorithm to clean up the collected data set, data integration and conversion, data dimensionality reduction, data discretization, etc.; (4) Early warning diagnosis: the control center of the station control room uses Bayeux The Sri Lankan algorithm analyzes the data processed in step (3), and through the methods of data classification, association and prediction, trains and models historical data and integrates the existing experience of staff to obtain strategies, and outputs corresponding early warning signals. The invention can simply and quickly extract effective data information, and provide more accurate early warning for the operation status of the primary / secondary equipment of the substation, so that the monitoring and early warning of the substation is more timely and accurate. The invention is suitable for risk assessment and early warning of all communication equipment in the power system.
Owner:ZHANGJIAKOU POWER SUPPLY COMPANY OF STATE GRID JINBEI ELECTRIC POWER COMPANY +1
Mapping search method and device for cache memory and main memory
ActiveCN105474215BRealize logical operation relationshipWith parallelismData acquisition and loggingNon-linear opticsControl signalParallel computing
A method and device for mapping lookup between a cache and a main memory; an optical comparator realizes a logical operation relationship xor by utilizing the above relation of a micro ring size in an optical field instead of negating a control signal A and an input signal B in an electric field; because of the parallelism of light, the optical comparator is capable of parallelism calculation and has small delay, and can be used to search the cache for the position of the data required by the CPU; and compared with a logic circuit implementing fully-associative mapping in the prior art, the logic circuit of the fully-associative mapping implemented by the optical comparator has a simpler structure, and the comparison delay does not increase with the increase of the number of compared bits.
Owner:罗三杰
A data-driven micro-fault diagnosis method and device
ActiveCN112348124BImprove troubleshooting speedWith parallelismCharacter and pattern recognitionNeural architecturesFlight vehicleTerm memory
The present application provides a data-driven micro-fault diagnosis method and device, wherein the method includes: preprocessing the acquired test samples of the test aircraft to obtain the feature data set of the test aircraft, and the feature data The collection is input into the pre-trained fault diagnosis model to obtain the fault probability of various fault types of the aircraft to be tested. The fault diagnosis model includes an input layer, a time convolutional network layer and a Softmax classification layer. The maximum failure probability is selected in the failure probability, and the failure type corresponding to the maximum failure probability is determined as the target failure type of the aircraft under test. In this way, this application realizes the capture of long-term information through the introduction of time convolution network, which has the characteristics of parallelism and low memory, and helps to improve the speed of fault diagnosis of hypersonic aircraft.
Owner:BEIHANG UNIV
Identity authentication method and authentication system based on EEG features
ActiveCN107196809BEasy to use and maintainEasy to buildData switching networksPhysiological signal biometric patternsDeep belief networkFeature extraction
The invention discloses an identity authentication method and an authentication system based on electroencephalogram features, wherein the method includes five steps of model training, password setting, alpha wave testing, electroencephalogram feature extraction and classification, and model updating. Among them, the brain wave feature extraction and classification method is composed of three parts: layer-by-layer pre-training, network fine-tuning, and classification. The present invention first utilizes the characteristic that α waves fluctuate greatly when a person tells a lie to carry out the first re-authentication on the tester, improves the existing method at the same time, establishes a feature extraction method based on a deep belief network, and uses the immune leapfrog algorithm to It is optimized to perform feature extraction and classification on the tester's brain waves when observing the password pattern, so as to perform the second verification. The two-layer verification can not only ensure the safety of users, but the improved deep belief network can also improve the existing recognition accuracy, reduce the error rate, avoid a large number of false alarms, and bring a good experience to users.
Owner:西安慧脑智能科技有限公司
A Path Planning Method for Mobile Robots Based on Variable Dimensional Particle Swarm Algorithm
InactiveCN104050390BReduce computational complexityOvercome the weakness that it is very complicated to build a topology mapSpecial data processing applicationsPosition/course control in two dimensionsAlgorithmState parameter
The invention belongs to the field of intelligent robot and control technology, and in particular relates to a path planning method for a mobile robot based on a variable-dimensional particle swarm algorithm. The steps of the present invention mainly include: modeling the environment, inputting path parameters, setting the fitness function, initializing effective path parent particles and sub-particles, classifying and reclassifying path particles, evaluating the fitness function, updating the position of the d-dimensional particle, and judging Whether the current particles have invalid dimensions, correct the particles with invalid dimensions and remove redundant dimension information, etc. The invention can quickly plan the optimal path suitable for walking of this type of robot by using the globally known information of the complex environment or the information of the local position of the complex environment and the state parameters of the undetermined mobile robot.
