Patents
Literature
310results about How to "Improve design efficiency" 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
A method, device and medium for fast retrieval of a three-dimensional model
The application discloses a three-dimensional model fast retrieval method, equipment and medium, and relates to the computer technical field. The method comprises the following steps: in response to receiving a part of a two-dimensional sketch of a three-dimensional model drawn by a user, performing image segmentation on the part of the two-dimensional sketch to obtain a plurality of sub-images; performing mapping processing on each sub-image to obtain an initial feature corresponding to each sub-image, and performing attention calculation on the initial features corresponding to the plurality of sub-images to obtain a plurality of first key features; fusing the plurality of first key features to obtain a second key feature corresponding to the part of the two-dimensional sketch; based on the second key feature, performing similarity matching on a feature database to obtain a plurality of target features, and searching a three-dimensional model set corresponding to the plurality of target features from a three-dimensional model database. In this way, the target three-dimensional model meeting the design intention of the user can be quickly and accurately retrieved according to the part of the sketch of the model drawn by the user, so that the design efficiency and the user experience are significantly improved.
Owner:SHANDONG HUAYUN 3D TECH CO LTD
A method and system for rapidly predicting a creep stress exponent of a tin-based solder alloy using an integrated model
ActiveCN115132293BCreep stress exponent is simpleCreep Stress Index QuickChemical property predictionEnsemble learningData setCreep stress
The application discloses a method and system for rapidly predicting a creep stress index of a tin-based solder alloy by integrating a model, collects element compositions, test temperatures and creep stress index values of the tin-based solder alloy from literature, and adds experimental data as dataset samples; the element compositions and test temperatures of the tin-based solder alloy are sorted out and used as features for modeling; the dataset is randomly divided into a training set and a test set in a 4:1 ratio; the collected creep stress index values of the tin-based solder alloy are used as target variables, and the features are used as independent variables; based on the divided training set, the independent variables are scaled by using RobustScaler, three learners are trained and integrated to obtain an R-X-L integrated model; and the R-X-L integrated model is used to rapidly predict the creep stress index of the test set samples and four independent experimental samples. The prediction model for the creep stress index of the tin-based solder alloy is simple, fast, low-cost, pollution-free and the like based on reliable literature data and a modeling method.
Owner:SHANGHAI UNIV +1
Key-value memory network-based active recommendation method for design knowledge of complex mechatronic systems
ActiveCN115859822BImprove and refine performanceWeaken barriers to information exchangeDesign optimisation/simulationNeural architecturesSystem design processNetwork output
The application discloses a kind of based on key value memory network's complex electromechanical system design knowledge active recommendation method.First, the scene feature semantic information extraction of software platform log file in the design process of complex electromechanical system is carried out, and scene ontology is established based on scene feature semantic information, and system scene knowledge base is formed by scene ontology and original knowledge base;Then, the scene maximum frequent sequence is used to describe the scene sequence feature similarity between designers;Again, the knowledge item interaction sequence of all designers is learned, and the knowledge item sequence preference vector corresponding to all designers is obtained;Further, input to key value memory network, and the initial design knowledge active recommendation sequence is obtained;Finally, after knowledge item selection, the final design knowledge active recommendation sequence of each designer is obtained.The application obtains design knowledge active recommendation sequence more in line with the needs of designers, and then improves the efficiency of complex electromechanical system design.
Owner:HANGZHOU INNOVATION RES INST OF BEIJING UNIV OF AERONAUTICS & ASTRONAUTICS
A wall corner element restraint method for a rocking wall structural system
ActiveCN116108538BImprove design efficiencyfast constraintGeometric CADSustainable transportationDrift angleClassical mechanics
This invention discloses a method for constraining corner elements in a rocking wall structure system. Based on the structural parameters of the rocking wall and the frame structure, the shear stiffness of the frame structure is obtained. A first lateral displacement expression and a second lateral displacement expression are constructed using the shear stiffness and the first and second equilibrium equations. An inter-story drift angle expression is then constructed using these expressions to solve for the stiffness of the rotating combined disc springs. The bending moment of the rotating combined disc springs is obtained from the stiffness of the rotating combined disc springs. The number of single disc springs stacked in the same direction and the number of combined disc springs are determined by the deformation requirement and the bearing capacity requirement of the combined disc springs obtained from the bending moment. This yields the combined height of the disc springs. The combined height of the disc springs is then used to constrain the length of the core rod and the support height in the corner element, achieving rapid constraint of the core rod length and support height in the corner element based on statics.
