26 results about "Design objective" patented technology

The invention discloses a digital circuit design method and device, equipment and a medium, and relates to the technical field of digital circuits, and the method comprises the steps: determining a target type of a to-be-designed digital circuit, and collecting a first training set corresponding to the target type; each training sample in the first training set comprises an input feature item used for designing a target type of historical digital circuit and the actual result quality of the historical digital circuit, and the input feature item comprises a high-level language source code and an HLS constraint parameter; constructing an initial prediction model based on the multi-layer perceptron mixed structure, training the initial prediction model by using the first training set, and searching and optimizing the model architecture based on the network structure to obtain a target prediction model; and predicting the actual result quality corresponding to each parameter combination in a target design space of a to-be-designed digital circuit by using the model, and designing the digital circuit by using the parameter combination of which the quality meets the Pareto frontier. And predicting an accurate actual quality result by utilizing a prediction model without intermediate representation so as to carry out digital circuit design.

Embodiments of the present disclosure provide enhanced systems and methods for predicting optimal design flow parameters for optimized output targets for physical design synthesis of a given IC design. A Variational Autoencoder (VAE) along with a regression network are trained using a dataset comprising synthesis design construction flows from historical IC designs to provide a training data representation of the dataset constrained to a latent space of the VAE. The system generates feature vectors based on the training data representation of the dataset and updates the feature vectors with initial design characteristics of the given IC design. The system iteratively performs an input gradient search of the updated feature vectors to optimize an objective function of the design targets to identify locally optimal design parameters. The system identifies globally optimal design flow parameters for optimized design targets based on locally optimal design parameters.

The invention discloses a fabricated wall BIM model multi-objective optimization method based on acoustic parameter analysis, and relates to the technical field of building information models. The method comprises the following steps: constructing an acoustic parameter analysis-prediction coupling model based on an auto-encoder and a deep network, and generating a lightweight performance prediction proxy model through a knowledge distillation technology; the performance prediction agent model, a cost calculation module and a weight calculation module are jointly packaged into an acoustic parameter analysis-model optimization plug-in conforming to BIM platform interface specifications; and realizing automatic multi-target optimization of the target fabricated wall in the BIM platform through the plug-in. According to the method, real-time prediction of high-precision acoustic performance and collaborative optimization of a multi-dimensional design target are achieved, interactive target and constraint setting of a user in a BIM environment is supported, the technical problems that in a traditional method, performance analysis efficiency is low, the optimization process is split, and multiple targets are difficult to balance are solved, and the method is suitable for popularization and application. And the scientificity, efficiency and engineering practicability of the assembly type wall body design are improved.

The invention relates to a microstrip network layout optimization method based on Fourier series, and belongs to the technical field of radio frequency circuit design. The method aims at solving the problems of topology dependence, boundary discontinuity and low efficiency caused by sharp increase of optimization parameters along with the area in a traditional rectangular spliced microstrip network. The core of the technical scheme is that a microstrip network boundary is modeled into a plane closed curve, the curve is represented through Fourier series decomposition, and a Fourier coefficient is optimized by taking circuit performance as a target drive, so that a microstrip layout with any shape and a continuous and smooth boundary is generated. According to the method, automatic design rule checking and port adding processes are integrated, and finally electromagnetic simulation iteration is carried out until a design target is met. The method can break through the limitation of the traditional topology, effectively improves the electromagnetic performance and design efficiency, and is widely applied to the microstrip network design of broadband matching, multi-band matching, harmonic control and the like.

The invention discloses an automatic optimization method based on a CAD model, and relates to the technical field of engineering design automation and computer aided design optimization. The method does not depend on manual predefined parameter dimensions any more, directly takes topology and geometric features of the CATIA model as core input, constructs a complete geometric optimization link through automatic extraction, structure mapping and optimization algorithm cooperative processing, ensures synchronous evolution of geometric features and design targets in the optimization iteration process, and improves the optimization efficiency of the CATIA model. And the optimization efficiency and stability are obviously improved. A reinforcement learning or evolutionary algorithm is built in the system to perform efficient search and multi-round adaptive iteration on geometric feature vectors, and global optimization of a complex structure can be completed without manual intervention. According to the method, automatic model generation of an optimization result is realized through a CATIA API and a macro command interface, and manual geometry reconstruction by an engineer is not needed. CATIA can be written back through one key according to optimized feature information, the closed-loop automatic process of input, optimization and output is achieved, the design period is remarkably shortened, and the iteration response efficiency is improved.

