16 results about "Global iteration" patented technology

The application relates to the technical field of tunnel engineering disaster prediction, and discloses a soft rock tunnel large deformation prediction method based on a particle swarm optimization neural network, which comprises the following steps: selecting geological construction characteristic parameters such as water content of surrounding rock and construction methods as input factors, and establishing a qualitative index and quantitative value mapping; based on orthogonal test design and numerical simulation, engineering sample data sets are constructed and normalized; subsequently, the initial connection weight and bias of an error back propagation neural network are globally iteratively optimized by using a particle swarm optimization algorithm, so that optimal initial values are obtained by taking network errors as fitness values; finally, data training is carried out to generate a mature model, and a predicted deformation value is output and combined with a double check mechanism to carry out engineering early warning. The application solves the problems that traditional neural networks are slow in convergence and prone to falling into local optimization due to random parameter initialization, and effectively improves the prediction accuracy of soft rock tunnel large deformation and the reliability of engineering early warning.

This invention provides a method for reconstructing continuous wheel-rail forces based on discrete fastener reaction forces, comprising the following steps: Step 1, establishing a wheel load distribution model; Step 2, establishing an algorithmic framework for the inversion method, by introducing a responsibility region and a local deconvolution strategy, mapping the discrete fastener reaction force FRF to the continuous wheel-rail force WRF. This invention effectively solves the problems of traditional onboard sensors' difficulty in long-term fixed-point monitoring of specific road sections, and the inability of existing ground discrete monitoring points to fully characterize the continuous wheel-rail interaction behavior. By constructing an inversion framework based on Gaussian load distribution using a "responsibility region + local deconvolution" approach, it fully utilizes the load transfer characteristics of the vehicle-track coupling system, achieving high-precision reconstruction of continuous wheel-rail forces using ground discrete fastener reaction force data. This eliminates the need for complex global iterative optimization and can accurately restore the dynamic characteristics of wheel-rail contact even with only a limited number of discrete measurement points.

This invention discloses a data-driven adaptive parametric modeling method for automotive rooflines, comprising the following steps: acquiring multiple sets of side-view roofline coordinate data of real vehicles, dividing the side-view roofline into a front guide segment and a rear main body segment; establishing a normalized coordinate system for each segment, and then discretizing it into several sampling points to form a bottom-level sample database; defining control variables, including the spatial position of control points in the normalized coordinate system, the number of control points, and the type of reconstructed model; employing a two-layer optimization strategy to perform global iterative optimization of the control variables for each segment to obtain the optimal control variables; constructing a comprehensive evaluation index, and selecting the control variable with the optimal comprehensive evaluation index from all parametric reconstructed models that have undergone inner-layer optimization as the optimal control variable, thus obtaining the optimal parametric reconstructed model. This invention aims to solve the problems of existing designs relying on manual experience and the difficulty in achieving an optimal balance between fitting accuracy and model complexity.

This application relates to the field of data processing technology and discloses a method for processing heterogeneous multi-scale data in Gaussian process regression. The method first acquires the raw data and performs feature transformation, standardization, and dimensionality reduction preprocessing to construct a multi-level validation dataset. Second, it reconstructs the Gaussian process regression kernel function, introducing a learnable Minkowski distance parameter to replace the traditional fixed Euclidean distance metric. Next, it uses a tree-structured Parzen estimator strategy to perform global iterative optimization on the validation set with the goal of minimizing prediction error, determining the optimal Minkowski distance parameter. Subsequently, it trains the model based on the optimal parameter and an automatic correlation determination mechanism, jointly optimizing hyperparameters such as signal variance and length scale through maximum likelihood estimation. Finally, it performs prediction and outputs the mean and variance. This invention, through adaptive parameterization of the kernel function metric, effectively overcomes the shortcomings of traditional models in adapting to non-spherical distributions and multi-dimensional heterogeneous features, significantly improving prediction accuracy and generalization performance.

Embodiments of the present application provide a method and device for constitutive model construction and parameter optimization of honeycomb materials, computer equipment and medium, relating to the technical field of electronic digital data processing, wherein the method comprises the following steps: obtaining original experimental data sets generated by mechanical experiments of the honeycomb material under multiple different working conditions; fitting to obtain a shape function representing the stress-strain relationship under a reference working condition, constructing a density function representing the influence of density on mechanical properties, using an exponential nonlinear function, constructing a strain rate function representing the influence of strain rate on mechanical properties, and constructing a temperature function representing the influence of temperature on mechanical properties; constructing a constitutive model of the honeycomb material based on multi-factor correction factors; taking the minimum mean square error between the stress-strain curve predicted by the constitutive model and the target curve as the optimization target, performing global iterative optimization and outputting. Through the constructed constitutive model, the problem of strong dependence of traditional model parameter fitting and poor universality is solved.

