256 results about "Basis function" patented technology

The invention relates to a thermal coupling over-reduced order prediction method and system for a multi-component assembled nonlinear system. The method comprises the following steps: collecting multi-scale physical field data; constructing an intrinsic orthogonal decomposition basis function space of a temperature field and a stress field, and establishing a double-field coupling constraint equation; constructing contact thermal resistance parameterized proxy models of a cylinder contact area, a bolt area and a free deformation area by adopting a domain discrete empirical interpolation method; constructing a parametric intrinsic mode tensor network to obtain a decline model which is used for realizing real-time reconstruction of a mode basis function through acquired tensor slices in an online stage; performing dynamic inversion based on a modal basis function reconstruction result, and outputting a predicted transient displacement field, a predicted temperature gradient field and a predicted contact stress field; and obtaining real-time parameters, and calculating a residual error with a corresponding prediction result so as to dynamically update the primary function and interpolation point distribution. Compared with the prior art, the real-time prediction of the transient thermal coupling of the multi-component contact system is realized on the premise of ensuring the precision.

The invention provides an earth and rockfill dam seepage pressure prediction method based on mechanism and data dual drive, and the method comprises the following steps: generating multi-working-condition seepage pressure data, optimizing a radial basis function neural network through an aurora optimization algorithm, and constructing an efficient proxy model; on the basis of a proxy model prediction result, fitting is carried out by combining measured data, and a seepage pressure prediction mechanism model based on a hysteresis effect function is established; with the predicted value of the mechanism model as a label, supervising the training model to learn the physical law of the seepage field; and freezing the physical feature extraction layer in the trained model, and finely adjusting the time sequence modeling layer by using actually measured data to realize approximation of an actually measured value. According to the method, mechanism driving and data driving are organically combined, the interpretability of a physical rule is reserved, complex factors which are not considered by a mechanism model are complemented by utilizing actually measured data, and the generalization ability of a deep learning model under extreme working conditions is remarkably improved while the prediction precision of the model is ensured.

The invention provides a traffic flow spatio-temporal feature adaptive extraction method based on a dynamic Kolmogorov-Arnold network, and aims to solve the defects of a traditional model in the aspects of dynamic spatio-temporal modeling, structural adaptability and feature expression efficiency. According to the method, a network structure is dynamically adjusted through differential topology search and a parameterized primary function library, a double-flow coupling architecture is designed to extract space and time dependent features respectively, and adaptive fusion of spatial and temporal features is realized by using a gating mechanism. The method comprises the following specific steps: performing space-time normalization and graph structure coding to generate a node feature matrix; extracting multi-scale spatial-temporal features in parallel by the dynamic graph convolution KANs and the time sequence convolution KANs; feature alignment and cooperative enhancement are realized based on a local correlation matrix and a bidirectional interactive attention mechanism; and the lightweight KAN decoder is combined with the dynamic basis function library to output a prediction result. Experiments show that the MAE is reduced to 1.78 vehicles per minute and the reasoning speed is improved by 1.8 times in a traffic accident emergency scene on a PeMS08 data set, and the method is suitable for an intelligent traffic management and vehicle-road cooperation system.

The invention relates to the field of remote sensing image processing, and provides a remote sensing image terrain radiation correction method, device and equipment, and the method comprises the steps: obtaining a to-be-corrected remote sensing image, calculating the terrain parameter of each pixel, dividing the to-be-corrected remote sensing image into a shadow region and a non-shadow region, and determining the direction weighting function of each pixel; calculating a topographic modulation spherical harmonic primary function corresponding to each pixel; training samples are selected in the shadow area and the non-shadow area respectively, a linear model is constructed, and spherical harmonic illumination coefficients representing overall three-dimensional illumination distribution of the shadow area and the non-shadow area are solved; and based on the topographic modulation spherical harmonic primary function, combining the spherical harmonic illumination coefficient corresponding to the area to which each pixel belongs, reconstructing the average incident radiance of each pixel, and then performing inversion to obtain the real reflectivity of the earth surface. According to the method, the problem of inaccurate radiation distortion correction of the remote sensing image under the complex terrain in the prior art is solved, and accurate inversion of the real reflectivity of the earth surface is realized.