Owner:SOUTHWEST JIAOTONG UNIV
Method for classifying electroencephalogram signals based on variable parameter recurrent neural network
PendingCN114580513AImprove signal-to-noise ratioImprove robustnessCharacter and pattern recognitionNeural architecturesAlgorithmMotor imagery
The invention provides a method for classifying electroencephalogram signals based on a variable parameter recurrent neural network, which comprises the following steps of: collecting the electroencephalogram signals of a user through an electrode cap, and then preprocessing the data; constructing a convex quadratic programming problem of a motor imagery electroencephalogram signal classifier by utilizing a soft interval support vector machine principle; introducing a convex quadratic programming problem into a penalty function to eliminate inequality constraints; solving a convex quadratic programming problem with equality constraints through a variable parameter recurrent neural network; substituting the optimal solution of the convex quadratic programming problem into the classifier decision function to obtain the classifier decision function; and judging the category of the newly input motor imagery electroencephalogram signal through the obtained classifier decision function, and outputting the result. According to the method, the convex quadratic programming problem of the motor imagery electroencephalogram signal classifier is solved by using the variable parameter recurrent neural network, and the method has the advantages of high convergence speed, good robustness and high classification accuracy.
Owner:SOUTH CHINA UNIV OF TECH
A progressive refinement transmission method and system for single-band matrix DEM data
ActiveCN110381076BMeet specific requirements in terms of accuracyImplementation of specific requirements in terms of precisionDigital video signal modificationTransmissionComputational scienceSingular value decomposition
The invention discloses a single-band matrix type DEM data progressive refined transmission method and system. The method comprises: segmenting DEM data, performing singular value decomposition, analyzing sub-slices, extracting effective singular vectors and singular values layer by layer, then performing compression, performing transmission and storage from a sending end to a receiving end layerby layer, and performing recovery at the receiving end. The system comprises a segmentation module, a decomposition module, an analysis and extraction module, a compression module, a transmission module and a receiving module. According to the method and the system, refined transmission under the control of precision parameters can be realized according to the requirements of actual precision, andthe specific requirements of numerical DEM data in the aspect of precision are met. After the DEM data fragments are compressed, the DEM data fragments do not need to be directly recovered to be in amatrix form at the receiving end and can continuously exist in a vector form, and the compression effect is reserved. When the data is recovered, the used calculation is basic calculation, so that the calculation burden is not increased, and the method has parallelism.
Owner:KUNMING UNIV OF SCI & TECH
OLED module gamma adjustment method and system based on multi-objective correlation mathematical model
ActiveCN108597442BOvercoming problems such as poor consistencyIncrease production capacityStatic indicating devicesRound complexityLinear correlation
The invention discloses a Gamma adjusting and recalibrating method and a Gamma adjusting and recalibrating device of an OLED module based on a multi-objective relational mathematic model. The method comprises the following steps: (1) establishing an objective function; and (2) establishing a constraint condition, wherein R, G and B register values of all the binding points of the module are in positive linear or negative linear dependence, the multi-objective relation optimization is converted into single-objective optimization, the double-objective optimization is converted into single-objective optimization, and finally, the R, G and B register values corresponding to F are solved and are output to R, G and B registers corresponding to the gray scale binding point m in a module chip according to setting. According to the technical scheme, the diversified mode of gamma adjustment is supported, the features of platformization and parallelism are realized, and the method and the systemhave the advantages of efficiency and stability in adjusting and good gradual change effect. The experimental result shows that with the provided mathematic model, the complexity can be effectively reduced, the conventional algorithm is converted into algorithm and calculation, the adjusting and recalibrating efficiency is very ideal, and the problems including that the module is poor in consistency are solved.
Owner:WUHAN JINGCE ELECTRONICS GRP CO LTD
Who we serve
- R&D Engineer
- R&D Manager
- IP Professional
Why Patsnap Eureka
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com