This invention relates to the fields of seismic resistance in geotechnical engineering and construction technology in hydraulic engineering, and particularly to a vibration energy dissipation feedback control system for core wall anti-liquefaction. The system includes: a data acquisition and positioning module that acquires vibration response signals and location information in real time during the compaction process; a signal processing and feature extraction module that extracts characteristic parameters representing the dynamic state of the fill material; a real-time liquefaction risk diagnosis module that outputs a quantified liquefaction risk index using a pre-trained discriminant model; an intelligent decision-making and control command generation module that generates control commands based on this index; an actuator adaptive adjustment and targeted compaction module that drives the roller to adjust its working parameters and performs targeted enhanced compaction in high-risk areas; and a global iteration and visualization monitoring module that coordinates the entire process and generates a quality risk distribution map. This invention achieves real-time perception and adaptive closed-loop control of anti-liquefaction performance during construction, fundamentally improving the seismic stability of the core wall.

This invention discloses a power device packaging optimization method based on a BPNN-WOA fusion model, relating to the field of power device thermal management. It addresses the problems of traditional optimization methods, such as neglecting nonlinear coupling of parameters, difficulty in global optimization, and handling complex discrete variables. This method relies on a database of heat dissipation influencing factors and a high-fidelity 3D thermal simulation model. Parameters are selected and standardized to generate a sample dataset, which is then used to train a BPNN thermal resistance prediction surrogate model. The WOA algorithm's whale position vector is mapped to a combination of packaging parameters, and the BPNN output is mapped to a fitness value. The optimal parameter combination is obtained through global iterative optimization using WOA, and verified by simulation after inverse normalization. If the deviation exceeds a threshold, additional samples are added to retrain the model. This method can capture the synergistic effect between parameters, achieving global optimization in a continuous-discrete hybrid design space, significantly reducing packaging thermal resistance, and improving optimization efficiency and accuracy.

The application discloses an FMD parameter self-adaptive adjustment method based on VPPSO optimization, and belongs to the technical field of bearing fault parameter self-adaptive adjustment methods.S1: obtaining a bearing vibration signal in a strong noise environment as original input information for FMD parameter optimization;S2: constructing a VPPSO double-population collaborative optimization model to determine the optimization range of the mode number n, filter length L and cutting frequency band number K of the FMD;S3: taking the reconstructed signal loss function value as the fitness index of VPPSO, and performing global iterative optimization on the key parameters of the FMD through the double-population collaborative optimization mechanism to output the optimal FMD parameter combination; through the VPPSO double-population collaborative optimization, the FMD parameter is automatically adjusted, manual trial and error are not needed, different fault types such as inner rings, outer rings and rolling bodies are adapted, and different fault diameter scenes of 0.007-0.021in are adapted, multiple fault types and diameters are adapted, the limitation of traditional one parameter and one working condition is solved; compared with the non-optimized PSO, the VPPSO reduces the convergence iteration number and greatly shortens the optimization time, and greatly improves the engineering application efficiency.

The application discloses a kind of transposed step-by-step optimization methods of cascade hydropower station interplant load distribution, comprising: obtaining basic hydrological engineering data and power grid load demand, generating the initial cumulative output sequence of each hydropower station;With cumulative output as state variable, in the order of first time period and then library, optimization is carried out using transposed step-by-step optimization algorithm;In the optimization process, the dynamic feasible region of the cumulative output of the current hydropower station is constructed, and the physical operation constraints of each reservoir are strictly checked;The optimal cumulative output point is selected by calculating the consumed energy storage objective function value;Based on the optimization result after global iteration convergence, the water head and load filling scheduling instructions are output.The application directly converts load balance into boundary constraint of iterative feasible region, reduces the calculation overhead of exploring infeasible solution, and improves the optimization efficiency of algorithm and cascade water utilization rate.

The application discloses a kind of underwater binocular camera calibration methods and systems of fusion polarization information, belong to marine environment monitoring technical field.The application constructs the isomerous binocular system under active polarization illumination, left eye camera is equipped with rotatable linear polarizer, right eye camera is full transparent;By collecting parallel and orthogonal polarization images, clear polarization feature image is solved using orthogonal polarization difference denoising model, left eye polarization feature point and right eye gray feature point are extracted respectively;Further construct joint error optimization objective function, water body real-time refractive index and polarization plate installation error are taken as state variable and included in objective function, and nonlinear optimization algorithm global iteration optimization is used.The application can effectively suppress water body backscattering noise, eliminate dynamic refractive index and mechanical error influence, realize the high-precision in-situ calibration of binocular camera under turbid water, and hardware change is small, engineering adaptability is strong.