Owner:ZHEJIANG SECOND CONSTR GRP CO LTD +1
Optimization design method of metamaterial absorber based on spatial mapping network and absorber
This invention discloses a metamaterial microwave absorber optimization design method and absorber based on a spatial mapping network. First, the trust region corresponding to the initial optimization point is determined. Then, sampling points for design parameters are obtained around the initial optimization point, and the electromagnetic response corresponding to the design parameters at the sampling points is obtained through CST full-wave simulation. Next, based on the structure of the metamaterial microwave absorber to be optimized, its equivalent circuit diagram and reflection coefficient expression are obtained. The reflection coefficient formula is used as a coarse model, and the result of CST full-wave simulation is used as a fine model. A mapping between the coarse and fine models is established through an artificial neural network model. A spatial mapping network is constructed to optimize and obtain the design parameters corresponding to the target electromagnetic response of the metamaterial microwave absorber and the predicted corresponding electromagnetic response curve. Electromagnetic simulation is used to verify whether the optimized design parameters meet the requirements. If they do, the optimization stops; otherwise, the trust region is updated, and the next round of optimization begins. This invention has high design efficiency and a short average time requirement.
Owner:HUAZHONG NORMAL UNIV
A design method for an inducer wheel of a liquid rocket engine turbopump and the inducer wheel itself.
ActiveCN120926129BImprove cavitation performanceAvoid the problem of poor batch delivery performance consistencyGeometric CADSustainable transportationTurbopumpMechanical engineering
This invention discloses a design method and inducer for a liquid rocket engine turbopump. Based on the overall design requirements of the liquid rocket engine, the operating conditions of the inducer are determined. According to the design requirements and empirical formulas, the basic parameters of the inducer are initially determined. Based on the range of core parameters, different combinations of core parameters are used to establish multiple initial three-dimensional models of the inducer. Simulations are performed on each of the established initial three-dimensional models to obtain the cavitation performance of the inducer. Based on the flow analysis results, the inducer with the optimal blade profile among all parameter combinations is obtained. The selected inducer with the optimal blade profile is then optimized and adjusted until the liquid rocket engine turbopump inducer with the best cavitation performance is obtained. This invention can design a liquid rocket engine turbopump inducer with good cavitation performance.
Owner:ORIENTAL SPACE (XIAN) AEROSPACE TECHNOLOGY CO LTD +3
Container ship side bulkhead and method of designing the same
The application relates to a container ship side water-tight strong frame and a design method thereof, and relates to the technical field of container ship hull structure design, which comprises the following steps: a piece of strong frame plate is enclosed by using a perimeter plate as the boundary of the side water-tight strong frame; the perimeter plate comprises an outer plate, an inner shell, an upper platform and a lower platform, and all are provided with longitudinal frames; a through hole is added at the position where the longitudinal frame of the perimeter plate penetrates the strong frame plate, and a water-tight patch plate is arranged; the longitudinal frames on the inner shell and the outer plate of the ship body are respectively sorted according to the vertical coordinates from small to large, then are sequentially and one-to-one matched, and a stiffener is arranged between each pair of longitudinal frames on the strong frame plate; the strong frame plate is divided into multiple panel grids by using the stiffeners, and the steel grade and the plate thickness of each panel grid are determined; the steel grade and the section size of each stiffener are determined; the panel grids and the stiffeners are grouped, and the plate thickness is combined and the size is combined respectively. The application can guarantee the rationality and reliability of the water-tight strong frame design, and is also helpful to improve the design efficiency.