This invention discloses a design method and apparatus for refractory high-entropy alloys based on a large model, relating to the field of alloy material design technology. The method includes: First, by generating and collecting data, pre-trained surrogate models for thermodynamics (covering phase diagrams, solidification processes, etc.) and mechanical properties are constructed, and calculation models for physical parameters such as valence electron concentration and density are established. Then, these models are integrated into a single ensemble model, and a large language model is introduced as the core of intelligent scheduling. The large model is responsible for parsing the design task, intelligently planning and calling the interfaces of each model based on a knowledge base and rule base. Finally, based on the set design objectives, a multi-objective genetic algorithm is initiated for iterative optimization and adaptive adjustment, efficiently outputting a final alloy design scheme that meets multiple performance requirements. This invention enables intelligent multi-objective design of refractory high-entropy alloys, solving the problem of high dependence on manual intervention.

This invention discloses a design method for a silicon-based mode divider. Based on a silicon-on-insulator (SiI) platform, it defines a bus waveguide, branch waveguides, and a functional region located between the bus and branch waveguides. The number of branch waveguides is n, determined by a linear combination of the input modes and bandwidths of the bus waveguides. The initial structures of the bus waveguides, branch waveguides, and functional regions are defined. Using an FDTD solver, each iteration requires m forward simulations and m adjoint simulations. The gradient field is calculated, and the dielectric constant parameters of the functional region are optimized using the gradient descent method based on the gradient field calculation results, updating the structure. Iteration continues until the design target is reached or further optimization is impossible, at which point it stops. The discrete dielectric constants are binarized, transforming the dielectric constants of the functional regions into a structure defined only by the functional regions. The final structure diagram is output. This invention also discloses a silicon-based mode divider designed using the above method.

This invention provides a base station design device and method for determining the placement of base stations in an environment where terminal usage fluctuates over time, by performing robust base station design with minimal computational effort. [Solution] The method using the station placement design device of the present disclosure involves arranging multiple base station installation candidates within a design target area, calculating the radio quality between each terminal and each base station installation candidate at each time step based on the movement prediction of multiple terminals within the design target area, creating a weighted directed graph for each time step with each terminal as a node, the connection between each terminal and each base station installation candidate as an edge, and each radio quality as the weight of each edge, obtaining a set of graphs, compressing the set of graphs in the time direction to obtain a compressed set of graphs, and selecting the placement location of a base station from multiple base station installation candidates based on the compressed set of graphs such that the sum of the probabilities of the existence of terminals whose radio quality is equal to or greater than the quality target value is equal to or greater than the design target value.

This invention discloses a multi-domain adaptive design rule method based on EMC data assets. It includes: acquiring the target application scenario identifier, functional type, IC combination, and design objectives; retrieving matching typical basic products and their derivative product cases to obtain their feature sets and mapping relationships; constructing an initial design concept by combining IC-level and system-level data assets; retrieving the standard essential characteristics of the target application scenario; constructing an EMC domain adaptation model, establishing an adaptive mapping based on verified mapping relationships, and quantitatively correcting and generating design rule parameters; outputting design rules containing constraint objects, types, target values, source metadata, and protection layer priorities; and structurally storing and assigning multi-dimensional tags. This invention establishes cross-application scenario mappings through typical basic products and derivative products, achieving quantitative adaptive generation of design rules, supporting cross-domain reuse, test prediction, rule self-evolution, and automatic generation of design documents, shifting EMC design from experience-based to data-driven.

The invention belongs to the technical field of circuit design, and discloses a method for impedance matching design based on a large language model, which comprises the following steps of: 1, collecting impedance matching design demand information for preprocessing, and outputting demand data; 2, semantic understanding, field extraction and structured mapping are carried out, a rule verification mechanism is adopted to carry out conversion unit unification, parameter range verification and missing information complementation, and standard structured design task description is generated; 3, judging a matching task type, selecting a corresponding design according to the matching task type, and generating a topological structure and parameters; 4, converting into a netlist file, performing S parameter simulation, analyzing key performance and indexes, judging initial structure performance, and generating a report and storing the report into an RAG knowledge base when a design target is met; 5, if not, parameter optimization is carried out, a report is generated and stored in an RAG knowledge base after optimization succeeds, modification suggestions are generated if the optimization succeeds for many times, and then redesign is carried out. According to the method, the matching structure design efficiency and accuracy of the large language model are improved.