The application discloses a kind of based on personalized federal learning identification method and system of differential privacy, the method is executed in client, comprising: obtaining initialization model parameter, and it is loaded into pre-built local model;Based on privacy budget, using personalized differential privacy algorithm is carried out noise processing to the local model of loading parameter, and the shared layer model parameter of the local model after noise processing is sent to server;The shared layer model parameter after aggregation is loaded into the shared layer of the local model after noise processing, then local fine-tuning is obtained personalized model;The personalized model is repeatedly trained until reaching global iteration number, and the personalized federal learning test model based on differential privacy is obtained.The client of the application can process local model according to the privacy budget selected by customer, to adapt to different privacy needs of different users, while combining Adam algorithm, reduce the influence of added noise on model accuracy.

This invention relates to the field of optical sensing and measurement technology, specifically to a design method and system for a matrix-type optical absorption cell based on a genetic algorithm. The method is as follows: First, a parameterized model of the matrix-type optical absorption cell is constructed, and the two side mirrors are divided into partitions. A genetic algorithm gene encoding sequence is constructed using the local curvature center of each partition as the optimization variable. A recursive relationship is established based on the law of light reflection, and the light propagation path is calculated successively, while optical performance indicators are statistically analyzed. A multi-objective comprehensive evaluation fitness function integrating indicators such as reflection count and spot coverage is constructed. An improved genetic algorithm incorporating partition-aware mutation and adaptive population update mechanisms is used for global iterative optimization. Finally, the optimal parameter combination and the corresponding light path and spot distribution characteristics are output. This invention can achieve automated optimization design of the absorption cell structure, improve design efficiency, and simultaneously optimize the effective optical path and mirror utilization, enhancing global optimization stability.

This invention relates to the field of tunnel boring machine (TBM) design technology and discloses a tool layout method for a circular cutterhead of a TBM. The method includes: constructing a mathematical model of tool layout incorporating cutting force data from a CSM model; determining the optimization objective and constraints for minimizing eccentric force and eccentric moment; initializing a gray wolf population by mapping discrete permutations to continuous vectors using a continuous encoding strategy and calculating fitness values; executing a global iteration of the gray wolf optimization algorithm, using social hierarchy to guide population updates; embedding a tabu search strategy in the global iteration to perform local neighborhood search on the optimal individual to generate and selectively replace the optimal solution; and finally outputting the tool layout scheme and verifying its feasibility. This invention effectively solves the problem of tool layout easily getting trapped in local optima through the synergistic cooperation of the gray wolf algorithm and tabu search, improving the mechanical balance and solution efficiency of the cutterhead design, and reducing design and production costs.

The application discloses a kind of optimal construction sequence determination methods of large-span space truss structure.The method includes: establishing the numerical simulation model of large-span space truss structure, and it is divided into multiple structure blocks;Joint state space including lifting node and construction time sequence is constructed, and the total strain energy area obtained by numerical integration of total strain energy of large-span space truss structure in construction process to cumulative characteristic displacement is established as evaluation objective function;Global iterative solution is carried out in joint state space using preset optimization algorithm, and the optimal joint strategy that makes the total strain energy area minimum is output, to determine the initial optimal lifting point and the initial optimal construction sequence corresponding;Real-time acquisition of stress deformation data in actual construction to calculate actual total strain energy area, when deviating from theoretical benchmark reaches preset threshold, trigger optimization algorithm to dynamically correct construction sequence.The application breaks the traditional decision limit, realizes joint global optimization and dynamic digital twin closed-loop control in time and space dimensions, and improves construction safety.

The application discloses a power device packaging optimization method based on a BPNN-WOA fusion model, relates to the field of power device thermal management, and solves the problems that traditional optimization methods ignore parameter nonlinear coupling, are difficult to globally optimize, and are complex in processing discrete variables. The method selects parameters and generates sample data sets by standardization in the aid of a heat dissipation influencing factor database and a high-fidelity three-dimensional thermal simulation model, and obtains a BPNN thermal resistance prediction proxy model through training. A whale position vector of a WOA algorithm is mapped into a packaging parameter combination, and BPNN output is mapped into fitness values. The optimal parameter combination is obtained through global iterative optimization of the WOA. After inverse normalization, simulation verification is carried out. If the deviation is over the threshold value, the model is retrained by supplementing samples. The method can capture the synergistic effect between parameters, realize global optimization of a continuous-discrete mixed design space, significantly reduce the packaging thermal resistance, and improve the efficiency and precision of optimization.