Owner:RES INST 708 OF CHINA STATE SHIPBUILDING CORP
New process for designing high-performance copper-iron alloy with high efficiency
PendingCN122503672Asolve blindnessInhibition of macrosegregation
This invention belongs to the field of copper alloy material design and preparation technology, and discloses a new process for efficiently designing high-performance copper-iron alloys. Based on the 5% Fe content vertical cross-sectional characteristics of the Cu-Fe-Si ternary phase diagram, the Si content is designed in stages (0-4%) to achieve targeted performance optimization. Combined with a three-stage heat treatment process of "homogenization-basic aging-gradient aging", a closed-loop system of "phase diagram design-preparation-detection-optimization" is constructed. Through precise composition control by EPMA and real-time feedback of mechanical properties, the problems of blind composition design and insufficient precipitation of strengthening phases in traditional processes are solved. The resulting alloy has excellent tensile strength, yield strength, and total elongation at break, with synergistic improvement in mechanical and magnetic properties, and a design efficiency improvement of more than 40%, making it suitable for electronic, mechanical and other fields.
Owner:GUILIN UNIV OF ELECTRONIC TECH
Aircraft automatic test system based on LabVIEW
PendingCN121858425AEasy to transplantEasy to updateError detection/correctionTotal factory controlTestwareOperational system
The invention provides an aircraft automatic test system based on LabVIEW. The aircraft automatic test system comprises a platform layer, a module layer, a scheduling layer and an interface layer, the platform layer is used for providing an operating system function interface, a hardware driving interface, a mathematical function library and a multi-thread library for the upper layer, and the module layer is used for defining a communication protocol with each test single-machine device of the aircraft, realizing module software for program control of each device, and realizing module software such as file processing and data analysis interfaces. The scheduling layer is used for starting module software in a multi-thread mode, loading a test file and responding to a user event of an upper layer to schedule the module software, and the interface layer is used for interacting with a user, generating the user event from an operation instruction of the user, transmitting the user event to the scheduling layer and displaying the execution progress of a test task. The modularized design mode facilitates transplantation and updating of the test software, the development complexity of the aircraft test software can be reduced, and the software design efficiency is improved.
Owner:SHANGHAI INST OF ELECTROMECHANICAL ENG
Diversion tunnel intelligent design method based on multi-module collaborative optimization
PendingCN121997428AEasy to find the economic optimal solutionImprove design efficiencyGeometric CADDesign optimisation/simulationControl engineeringDesign methods
The invention relates to an intelligent diversion tunnel design method based on multi-module collaborative optimization, which solves the problem of a linear serial mode of a traditional diversion tunnel design process, breaks through the limitation of artificial experience and finds a global or approximately global structural economic optimal solution. Establishing a model basis library, inputting engineering boundary conditions, and standardizing data; the intelligent hydraulic calculation and section optimization module is used for performing hydraulic simulation calculation to generate a plurality of section design schemes; the preliminary support design module is used for matching an initial support design scheme with known sections and geological conditions; the structural calculation module is used for calculating, checking and optimizing mechanical safety and economical efficiency by integrating the design schemes of the section and the preliminary support; the economic comparison module quantifies the economic cost of each design scheme and provides a target function value; and through an intelligent closed-loop mode, closed-loop processing is carried out on the calculation module by adopting modes of design, verification, feedback, optimization and result output in sequence.
Owner:GUANGDONG ELECTRIC POWER PLANNING SURVEY & DESIGN INST
Fabricated steel structure and mounting method
PendingCN121976611AGuaranteed rigidityGuarantee efficiencyBuilding material handlingSteel columnsIndustrial engineering
The invention relates to the technical field of the building industry, and discloses a fabricated steel structure and an installation method, the fabricated steel structure comprises frame columns and frame beams, and the frame columns and the frame beams are connected in a matched mode; the modular node units are detachably connected with the frame columns and the frame beams respectively; the modularized joint unit comprises a first connecting plate and a second connecting plate, the sides, deviating from each other, of the first connecting plate and the second connecting plate are suitable for being connected with one ends of the adjacent frame columns correspondingly, and a box type steel column section is arranged between the first connecting plate and the second connecting plate; the two ends of the box type steel column section are connected with the first connecting plate and the second connecting plate correspondingly. The modularized joint units are further provided with I-shaped steel beam ends, and the two I-shaped steel beam ends are connected with the two ends of the box type steel column section respectively in the extending direction of the frame beam.