This invention discloses a passive component selection method based on EMC data assets. It includes: acquiring the application scenario, design objectives, housing structure, and EMC standard requirements of the target product; acquiring port and PCB interconnect information to determine the protection priorities of emission suppression, immunity enhancement, and transient protection; retrieving IC-level and system-level EMC data assets to identify interference source pins and sensitive pins; searching historical design cases to extract successful solution characteristics and failure modes; calculating the required emission suppression, immunity enhancement, transient protection level, housing shielding effectiveness, and common-mode suppression requirements; establishing an inside-out forward design method to determine IC-level, PCB-level, and system-level three-layer protection and dynamically allocate weights; establishing a weighted scoring model, introducing a multi-dimensional labeling system for filter topology, generating and ranking candidate solutions; and integrating selection records into a closed-loop protection selection data asset. This invention achieves quantitative selection of passive components.

This invention belongs to the field of flowmeter technology and discloses a method for optimizing the impeller of a turbine flowmeter. The method includes: S1: collecting data on the physical properties, operating range, and performance indicators of the medium through a system to establish clear design constraint data; S2: integrating streamlined and multi-lead design concepts, and determining the blade structural parameters through mathematical modeling and simulation analysis; S3: ensuring precise matching between the processing process and design objectives through parameter modeling, accuracy threshold setting, process coordination, and data linkage; S4: manufacturing an impeller prototype based on the processing technology, and calibrating the blade size and angle parameters using equipment; S5: verifying linearity and starting flow rate indicators through data acquisition and analysis, and iteratively optimizing design and processing parameters; S6: establishing a large-scale batch production process based on the optimized final parameters. This invention solves the problems of large starting flow rate, poor linearity, and insufficient consistency of metering accuracy in turbine flowmeters under low Reynolds number conditions.

The application discloses a training data obtaining method and device applied to electromagnetic device design, a medium and equipment, and relates to the technical field of electromagnetic device design. The application utilizes the inverse mapping relationship between the electromagnetic response curve and the topological distribution of the frequency selective surface structure, firstly performs topological modeling and full-wave simulation, and utilizes a binary matrix to express the topological structure. In the iteration of the target function, the topological binary matrix and the electromagnetic response reserved in each iteration are taken as a training data set, more useful design information can be reserved, the electromagnetic response has more similar characteristics with the design target, misleading information is less, and the quality of the training data in the electromagnetic device design is improved.

A computer-implemented physics solver operates within a neural network framework. A neural network accepts a coordinate for each location within a design domain and outputs a local composition for each location. A model of the design domain is formed in a physics solver. The model includes discrete elements that encompass the design domain. A solution of the model provides a value of a design objective. For a plurality of iterations, the following is performed: the neural network determines current local compositions for the locations in the design domain corresponding to the discrete elements; the current local compositions are input into the discrete elements of the physics solver to obtain a current value of the design objective; and the current value of the design objective is used to find a loss gradient of a loss function. The loss gradient is used to update the neural network during the iterations.

The invention discloses a power via hole design method and device, equipment and a storage medium. The method comprises the following steps: dividing the PCB into a plurality of functional areas according to electrical functions, and distributing a current weight factor, a thermal coupling weight factor and a signal integrity sensitivity weight factor for each functional area; the method comprises the following steps: dispersing a power supply via hole of a PCB into a plurality of infinitesimal segments with the same layer number as the PCB in the axial direction, and defining a geometric parameter set containing aperture, side wall copper thickness and anti-pad size for each infinitesimal segment; performing iterative optimization on the geometric parameter set of the infinitesimal segment based on the weight factor of the functional region to obtain a Pareto optimal solution set; and based on a preset design target of the PCB, selecting a target geometric parameter set from the Pareto optimal solution set, and designing the power supply via hole of the PCB according to the target geometric parameter set. The method solves the problem of cross-domain failure chain of the traditional equal-diameter power supply via hole design, and has the effects of realizing multi-physics field system-level collaborative optimization and remarkably improving the comprehensive performance of a power supply distribution network.

This application discloses an intelligent design method, apparatus, device, and readable storage medium for adjacent tunnel spatial relationships, relating to the field of tunnel engineering technology. It includes: acquiring basic data on existing tunnels and the engineering environment; determining the feasible spatial domain of the new tunnel based on the basic data and the design objectives of the new tunnel, and establishing a spatial relationship constraint model for adjacent tunnels; generating a set of candidate spatial layout schemes for the new tunnel within the feasible spatial domain, based on the spatial relationship constraint model; constructing an intelligent evaluation model to score and rank the feasible schemes in the candidate spatial layout scheme set; and outputting a recommended design scheme that meets the scoring requirements and its key spatial relationship parameters with existing tunnels. This method significantly improves design efficiency, reduces reliance on manual experience, and ensures the quantifiability and reproducibility of the design results.