Examples described herein generally relate to alerting metric baseline behavior change. The examples include performing at least one of a radial basis function (RBF) kernel procedure and an autoencoding procedure for a time-series data; determining whether one or more change points occur in a seasonal pattern of the time-series data based on at least one of the RBF kernel procedure and the autoencoding procedure; and transmitting, to a user, an alert indicating the one or more change points based on a determination that the one or more change points occur in the seasonal pattern of the time-series data.

The invention discloses a parameter optimization method for a rotary inertia double-tuned mass damper and a vibration reduction system. The parameter optimization method comprises the steps that S1, a main structure motion equation comprising the rotary inertia double-tuned mass damper is constructed; s2, selecting combined variables from a design variable space of the rotary inertia double-tuned mass damper, adopting a genetic simulated annealing algorithm for each combined variable, and optimizing parameters by taking minimization of the maximum acceleration frequency response of a main structure as an objective to obtain an optimal parameter data set; s3, cleaning an optimal parameter data set based on an IQR criterion; s4, constructing a correction factor, and establishing a parameter correction factor nonlinear mapping model through tensor basis function expansion; and S5, multiplying the theoretical closed solution by the parameter correction factor to obtain design parameters of the rotary inertia double-tuned mass damper. Compared with an analytical formula design, the RIDTMD designed by using the corrected formula has the advantages that the vibration reduction efficiency is improved by about 10% at most, and the asymmetric three-peak phenomenon of a core frequency band can be effectively eliminated.

The invention relates to the field of rotating machinery fault diagnosis, and discloses a bearing fault diagnosis method based on a physical perception KAM network, and the method comprises the steps: carrying out the discretization of a bearing vibration signal through a Gabor filter group based on physical prior initialization, and generating modal feature lexical elements with physical frequency band meanings; and inputting the lexical elements into a PC-KAM backbone network, calculating a hidden state vector by using a state space model branch, and dynamically adjusting the position of a primary function node of a Kolmogov-Arnod network branch to realize collaborative dynamic feature extraction. In the training stage, an orthogonal subspace constraint and physical perception low-rank adaptation fine tuning mechanism is introduced. And finally, searching a historical fault case, performing multi-modal fusion on the historical fault case and the deep feature sequence, mapping a fusion representation into a soft prompt, and inputting the soft prompt into a large language model to generate a diagnosis report. According to the method, the problems of poor physical interpretability of characteristics and few-sample diagnosis under variable working conditions are effectively solved, and the generalization and decision-making ability of a diagnosis system are improved.

The invention belongs to the field of automatic control, and particularly relates to a nonlinear system dynamic gain global trajectory tracking control method based on a self-adaptive neural network observer, which comprises the following steps: converting an uncertain nonlinear system model with unknown interference according to the requirement of output feedback control to obtain a first nonlinear system model; enabling an unknown nonlinear continuous function in the first nonlinear system model to be expressed as a function only containing an output signal and other system state estimation values; performing approximate approximation on an unknown nonlinear continuous function in the first nonlinear system model by using the first radial basis function neural network vector to obtain a second nonlinear system model; designing a second radial basis function neural network vector to construct a dynamic gain state observer of the second nonlinear system model, and obtaining an estimated value of a system state and an estimated value of an unknown nonlinear continuous function in the model; according to an inversion control method, defining a dynamic system tracking error index containing an intermediate virtual control signal, designing a Lyapunov function, and obtaining a change rate of a dynamic gain, a weight adaptive update rate of a second radial basis function neural network vector, and a control rate of the intermediate virtual control signal and a system input signal; different from most existing nonlinear system control methods, the method combines a novel dynamic gain neural network observer with an inversion controller with dynamic gain, and provides a robust global trajectory tracking control solution for a dynamic system operating under uncertain conditions.

