20results about How to "Improve inversion efficiency" patented technology

The invention discloses a rock-fill dam material parameter uncertainty inversion method based on Bayesian inference, and the method specifically comprises the steps: building a rock-fill dam finite element model, and determining the prior distribution of to-be-inverted parameters; constructing an sPCE proxy model based on prior distribution of inversion parameters; a rock-fill dam displacement statistical model is established, and actually-measured displacement water pressure components are separated; constructing a rock-fill dam material likelihood function; carrying out uncertainty inversion on the rock-fill dam material parameters; and updating and inverting the uncertainty of the rock-fill dam material parameters. According to the method, a Bayesian inference method is utilized, and uncertainty distribution characteristics of different partition material parameters of a dam body are obtained through actual measurement displacement data, parameter prior distribution and a likelihood function of the dam. According to the method, the defect that parameter uncertainty cannot be considered in a traditional inversion method is overcome, and obtained parameter values are closer to engineering practice; compared with a traditional inversion method, parameter updating inversion can be carried out through monitoring data in different periods, and continuous tracking of parameter timeliness evolution is achieved.

The invention belongs to the field of intelligent construction, and discloses a deep foundation pit displacement prediction method combining finite element and data driving, which comprises the following steps: calculating a theoretical elastic modulus according to an investigation report, and constructing a sample space by adopting a Latin hypercube sampling method and an orthogonal experiment design method; a proper constitutive model is selected, and corresponding lateral displacement is calculated through finite elements; dimensionality reduction processing is carried out on the data set based on global sensitivity analysis of variance so as to accelerate convergence during subsequent inversion; training an SKRR-LSBOOST model through an ensemble learning method to replace a traditional model; combining an improved Hemma optimization algorithm with a substitution model, and applying the combined model to parameter inversion; and performing forward analysis based on the finite element model defined by the inversion parameters to predict the deformation of the deep foundation pit in the next excavation stage. According to the method, the advantages of numerical simulation and data driving are fully exerted, the calculation cost is reduced according to the rock and soil parameters, higher prediction precision is achieved, and the accuracy of foundation pit deformation prediction is improved.

The invention provides an aviation transient electromagnetic fast Bayesian inversion method, device and equipment, and relates to the technical field of geophysical signal processing and analysis, and the aviation transient electromagnetic fast Bayesian inversion method comprises the steps: obtaining a resistivity model and the height of a transmitting coil; the resistivity model and the transmitting coil height serve as input data, electromagnetic response data serve as output data, a deep learning network model is trained, and a forward modeling model is obtained; coupling the forward modeling simulation model to a parallel Bayesian inversion framework to generate a target network model; and performing resistivity inversion and quantitative uncertainty analysis on to-be-analyzed aviation transient electromagnetic data based on the target network model to obtain an analysis result. By integrating a deep learning forward simulator into a Bayesian inversion framework, rapid Bayesian inversion and quantitative uncertainty estimation of aviation transient electromagnetic data can be realized, and the inversion efficiency is greatly improved.

The invention discloses a particle energy spectrum inversion method and a particle energy spectrum inversion system based on double regular coefficients, so as to achieve the purpose of one-time inversion of electron and proton energy spectrums by using fewer detection devices. The detector is a single detector and detects electrons and protons at the same time, and the method comprises the following steps: generating a first inversion model based on the detector; wherein the first inversion model comprises a local relaxation coefficient and a biregular coefficient, and the biregular coefficient comprises a first regularization coefficient corresponding to electrons and a second regularization coefficient corresponding to protons; presetting a local relaxation coefficient set of local relaxation coefficients and a biregular coefficient set of biregular coefficients, performing iterative calculation on the first inversion model by adopting a Boosted-Gold algorithm based on the biregular coefficients on the basis of the local relaxation coefficient set and the biregular coefficient set, and recording an energy spectrum norm and a residual norm which meet a preset convergence condition; and generating an L curve based on the spectral norm and the residual norm, and taking an inflection point of the L curve as an optimal solution for inversion.

The invention discloses a crustal stress field inversion method and device based on Bayesian optimization and a medium. The inversion efficiency and precision can be improved. The method comprises the following steps: acquiring parameters generated by an agent model, wherein the parameters comprise boundary parameters and physical and mechanical parameters; the prior probability of Bayesian model parameters is obtained through calculation according to the parameters; simulation stress is obtained according to the parameter simulation, and a Bayesian model parameter likelihood probability is obtained based on the simulation stress and the corresponding actually measured stress; obtaining an objective function based on the prior probability and the likelihood probability, and adding the objective function and corresponding parameters into an optimization data set; adding the parameters and the objective function into a sample set of an agent model to update the agent model, and circulating after new parameters are obtained; and finally, determining a minimum value of an objective function in the optimized data set, and performing simulation based on parameters corresponding to the objective function to obtain a crustal stress field.