The invention discloses an automatic design method for an electrical system of a traction substation kiosk based on characteristic value matching. The method comprises the following steps: carrying out structured hierarchical module division on the electrical system of the traction substation kiosk; carrying out feature value definition and weight distribution on each hierarchical module, and establishing a module feature value database; according to project design requirements, generating a design target feature value set of each level of target module; on the basis of the design target feature value set of each level of target module, obtaining a candidate module list of each level of target module in a module feature value database through similarity calculation, and selecting and obtaining a final matching result on the basis of the candidate module list; and finally, combining the matching results of the selected layers of modules, and automatically generating an electrical system design result file. Modules and characteristic value systems of corresponding levels are established according to industry characteristics, and automatic matching, recursive combination and local replacement of the modules are realized by using the method disclosed by the invention, so that the design efficiency and the reuse rate are remarkably improved.

The application discloses a circuit automatic design method based on large language model distillation and self-evolution, comprising the following steps: inputting a circuit type and a design target specification; constructing a reference netlist set, and performing initial supervision fine-tuning on a generation strategy model; generating a candidate netlist and performing simulation, calculating a target accuracy and a standard mark of each candidate netlist; for the candidate netlist that does not meet the standard, calculating a constraint violation degree and attempting automatic repair; for the candidate netlist that meets the standard, calculating a comprehensive performance vector and performing a Pareto sorting and diversity screening; performing preference optimization training by using good and bad sample pairs; performing security strategy optimization; and outputting a finally generated netlist. The application can reduce invalid candidate netlists and invalid simulation, improve the output rate of candidate netlists meeting the design target specification, and enhance the training stability and the circuit automatic design effect under multi-index constraints.

The invention discloses an antenna artificial intelligence design system and method, electronic equipment and a storage medium, and belongs to the technical field of antenna design. The system executes stored instructions through a processor, receives a design target, generates a geometric structure image, performs boundary correction based on an edge effect physical model to obtain an electrical size image, and generates a candidate antenna structure under the condition of the design target by using a conditional generative adversarial network. A multi-modal neural network is adopted to carry out fusion processing on multi-source features such as geometric structures, contours, electrical dimension images and key parameters, S parameters are predicted, and iterative optimization is carried out until design requirements are met. According to the method, a physical model is integrated to improve the prediction precision and generalization ability, joint automatic optimization of the antenna structure and size is realized, the design period is shortened to be within 3 hours from traditional 72 hours, and the problems that traditional design efficiency is low, and an existing AI method is insufficient in degree of freedom and ignores a physical mechanism are effectively solved.

This invention belongs to the field of microwave vacuum electronic device design technology, specifically relating to a strip-beam electron gun design method based on intelligent collaborative optimization. The method includes: S1, selecting an initial model from a database according to the design objectives; S2, establishing the model and extracting electron beam performance parameters through a CST and MATLAB co-simulation platform; S3, calculating the objective evaluation function; S4, optimizing the electron gun structural parameters using an intelligent collaborative optimization algorithm; S5, performing performance evaluation and redundancy analysis; and S6, outputting the final parameters and storing them in a database. This invention achieves full automation of the strip-beam electron gun design process. By intelligently adjusting the optimization strategy through algorithms, it significantly improves design efficiency and accuracy, effectively optimizes the laminar flow and uniformity of the electron beam, and supports both single-sided and double-sided compression strip-beam electron gun configurations.

The application provides a MIMO antenna design optimization method, device, equipment and medium, and belongs to the antenna design field. The method comprises the following steps: obtaining historical design parameters and corresponding performance indicators of a MIMO antenna; preprocessing the historical design parameters and the performance indicators, constructing a mapping relationship between input data and output data, and training a forward prediction model and a generative model based on the mapping relationship; receiving a target performance indicator input, generating a candidate design parameter set by using the trained generative model, taking the candidate design parameter set as an initial population, combining the forward prediction model as a surrogate model, adopting a multi-objective evolutionary algorithm for iterative optimization, and outputting a Pareto optimal solution set as a target design parameter set. In this way, the application greatly speeds up the performance evaluation link with the help of the surrogate model, provides diversified non-inferior solution selection through Pareto frontier analysis, and can flexibly trade off different design targets according to actual requirements.