Owner:CHINA CONSTR STEEL STRUCTURE ENG CO LTD
A fast design method of low sidelobe reflector antenna for complex requirements
PendingCN122241925AEnsure vertices are coplanarPrecise adaptation of asymmetric structuresGeometric CADDesign optimisation/simulation
This invention discloses a rapid design method for low-sidelobe reflector antennas applied to complex requirements, comprising the following steps: S1: Calculate the initial focal coordinates of the long-axis reflector based on the structural parameters of the long-axis reflector, and calculate the half-angle of the main reflector and the distance from the feed to the vertex of the sub-reflector; S2: Calculate the initial focal coordinates of the short-axis reflector based on the principle of equal optical path length; S3: Introduce a coefficient set to construct a normalized focal distribution feature; S4: Keep the focal points fixed and dynamically reconstruct the focal distribution of the long and short axes within the full illumination angle range; S5: Calculate the coordinates of the main reflector and the sub-reflector; S6: Iteratively adjust the boundary focal points and the coefficient set to converge and obtain a reflector antenna that meets the low-sidelobe index. This invention has strong adaptability and extremely low algorithm freedom, with only 7 degrees of freedom; it can quickly achieve high-performance optimization of complex reflectors, perfectly adapt to irregularly shaped reflector antennas with large aspect ratios, and solve the design problem of asymmetric structures.
Owner:JING LIN CHENGDU SCI & TECH
Packaging method, device and equipment of circuit module and storage medium
The present disclosure provides a packaging method and device of a circuit module, equipment and a storage medium, relating to the technical field of integrated circuit design, comprising: obtaining a first identifier corresponding to a circuit module to be packaged, a target parameter contained in the circuit module, a sub-circuit module and a corresponding second identifier, and a signal flow direction corresponding to the circuit module; based on the signal flow direction, processing the target parameter and the second identifier of the circuit module to obtain a topological connection mode corresponding to the circuit module; determining an external port corresponding to the circuit module; and packaging the circuit module based on the first identifier, the target parameter, the topological connection mode and the external port. Thus, based on the target parameter, the topological connection mode and the external port corresponding to the circuit module, the circuit module is packaged, which facilitates the instantiation of the circuit module, improves the reusability of the circuit module, and thus provides a condition for improving the design efficiency of analog circuits.
Owner:TSINGHUA UNIVERSITY
A modified white carbon black silicone rubber performance prediction and formula optimization method and system
The application provides a modified white carbon black silicone rubber performance prediction and formula optimization method and system, the method establishes a candidate formula set containing white carbon black addition amount omega, modifier type and dispersion process parameters, obtains characteristic parameters such as dispersion uniformity coefficient D, surface silane load molar number n, mechanical property correction coefficient and electrical property correction coefficient of each candidate formula; respectively through a mechanical property prediction model and an electrical property prediction model, the prediction values of tensile strength, elongation at break, Shore hardness, volume resistivity and breakdown field strength are obtained; then the prediction values are compared with the target performance requirements, unqualified formulas are removed and the target formula is determined, and the corresponding addition amount, modifier and dispersion process parameters are output; the application can effectively reduce the repeated sample preparation and test screening workload, realize performance quantitative prediction and formula collaborative optimization, improve the formula design efficiency and the performance regulation accuracy and reliability.
Owner:STATE GRID HUNAN ELECTRIC COMPANY DISASTER PREVENTION & REDUCTION CENT +2
Signal testing methods, systems, equipment, storage media, and software products for chips.