The invention discloses a cement stabilized macadam mesoscopic fracture parameter rapid inversion method, and the method achieves the multi-dimensional improvement of model scale selection, automatic data processing, machine learning model construction, parameter result optimization and the like through the construction of an inversion process of a discrete element model-XGBoost model-particle swarm optimization model frame. The inversion efficiency is improved; the time and energy consumed by repeated manual trial calculation are reduced; the multi-objective optimization advantage of a vectorization particle swarm optimization algorithm is fully utilized, and the vectorization particle swarm optimization algorithm is combined with an XGBoost model, so that the optimal solution of a target vector is realized; compared with a manual trial and error method, the method has the advantages that multi-dimensional targets can be optimized at the same time, so that a real test result is approached at the fastest efficiency, and the precision of the cement stabilized macadam microscopic fracture parameters is greatly improved.

The invention provides an insect body axial ratio inversion method and device based on RCS features and a medium, and belongs to the technical field of insect radars. The method comprises the following steps: determining an incidence relation between each directional diagram characteristic parameter and an insect body axial ratio based on an analysis result of a preset insect collinear polarization directional diagram, and determining a plurality of insect body axial ratio estimators based on the incidence relation; wherein the association relationship at least comprises the association relationship with the insect body length and the association relationship with the insect body width; the insect body axial ratio estimator is used for representing the quantitative relation between the insect polarization characteristic and the insect body axial ratio; according to each insect body axis ratio estimator and a preset XGBoost regression algorithm, constructing a pre-trained body axis ratio inversion model; wherein the body axial ratio inversion model establishes a mapping relation between each insect body axial ratio estimator and the insect body axial ratio; and deploying the body axial ratio inversion model to a user terminal, so that the user terminal performs insect body axial ratio inversion based on the input insect collinear polarization directional diagram and the body axial ratio inversion model.

The application discloses a reservoir intelligent inversion method and related device, relates to the technical field of reservoir intelligent inversion, and comprises the following steps: obtaining multi-source heterogeneous data of a target area and preprocessing; constructing an initial geological model generator based on a conditional diffusion model, and optimizing hidden space parameters in a manner of seismic wave forward simulation; constructing a reservoir knowledge graph and adaptively pretraining and migrating the model online; performing task demand analysis, task decomposition and dynamic collaborative execution on a reservoir modeling task based on a thinking chain technology and a language large model technology, and obtaining iteratively optimized hidden space parameters; inputting the iteratively optimized hidden space parameters into the optimized geological model generator, outputting a three-dimensional distribution model, and extracting a confidence interval feature; incrementally updating the reservoir knowledge graph, determining a reservoir model conforming to a geological mode feature and matching a seismic waveform, and realizing reservoir intelligent inversion. The application can improve the accuracy and efficiency of reservoir inversion.

The invention discloses a data decryption method and device based on an SP network structure, equipment and a medium, and relates to the technical field of information security, and the method comprises the steps: obtaining ciphertext data sent by a user side; the ciphertext data is data obtained after the user side encrypts plaintext data based on an SP network structure, a P box in the SP network structure is constructed based on a preset Vandermonde-like fusion matrix, and the Vandermonde-like fusion matrix is constructed based on an original Vandermonde matrix and a Vandermonde-like matrix obtained after the original Vandermonde matrix is transformed; acquiring encryption parameters respectively corresponding to an S box and a P box in the SP network structure; the encryption parameter comprises a Vandermonde-like fusion matrix; and performing inverse operation on the Vandermonde-like fusion matrix to obtain a target inverse matrix, determining a first decryption parameter corresponding to the P box based on the target inverse matrix, and decrypting the ciphertext data by using the first decryption parameter and a second decryption parameter corresponding to the S box to obtain plaintext data. According to the method, the inversion efficiency can be improved during decryption calculation.

The invention discloses a sound velocity-density inversion method in quantitative ultrasonic tomography and a related device, and relates to the technical field of medical imaging, and the method comprises the steps: firstly obtaining an original ultrasonic signal collected by the quantitative ultrasonic tomography as a to-be-processed ultrasonic signal, and then carrying out the reconstruction through a delay summation method to obtain an initial reconstruction image; and inputting the two into a sound velocity-density inversion model, and outputting a sound velocity distribution diagram and a density distribution diagram. The model is of a supervised double U-Net architecture and comprises double encoders, double decoders and a cross-path feature fusion unit, the double encoders extract ultrasonic signal physical features and initial image structural features respectively, the double decoders are responsible for reconstruction of two parameters respectively, and the cross-path feature fusion unit achieves feature interaction. According to the scheme, the calculation bottleneck of a traditional physical constraint method is avoided through the double-architecture design, the problem of unstable training is solved through cross-path fusion, the inversion precision, efficiency and stability are finally considered, and the physical consistency of reconstruction results is ensured.