The invention relates to the technical field of water supply pipe networks, and discloses a multi-pump-room collaborative intelligent water supply frequency conversion regulation and control method, which comprises the following steps: initializing sensor calibration and data acquisition, generating a standardized data packet by using Kalman filtering, constructing a differential algebraic equation model, and online updating a pipe network topological characteristic matrix in combination with a mixed basis function observer. Layered optimization control and frequency conversion cooperative regulation and control are driven, 30-100ms-level real-time solution of a differential equation is realized by means of a hardware acceleration module, a three-dimensional pressure cloud picture is finally generated, and early warning is triggered based on a current harmonic threshold value. The transient fault capturing capability is enhanced by dynamically adjusting the data acquisition frequency, and the problem of signal distortion caused by traditional fixed sampling is solved; resource allocation is optimized and calculated in combination with a pipe network topology model dimension reduction technology, and calculation power waste in large-scale monitoring is reduced; and dynamic reconstruction of the model is driven by abnormal data to realize system self-adaption, so that manual parameter adjustment deviation is avoided, and the prediction precision and the diagnosis reliability are remarkably improved.

The invention discloses a fracture pressure prediction and internal friction angle joint inversion method for a heterogeneous stratum. The invention belongs to the technical field of petroleum engineering and artificial intelligence cross application, well depth is used as an independent variable, the minimum / maximum horizontal stress, pore pressure and cohesion are combined to construct well depth mixed input, only multi-scale periodic basis function transformation is carried out on the well depth, and a linear well depth channel is reserved; establishing a rupture pressure depth mapping model, introducing an augmented Lagrange physical constraint containing dual variables and a dynamic threshold, adaptively adjusting a physical weight according to a physical residual error and a threshold difference value, realizing dynamic balance of a rupture criterion and data fitting, and searching and optimizing a model topology by using a multi-target network structure; under a limited sample, a fracture pressure profile and a physically reasonable internal friction angle can be obtained at the same time, and the prediction precision and the physical consistency are improved.

The invention discloses a split type three-box girder nonlinear vortex vibration steady-state amplitude intelligent prediction method, belongs to the field of bridge vortex-induced vibration control, and aims to solve the problem of low accuracy of three-box girder vortex vibration nonlinear modeling and amplitude prediction. The method comprises the following steps: acquiring displacement time sequence data through a reduced scale model wind tunnel test, preprocessing to obtain displacement, speed and acceleration dimensionless data, and dividing a data set; constructing a candidate function library containing high-order polynomial terms of the primary function; performing nonlinear system sparse recognition by taking acceleration data as a target item and combining an improved algorithm to obtain a preliminary feature set; a final feature set is obtained through energy-statistics-self-adaption three-stage screening; and finally, constructing a control equation to predict the steady-state amplitude. The method can accurately predict amplitude, coincide test and real bridge observation results, and is less in computing resource occupation and high in numerical value precision.

The invention provides a temperature distribution inversion method based on a multi-vertex grid and a high-robustness error functional, and the method comprises the steps: 1, obtaining a temperature gradient change rule of a measured structure under a typical thermal load condition, and determining a three-dimensional multi-vertex grid division scheme and a temperature sensor layout form of the measured structure; step 2, establishing an m vertex grid basis function model, and constructing a conversion matrix between a regular grid and an irregular grid; 3, establishing a high-robustness error functional between the theoretical temperature of the centroid position of each surface of the m-vertex grid and the actually measured temperature of the sensor; 4, a temperature matrix equation containing a temperature coefficient matrix and a temperature amplitude matrix is constructed, temperature vectors of all grid vertexes are solved, the m-vertex grid basis function model is combined, three-dimensional temperature distribution in the structure space is inverted, and according to the method, only few sensors need to be adopted, the global temperature response of the structure can be obtained, and the inversion precision is high; the method is simple, convenient and high in real-time performance.

The invention belongs to the field of magnetic target body positioning, and relates to a measuring line vertical gradient-based magnetic multi-target positioning and magnetic moment inversion method, which comprises the following steps of: obtaining a unit orthogonal primary function and a primary function coefficient corresponding to each component of magnetic anomaly gradient data vertical to a measuring line, and calculating an energy curved surface of the magnetic anomaly gradient data vertical to the measuring line; detecting a local energy peak value of the energy curved surface, obtaining an energy peak value ratio sequence, obtaining a target number and horizontal position estimation corresponding to the target according to an energy sudden change value in the energy ratio sequence, and calculating a target depth corresponding to a measuring line vertical magnetic anomaly gradient energy extreme value by using a dichotomy method, and constructing a linear equation set about the target magnetic moment by using the measuring line vertical magnetic anomaly gradient data and the unit orthogonal basis function, and obtaining target magnetic moment parameter estimation through a least square method. According to the invention, the anisotropy difference of the estimation precision of the horizontal position of the target is effectively reduced, and the problem of target detection omission caused by large difference between the target burial depth and the magnetic moment module value in multi-target positioning is improved.