ActiveCN114371387BImplement batch acquisitionimplement testElectronic circuit testingDesign testingSoftware engineering
This disclosure provides a signal testing method, system, device, storage medium, and program product for chips, relating to the field of semiconductor technology. It addresses the technical problem of low chip testing efficiency. The chip has multiple input signals. The signal testing method includes: acquiring all input signals of the device chip; obtaining the time parameters of each device input signal; and generating a result table of the time parameters of all device input signals. By acquiring all input signals of the chip and obtaining the time parameters of each input signal, and obtaining a result table of the time parameters of all input signals, batch testing of the chip's input signals is achieved, improving the design efficiency of circuit designers, facilitating the design of test circuits, and increasing testing efficiency.
Owner:CHANGXIN MEMORY TECH INC
A method, system, device and medium for calculating the outer diameter of an automobile wire harness
ActiveCN116226583BImprove design efficiencyimprove accuracyManufacturing computing systemsComplex mathematical operationsAdhesive beltElectrical design
The application discloses a kind of automobile wire harness outer diameter calculation method, system, equipment and medium, it is related to automobile electrical design field, the method includes: obtaining component pin information;Component pin information includes the wire diameter of each wire in wire harness, wire type and the number of each wire diameter wire;According to the wire diameter and wire type of each wire, the maximum outer diameter of each wire is determined respectively;According to the maximum outer diameter of each wire and the number of each wire diameter wire, the initial wire harness outer diameter is calculated;Determine the outermost wire harness protection material and the outermost wire harness protection mode;Wire harness protection material includes: adhesive tape and tubular object;Wire harness protection mode includes: full winding, point winding and flower winding;According to the outermost wire harness protection material, the outermost wire harness protection mode and the initial wire harness outer diameter, the final wire harness outer diameter is calculated.The application can quickly and accurately calculate the outer diameter size of low-voltage wire harness, thereby improving the 3D design efficiency of low-voltage wire harness, and improving the accuracy of low-voltage wire harness 3D design.
Owner:江苏开沃汽车有限公司
Snow surface overlay layering design method for ice structure bearing
This invention discloses a layered design method for snow surface overlay on ice-structured load-bearing structures, relating to the field of ice and snow engineering technology. The method includes the following steps: S1, determining the design constraints and performance indicators of the snow surface overlay based on target working conditions and environmental data; S2, obtaining the microstructural parameters of the snow material to be laid, including initial porosity, ice crystal size distribution, coordination number, and contact geometry. This invention establishes a cross-scale mechanical performance evolution model to accurately capture the microscopic mechanisms of ice crystal sintering, particle creep, and fatigue damage of snow material under long-term load and temperature, solving the problem of performance prediction deviations caused by the inability of existing macroscopic models to reflect microscopic evolution. By constructing a quantitative mapping relationship between microstructural parameters and macroscopic mechanical properties, combined with thermo-coupling load history simulation, accurate performance evolution data of the overlay can be obtained.
Owner:POLAR RES INST OF CHINA +2
Parameter configuration method and device, electronic equipment, storage medium and program product
ActiveCN121009856BImprove reliabilityImprove design efficiencyComputer aided designSpecial data processing applicationsDependabilityComputer engineering
This application provides a parameter configuration method, apparatus, electronic device, storage medium, and program product. The method includes: obtaining a functional description of a target chip design; obtaining a behavioral-level parameter model, a system-level parameter model, and an RTL-level parameter model based on the functional description; the behavioral-level parameter model includes behavioral-level parameters, the system-level parameter model includes system-level parameters, and the RTL-level parameter model includes RTL-level parameters; establishing dependencies between the parameters in the behavioral-level parameter model, the system-level parameter model, and the RTL-level parameter model to obtain a parameter set; optimizing the parameter set based on the dependencies to obtain a target parameter set that ensures the target chip design is free from timing violations; and outputting parameter configuration based on the target parameter set. This application can reduce the risk of parameter version inconsistencies, improve the reliability of parameter configuration in the chip, and improve chip design efficiency.