The application belongs to the field of microwave technology and electromagnetic optimization design, and particularly discloses an electromagnetic device optimization design method based on a particle swarm optimization algorithm assisted by multi-view semi-supervised learning. The method disclosed by the application comprises the following steps: defining design variables and optimization objectives, and generating initial samples; updating particle positions by using an improved social learning particle swarm algorithm, selecting a prediction model to perform real simulation on potential particles, and storing the remaining particles in an unlabeled sample pool; screening high-confidence samples by using a multi-view model, generating pseudo-labels by using convex combination optimization, and expanding a training set; constructing a semi-supervised learning model, calculating a prediction error, and setting a tolerance interval; selecting a prediction model according to a three-mode decision rule, iteratively updating the model and the particle swarm, and stopping until the design objectives are met. The application can efficiently optimize the performance of electromagnetic devices with a small number of simulation times. The application is suitable for electromagnetic design problems with high-cost simulation evaluation.

The invention belongs to the technical field of product process design of intelligent manufacturing, and discloses a target essential process design method based on double functions of demand and constraint, which comprises the following steps: acquiring user demand information and process constraint information; converting the user demand information into corresponding elements; normalizing the process constraint information to obtain a constraint group; constructing an attachment between the constraint and the element to update the standard element group; and according to the updated standard element group, determining an association relationship among the elements, constructing an element chain, and determining a design target. According to the method, fuzzy demands are automatically mapped into executable process parameters through a structured demand-element conversion model, so that subjectivity is reduced; and through the dynamic association of the elements and the constraints, the process route can be dynamically adjusted, and the method has certain feasibility and applicability.

This application discloses a full-link design method for an intelligent operation monitoring system in a nuclear power plant, comprising: acquiring monitoring characteristic parameters and design input parameters; the monitoring characteristic parameters including one or more of the monitoring frequency, monitored objects, anomaly identification methods, and intervention methods under manual monitoring conditions; and the design input parameters including the design objectives and constraints of the intelligent operation monitoring system; determining the application constraint parameters, overall configuration parameters, and functional configuration parameters of the intelligent operation monitoring system based on the monitoring characteristic parameters and design input parameters; obtaining the intelligent operation monitoring system based on the overall configuration parameters; calibrating and verifying the intelligent operation monitoring system; and iteratively optimizing the intelligent operation monitoring system based on the verification results. This application achieves standardized design and stable deployment of an intelligent operation monitoring system for nuclear power plants by constructing a full-link design method covering design input analysis, system configuration determination, core element development, and calibration iteration.

The invention provides a parameter design method and device for a double-core symmetrical controllable phase shifter, and relates to the technical field of power grids. According to the method, firstly, key performance indexes such as a phase shifting angle, rated capacity, rated voltage, rated current and a short-circuit impedance range of the controllable phase shifter are determined based on system design requirements and constraint conditions, then accurate calculation is carried out according to the key performance indexes, the rated transformation ratio of a series transformer and a parallel transformer of the controllable phase shifter is determined, and the controllable phase shifter is obtained. The core component parameters such as the rated voltage, the rated current and the rated capacity of each winding are obtained, and the layered parameter design thought of first integration and then component is achieved; the overall performance index of the phase shifter is determined from the global demand of the power system, it is ensured that the design target is highly consistent with the system demand, the problem of disjunction with the overall operation condition possibly caused by direct component design is avoided, the overall performance optimization can be achieved, and the problem that a traditional parameter design method of the controllable phase shifter is unreasonable is solved.

The application discloses a material library constraint-based radio frequency matching network design method and device, equipment and storage medium, and the method comprises the steps of obtaining a material index space, determining a target impedance and a target matching configuration according to a current matching demand; constructing a current search space based on the target matching configuration and the material index space, and running a differential evolution algorithm in the current search space to obtain an optimal nominal value combination; mapping the optimal nominal value combination to an integer index in the material index space, enumerating all candidate material combinations in the neighborhood of the integer index, determining impedance errors of the candidate material combinations, selecting the candidate material combination with the minimum error as a current optimal discrete solution, converting the current optimal discrete solution back to a continuous value, and returning to iteration until an optimal material combination for designing the target matching network is output. The application embeds the material library discrete constraint in the optimization process, adopts a double-layer search architecture of an outer continuous differential evolution and an inner discrete neighborhood enumeration, eliminates the disconnection between design and material selection, and thus improves the efficiency and accuracy.