The present specification provides a determination method and device of a multi-task inversion model and a multi-task elastic parameter inversion method. The determination method of the multi-task inversion model comprises: inputting pre-stack seismic data samples into the multi-task inversion model to obtain a plurality of different elastic parameter data; inputting the plurality of different elastic parameter data into a forward model to obtain predicted pre-stack seismic data; determining a pre-stack seismic loss according to the pre-stack seismic data samples and the predicted pre-stack seismic data; determining an elastic parameter loss according to label data corresponding to the plurality of different elastic parameter data and the pre-stack seismic data samples; adjusting parameters of the multi-task inversion model and the forward model according to the pre-stack seismic loss and the elastic parameter loss; wherein the adjusted multi-task inversion model is used to determine a plurality of different target elastic parameter data according to target pre-stack seismic data. The above method can improve the prediction accuracy and efficiency of different elastic parameter data.

The invention discloses an inverse kinematics solution method and device for a biased robot and a medium, and the method comprises the steps: firstly, carrying out the modeling of a biased robot as an unbiased robot, carrying out the inverse solution to obtain a joint angle approximate analytical solution, and setting the dynamic search range of an initial population and a particle swarm according to the approximate solution; taking the difference between the actual tail end pose and the tail end pose corresponding to the approximate solution as a fitness function, solving the minimum value of the fitness function, improving the adjustment mode of an inertia factor of the particle swarm algorithm, shortening the iterative calculation time, and finally solving an accurate numerical solution of the robot through particle swarm iteration to complete the inverse solution of the joint angle of the robot. According to the method, the inverse kinematics efficiency of the bias robot can be improved, and practicability is provided.

This invention discloses a method, device, equipment, and storage medium for geometric correction of hidden road defects based on mechanical drive. The method includes: constructing a finite element simulation sample library for road defects; extracting and fusing multidimensional features from the sample library to obtain a fused feature vector; using this vector to construct a dataset for training to obtain a proxy model for defect parameter inversion; locating abnormal areas of the road medium using 3D ground-penetrating radar and activating the corresponding strain sensor array; acquiring measured strain data and fusing it into a feature vector to be identified; inputting this vector into the proxy model for inversion; and outputting the geometric parameters of hidden road defects. This effectively combines the advantages of 3D ground-penetrating radar and strain sensor arrays, compensating for the shortcomings of single detection methods, and effectively supplementing the inaccuracy of ground-penetrating radar in detecting the boundaries of hidden defects. Through multidimensional feature extraction and fusion and neural network model training, the deep information of strain data is accurately mined, improving the estimation accuracy of geometric parameters of hidden road defects.

This invention relates to the field of bridge structural health monitoring and safety assessment technology, and particularly to a method for detecting the deflection deformation of orthotropic steel bridge decks based on influence surfaces. The method includes: S1, establishing a layered spatiotemporal traffic flow information database specific to the orthotropic steel bridge deck to be tested, generating virtual loads based on time windows and vehicle tail reinforcement coefficients to simulate bridge traffic; S2, acquiring the static stress response of key bridge points when vehicles cross the bridge; S3, inverting the influence surface based on dimensional adaptive prior information, constructing a hybrid likelihood function, and reconstructing a two-dimensional influence surface matrix by integrating typical traffic and tail reinforcement information; S4, comparing the changes and responses before and after the influence surface to infer the deflection deformation. This application can generate virtual loads using traffic flow statistics and invert the entire field influence surface through stress measurement at a single key point; the innovative dimensional adaptive prior setting method significantly improves the inversion efficiency, accuracy, and stability.

This invention discloses a method for decoupling and inverting the emission flow rate and concentration of thermally hazardous gases based on intrinsic orthogonal decomposition, belonging to the field of industrial building ventilation and air pollution control technology. It addresses the pain points of dual-parameter inversion of thermally hazardous gases, such as large sample requirements, low inversion accuracy, and inability to adapt to scenarios with significant disturbances in the flow field caused by pollution sources. This invention constructs a physical model of pollution source diffusion and divides the flow field region. Based on emission parameters and reconstruction error analysis, it determines the optimal deployment area for temperature sensors. Temperature field data is reconstructed through adaptive sampling and intrinsic orthogonal decomposition. The emission flow rate is obtained by matching and optimizing with measured temperatures, and the uniqueness of the inversion results is verified. Simultaneously, based on the actual flow rate obtained through inversion, the emission concentration can be further inverted using the linear proportional relationship of concentration. This invention offers high inversion accuracy and stable, reliable results, suitable for rapid and accurate inversion of thermally hazardous gas emission parameters in industrial scenarios.