Owner:上海芯睿域智能科技有限公司
System and method for detecting the fit of a manufactured piece
PendingCN122197203AAvoid misdiagnosis as design issuesImprove detection accuracyMachine part testingGeometric CAD
The application provides a manufacturing piece fitting detection system and a fitting detection method. The fitting detection system comprises a determination module configured to determine a plurality of to-be-detected parts in a manufacturing piece based on detection requirements; and an artificial intelligence model configured to determine a connection type between the plurality of to-be-detected parts based on the determined plurality of to-be-detected parts. The artificial intelligence model is further configured to detect a fitting relationship between the plurality of to-be-detected parts and determine whether the fitting relationship meets a design specification of the manufacturing piece according to the connection type. The fitting detection system provided by the application first determines the connection type between the to-be-detected parts by using the artificial intelligence model, and then determines whether the fitting relationship meets the requirements based on the connection type and the design specification, so that the reasonable fitting relationship in design can be avoided from being misjudged as a design problem, the accuracy of detection and judgment is improved, and the artificial intelligence model is used to automatically check whether each fitting relationship of the to-be-detected parts is reasonable, so that the detection accuracy and efficiency of the fitting relationship of the manufacturing piece are improved.
Owner:斯特兰蒂斯汽车集团
Rear side wall inner plate, door ring structure and vehicle
ActiveCN224146019UImprove design efficiencyIncrease productivitySuperstructure subunitsStructural engineeringMechanical engineering
Owner:ZHEJIANG ZEEKR INTELLIGENT TECH CO LTD +1
A four-factor constraint-based brake disc optimization method, device and medium
ActiveCN116151142BSolve the problem of suboptimal optimizationImprove cooling efficiencyThermodynamicsMechanical engineering
This invention relates to a brake disc optimization method, device, and medium based on four-factor constraints. The method includes: obtaining the mass of a target brake disc; establishing a model with an undercarriage structure based on the parameters of the target brake disc, and obtaining the maximum limiting temperature of each part; performing optimization design without exceeding the original mass to obtain the optimized brake disc; performing CFD steady-state simulation to calculate the steady-state cooling pump air ratio; when the maximum temperature is lower than the maximum limiting temperature, determining whether it meets the preset optimization target based on the steady-state cooling pump air ratio; performing CFD transient simulation analysis on candidate brake discs to calculate the transient cooling pump air ratio; when the maximum temperature is lower than the maximum limiting temperature, determining whether it meets the optimization target based on the transient cooling pump air ratio; if so, the target brake disc is the preferred brake disc. Compared with the prior art, this invention solves or partially solves the problem that existing brake disc optimization methods lack constraints during the optimization process, leading to unsatisfactory optimization.
Owner:TONGJI UNIV +1
Integrated mask circuit generation and detection method and platform
PendingCN122242401ASignificant practicalitySignificant technical advantagesCAD circuit designSpecial data processing applications
This invention belongs to the field of integrated mask circuit generation technology, and provides a method and platform for generating and detecting integrated mask circuits. The method involves obtaining an S-box configuration file, parsing it, extracting the S-box, generating hardware circuits from the extracted S-box, and obtaining an S-box file. A test environment is constructed based on the S-box file, and corresponding test stimulus files are generated. Behavioral-level simulation is performed in the test environment, and simulation logs are recorded. The simulation logs are parsed to determine whether the S-box design is functionally correct. The RTL-level design of the S-box file is converted into a gate-level netlist based on the target technology library, generating a netlist file and an area report. Timing verification and power estimation are performed under preset timing and load constraints, generating timing and power reports respectively. Multiple reports are synthesized, key performance indicators are extracted, structured, and a comprehensive evaluation report is generated. This invention can improve the design efficiency, reliability, and overall performance of mask circuits.