The invention provides a frequency dispersion curve calculation method based on a decoupling type Legendre-Gauss-Labatto spectral element method, and belongs to the technical field of discrete solution of guided wave propagation in a thin plate, and the method comprises the following steps: S1, constructing a wave equation which comprises a Cartesian space coordinate system based on an orthotropic plate, and combining a constitutive equation, a geometrical relationship under the coordinate system and guided wave displacement to obtain a wave dispersion curve; substituting into a control equation when the volume force is zero to obtain an improved three-dimensional elastic wave equation; s2, a differential matrix is constructed, wherein discretization and decoupling are carried out on a wave equation on the basis of a Lagrange interpolation basis function of a Legendre-Gauss-Labatto node; s3, explicitly embedding the stress free boundary condition into the solving process through matrix line transformation, and solving to obtain a frequency dispersion curve, the derivation process is simplified, the boundary and the interior of the material are decoupled, and the physical significance is clearer. And meanwhile, the high-precision advantage of the spectrum method is maintained, and the calculation rate is remarkably improved.

The invention discloses a shear wave velocity prediction method, device and equipment and a readable storage medium, and is applied to the technical field of geophysics, and the method comprises the steps: constructing a shear wave velocity prediction model in advance based on a KAN network and a sequence displacer; an activation function of the shear wave velocity prediction model is a B-spline primary function; obtaining logging data of the target area; and inputting the logging data into the shear wave velocity prediction model to obtain a prediction result. The shear wave velocity prediction model based on the KAN network has robustness and accuracy, the model is combined with a sequence displacer architecture to be fused with the KAN network to perform feature extraction, efficient modeling of logging data is realized through a dynamic weight fusion mechanism and a learnable B-spline basis function, and the prediction accuracy is improved. The model can be used in various large oil fields in a large scale, deals with various complex geological conditions and accurately predicts the shear wave velocity, so that the exploration and development risk is reduced, and the benefit development of each large oil field is guaranteed.

The invention aims to provide an electromagnetic parameter extraction method based on a non-conformal partial element equivalent circuit method, and belongs to the technical field of computational electromagnetism. According to the method, the multi-scale electromagnetic structure is divided into different areas according to the size difference, each area is subdivided by adopting a triangular prism grid with the corresponding size, and meanwhile, an h-SPV primary function is defined at the non-conformal junction, so that the continuity of current at the non-conformal junction is ensured. According to the method, in the calculation process of the multi-scale structure problem, a large number of transition grid units at the interfaces of different areas are avoided through non-conformal grid subdivision, and the calculation complexity is reduced while the calculation precision is guaranteed by establishing the current continuity boundary conditions at the interface ports.

This disclosure describes a set of advanced video coding technologies and is particular related to an intra warp prediction of a sample in a video block where the sample is intra predicted using a weighted sum of basis functions of pixel coordinate position of the sample. A set of weighting coefficients for the weighted sum may be predefined or may be derived by an encoder and signaled in a bitstream containing the video block at various coding levels.

Various embodiments of this disclosure relate to generation and reception of signals comprising cyclically shifted basis functions. A device may obtain a set of modulation symbols and modulate a plurality of basis functions based on the set of modulation symbols to generate a signal. The plurality of basis functions may comprise cyclically shifted versions of a basis function and a periodic autocorrelation function of the basis function may comprise an ideal (e.g., perfect) periodic autocorrelation function.

The invention belongs to the technical field of hydraulic engineering digital twinning, and provides an arch dam static response rapid calculation method which comprises the following steps: sampling and simulating based on a finite element model in an offline stage to obtain a displacement field snapshot matrix, extracting a dominant mode through intrinsic orthogonal decomposition to construct a reduced-order subspace, and calculating a mode coefficient; and training a radial basis function neural network to establish a nonlinear mapping relation by taking the working condition parameters as input and the modal coefficient as output, and finally, inputting the target working condition parameters into the trained network to predict the modal coefficient in an online stage, and quickly reconstructing a complete displacement field response through linear combination with the POD modal. The method achieves the quick and accurate prediction of the static physical field of the arch dam, and remarkably improves the calculation efficiency of the displacement field of the arch dam.