The invention discloses a neural network alloy layer material mechanical property identification method based on a genetic algorithm. The method comprises the steps that 1, reasonable test parameters are selected according to a bearing bush alloy layer material microstructure and a bearing bush alloy layer material macrostructure; 2, the bearing bush alloy layer is tested through an indentation test, and an indentation force-indentation depth curve is obtained; 3, selecting a material constitutive model; 4, creating a neural network; 5, on the basis of the Latin hypercube theory, adopting Abaqus numerical analysis software to construct a network training database; 6, training the neural network by using a database constructed by Abaqus, and constructing a nonlinear mapping relation between the constitutive parameters of the material and the indentation force-indentation depth curve by the neural network through training; and 7, through a genetic algorithm, the least square error between the press-in force-press-in depth curves obtained through the test and the neural network calculation is made to be minimum, and therefore the mechanical property of the bearing bush alloy layer material is obtained.

This invention discloses a method and apparatus for crack morphology inversion combining full-time step and physical model, comprising: inputting crack height and crack offset in the crack height direction into a physical model; first, processing the crack height through a crack propagation sub-model to output crack geometric parameters for the full-time step; then, processing the crack geometric parameters and crack offset in the crack height direction through a fiber strain response sub-model to output axial strain rate at simulated measurement points for the full-time step; adjusting the axial strain rate at simulated measurement points based on positive and negative strain rate scaling factors; calculating the loss value between the adjusted simulated axial strain rate and the measured axial strain rate; if the loss value is greater than a preset loss threshold, adjusting the corresponding optimization control parameters until the loss value is less than or equal to the preset loss threshold, or a preset iteration stop condition is reached; and outputting the target crack geometric parameters for the full-time step based on the finally adjusted optimization control parameters. This invention can improve the accuracy and efficiency of crack morphology inversion.

The invention discloses an efficient inversion algorithm for large-scale geophysical data, which comprises the following steps of: reducing the number of observation data participating in inversion by using observation data constraint downsampling based on information monitoring before inversion, and realizing preliminary reduction of the scale of an inversion problem, a model updating direction is solved by adopting a precondition conjugate gradient method, a physical property model is updated by adopting an inversion-verification mode, and on the basis of a downsampling data set, the number of observation data participating in iteration is further reduced, and secondary reduction of an inversion problem scale is realized, so that high-efficiency and low-memory-occupation inversion of large-scale data is realized. According to the method, all data do not need to participate in inversion, the inversion precision is kept unchanged, more storage requirements and time consumption can be saved, and rapid inversion of large-scale geophysical data is realized by combining data downsampling and a training-verification iteration mode.

The invention relates to the technical field of seismic data processing, and relates to a simulated annealing nonlinear pre-stack inversion method based on tectonic dip angle constraint. Comprising the following steps: S1, dip angle information extraction: obtaining a stratigraphic dip angle theta through a plane wave deconstruction method; s2, constructing an inversion objective function: inverting an AVO model based on a Bayesian theory framework, introducing a tectonic stratigraphic dip angle theta as prior information, and forcing an inversion result to follow preset dip angle distribution by adding a penalty term in the objective function; s3, performing inversion calculation based on the objective function: introducing an adaptive temperature adjustment strategy in the calculation process, so that the inversion calculation can widely explore a solution space in the initial stage and can perform refined search in the later stage; meanwhile, a multi-scale inversion framework is introduced in the calculation process, inversion is carried out step by step from coarse to fine, the low-scale level is used for determining a large-scale structure, and the high-scale level is used for revealing detail features. According to the invention, the geological rationality and accuracy of inversion are significantly improved.

The present invention discloses an efficient inversion algorithm for large-scale geophysical data, including using observation data constraint downsampling based on information monitoring to reduce the number of observation data involved in inversion before inversion, so as to achieve a preliminary reduction in the scale of the inversion problem. The inversion algorithm is based on the Gauss-Newton iteration algorithm, and uses the preconditioned conjugate gradient method to solve the model update direction, and uses the inversion-verification mode to update the physical property model. On the basis of the downsampled data set, the number of observation data involved in the iteration is further reduced, so as to achieve a secondary reduction in the scale of the inversion problem, thereby achieving efficient and low memory usage inversion of large-scale data. The present invention does not require all data to participate in the inversion, keeps the inversion accuracy unchanged, can save more storage requirements and time consumption, and combines data downsampling and training-verification iteration mode to achieve rapid inversion of large-scale geophysical data.