Owner:SHANDONG UNIV
Intelligent top pipe splitting method based on BIM (Building Information Modeling) technology
PendingCN121936145ARespond quickly to changesreduce statistical errorGeometric CADDesign optimisation/simulationSoftware engineeringSelf adaptive
The invention discloses an intelligent top pipe splitting method based on a BIM technology. The method comprises the following steps that 1, a first Revit self-adaptive pipe joint family is manufactured; 2, generating an overall model of the pipeline according to a projection path of the pipeline by using the first Revit self-adaptive pipe joint family; 3, by means of Dynamo software, the splitting fixed length and the splitting distance of the pipeline are preset, segmentation is conducted, and multiple segments of segmented curves are obtained; 4, processing all the segmentation curves through Boolean operation by utilizing Dynamo software to obtain corresponding segmentation curves segmented into accurate lengths; 5, by utilizing Dynamo software, generating all pipe joints after the pipeline is segmented through the segmentation curve; 6, according to a preset corner threshold value of the adjacent pipe joints, the corner between the adjacent pipe joints is calculated; and 7, exporting a detailed list of all the pipe joints. According to the method, design and machining can be quickly joined, the material statistical error is reduced, and the design and machining efficiency is improved.
Owner:HEFEI MUNICIPAL DESIGN INST
Terahertz metamaterial sensor reverse design method based on transfer learning and reinforcement learning
PendingCN121859736AImprove design efficiencyShorten the deployment cycleDesign optimisation/simulationInference methodsLearning machineTerahertz metamaterials
The invention discloses a terahertz metamaterial sensor reverse design method based on transfer learning and reinforcement learning, and belongs to the technical field of terahertz metamaterial sensor design. In order to solve the technical problems that an existing terahertz metamaterial sensor is low in reverse design efficiency and insufficient in generalization ability, and structural parameters meeting a preset performance target are difficult to rapidly obtain, a hybrid deep learning model is firstly constructed and pre-trained, and then a generalization prediction model is obtained through an adversarial transfer learning mechanism. Then, a reinforcement learning optimization environment is established, finally, a depth deterministic strategy gradient algorithm is utilized, a generalization prediction model is used as an environment simulator for training, optimal metamaterial structure parameters are output, and reverse design of the terahertz metamaterial sensor can be efficiently achieved. The method is suitable for quickly acquiring sensor structure parameters meeting the requirements of a target frequency band, an absorptivity peak value and a quality factor.
Owner:HENAN UNIVERSITY OF TECHNOLOGY
Method for modeling a high-speed communication link transmitter based on a transformer decoder
ActiveCN118826961BImprove design efficiencyEfficiently describe nonlinear characteristicsTransmitter specific arrangementsDesign optimisation/simulationEncoder decoderCommunication link
This invention discloses a high-speed communication link transmitter modeling method based on a Transformer decoder. This method parameterizes the input signal and establishes a deep learning model with an encoder-decoder architecture, including a non-sequential encoder and a Transformer sequential decoder, to predict the output signal of the transmitter given the input signal. The encoder processes the input signal parameters, generates a context vector, and then passes it to the decoder. The decoder uses the context vector and the transmitter output signal sequence to generate a class probability distribution for each point in the sequence. The model is trained using a random masking strategy and employs non-autoregressive decoding and filtering techniques for inference, achieving parallel output sequence prediction. The final output is obtained through a single signal filtering step. Compared with traditional simulation techniques, this method significantly improves simulation speed while maintaining a very small error.
Owner:ZHEJIANG UNIV
A project design and optimization method, system, and storage medium
ActiveCN121168254BConvenience for follow-up workImprove design efficiencyDesign optimisation/simulationConstraint-based CADResource consumptionKnowledge graph
This invention provides a project design and optimization method, system, and storage medium. The method includes: acquiring initial project information; cleaning, normalizing, and performing correlation processing on the initial project information to obtain processed project information, wherein the processed project information includes processed historical project information and processed current project information; constructing a knowledge graph by processing the processed historical project information using a natural language processing algorithm and a fusion model; constructing a project performance prediction model using machine learning methods based on the processed historical project information and the knowledge graph; acquiring project requirements; determining constraints based on the project requirements; obtaining candidate projects based on the constraints, the project performance prediction model, and the processed current project information; and determining the final project based on the candidate projects. This application can improve design efficiency and reduce resource consumption.
Owner:HANGZHOU ZHONGCHENG CONSULTING SUPERVISION CO LTD
Method for optimizing hafnium-based ferroelectric transistor performance based on bayesian theory
This invention discloses a method for optimizing the performance of hafnium-based ferroelectric transistors (FFETs) based on Bayesian theory. The invention first constructs a FeFET device with an MFMIS structure and calculates the effective area, extracting the corresponding memory window (MW) and subthreshold swing (SS) to build an initial sample dataset. Secondly, based on the initial sample dataset, a Gaussian process surrogate model is established to predict the MW at a preset value. R Within range R The changing continuous response curve. Then, based on the Gaussian process surrogate model, the predicted... R‑ MW continuous response curve, finding the optimal R Finally, a comprehensive evaluation index is constructed for the identified optimal index. R An evaluation will be conducted, and the one with the highest comprehensive evaluation index will be selected. R As a preferred option R This invention optimizes the performance of hafnium-based ferroelectric transistors. It reduces the number of experiments or simulations, improves prediction stability under limited sample conditions, and increases device design efficiency.
Owner:HANGZHOU DIANZI UNIV
Force Feedback Nonlinear Active Disturbance Rejection Controller and Method Based on Particle Swarm Optimization
This invention discloses a force feedback nonlinear active disturbance rejection controller and method based on particle swarm optimization, comprising: a signal input module, a tracking differentiator, a feedback control law, an extended state observer, a force feedback system, and a particle swarm algorithm module; the control method includes: step one, building a model; step two, determining the spatial expression; step three, transient processing of displacement signals; step four, processing feedback control signals; step five, processing extended state observation signals; and step six, compensating for the control quantity feedback system. This invention uses the extended state observer for real-time estimation to compensate for the total disturbance received by the force feedback system during operation. Furthermore, the designed nonlinear active disturbance rejection control technology can perform force feedback control without relying on a specific system model, exhibiting strong adaptability and robustness. The particle swarm optimization algorithm can automatically tune parameters, improving the design efficiency of the nonlinear active disturbance rejection controller and optimizing control performance.
Owner:HARBIN INST OF TECH
Test case generation method, device and product based on weighted graph double-path fusion
The application discloses a test case generation method and device based on weighted graph double-path fusion and a product, relates to the technical field of software testing, and automatically extracts an entity list and a relationship triple from a requirement document and constructs a directed weighted graph, thereby replacing a traditional manual design and preset rule mode and improving the design efficiency of test cases; forward paths and reverse paths are generated by respectively performing forward search and reverse search, and a defect type value is added in the reverse search process, so that defects of the prior art which focus on normal flow verification and are insufficient in abnormal scene coverage are overcome, bidirectional coverage of normal flow and abnormal scene is realized, and test coverage and defect discovery capability are significantly improved; test cases are generated by matching and fusing the forward paths and the reverse paths and performing priority sorting on the fused paths, automatic and standardized generation of a test case set is realized, and the efficiency, comprehensiveness and quality stability of the test case set are improved.
Owner:HANGZHOU YOUYUN TECH CO LTD
Method and system for layout optimization of high-cleanliness plant pipeline spaces
PendingCN122088009AConstraint placement priorityconstraint space
This invention discloses a method and system for optimizing the layout of pipeline spaces in high-cleanliness industrial plants, relating to the field of computer intelligent construction technology. The method includes: acquiring a three-dimensional model of the building structure, equipment layout, and pipeline attribute information within the plant; dividing the pipeline layout area into grids based on the column grid and process equipment layout; calculating a location convenience index, structural impact index, and equipment association index for each grid, fusing them to obtain a comprehensive spatial importance index, and assigning functional area attribute codes accordingly; constructing a multi-objective optimization function with objectives including comprehensive pipeline cost, inter-pipeline conflict degree, and secondary connection pipeline space; constructing the pipeline system as a graph structure, using a graph neural network to solve the multi-objective optimization function, and generating a preliminary layout scheme; and outputting the final layout after collision detection and local adjustments. This invention achieves automated and intelligent layout optimization of pipeline spaces in high-cleanliness industrial plants, improving space utilization and reducing construction costs.
Owner:CHINA CONSTR FOURTH ENG DIV CORP LTD +1