The invention discloses a method and device for compensating the dispensing height through curved surface fitting. The method comprises the steps that three-dimensional coordinate point information of a plurality of discrete sampling points on the surface of a to-be-dispensed target is collected; building a polynomial of a curved surface model for compensating dispensing according to the number of the sampling points; constructing a matrix, and constructing a linear equation set based on the primary function; and the weight coefficient of the linear equation set is solved through a Gaussian elimination method, a fitted quadric surface equation after interpolation calculation is obtained, and the dispensing height is compensated based on the quadric surface equation. According to the method, the polynomial is fitted through the three-dimensional coordinate point information, the linear equation set is constructed, the predicted compensation value is obtained after interpolation calculation is carried out according to the fitting result so as to compensate the dispensing height, the high-precision continuous curved surface model can be constructed by collecting discrete sampling points, real-time interpolation calculation of any coordinate is supported, and the precision of the dispensing height is improved. The method can solve the problem of insufficient fitting precision of dispensing height compensation based on a single plane.

The invention discloses a front subframe structure MaOP method and device based on TPT, a medium and a program product, and the method comprises the steps: (1) constructing a MaOP design model capable of optimizing the weight, modal, strength and rigidity at the same time based on front subframe longitudinal beam and load analysis; (2) generating an elite population based on a diversity criterion and Latin hypercube, obtaining target values of the elite population, establishing a data set, and constructing a radial basis function model; (3) designing co-evolution operation driven by TPT to generate a candidate frame set; (4) an individual potential evaluation criterion is constructed based on the advantage and disadvantage target sets, and an optimal candidate frame is screened; and (5) obtaining each target value of the optimal candidate frame, updating the data set and the radial basis function model, returning to the step (3) until all optimization targets meet requirements, and outputting an optimal parameter value. According to the method, a TPT-driven coevolution mechanism is adopted, radial basis function model prediction is combined, and the multi-performance index optimization process for the front auxiliary frame can be effectively balanced.

The invention discloses a mixed time-varying reliability analysis method for a solid rocket engine shell structure. Uncertain parameters and distribution are determined according to complex characteristics such as high-temperature and high-pressure time-varying load borne by a solid rocket engine shell structure, geometric sudden change in a star hole charging combustion stage, a throat ablation non-stationary interval process and a time-varying correction coefficient random process, and the random and interval processes are processed in a unified mode through equivalent time-varying uncertainty conversion. Input dimension explosion is avoided, specifically, a random process vector and an interval process vector are converted into an equivalent random vector, so that a time-varying limit state function is equivalently converted into a static function, the input dimension of the RBF combination model is effectively prevented from being increased, the advantages of multiple basis functions are fused through the radial basis function RBF combination model, the calculation cost is reduced, and the calculation efficiency is improved. The modeling precision of the strong nonlinear limit state function of the solid rocket engine shell is improved, the nonlinear mixed time-varying reliability analysis problem is efficiently solved, and the calculation precision is high.

The invention belongs to the field of real-time monitoring, particularly relates to a chemical process real-time monitoring method and system based on digital twinning, and aims to solve the technical problem that an existing method lacks an online model correction and updating mechanism. The monitoring method comprises the following steps: S1, constructing a state snapshot data set, and synchronously collecting real-time measurement data; s2, identifying a process fluctuation interval, and selecting to-be-selected state snapshots to form an optimal snapshot matrix; s3, calculating a reduced-order truncation error energy ratio, and constructing a reduced-order model based on a reduced-order basis function; s4, in the real-time monitoring stage, if the norm of the prediction residual error is smaller than a monitoring threshold value, a Kalman filtering algorithm is adopted to correct the state coefficient of the reduced-order model; otherwise, calculating the projection error of the prediction residual on each primary function in the standby primary function library; and outputting a full-order state vector reconstructed based on the corrected or updated reduced-order model. The monitoring method provided by the invention ensures the continuous effectiveness and accuracy of monitoring when the chemical process changes.

The invention belongs to the technical field of space geodetic survey observation data inversion regional land water reserve change, and provides a method and related equipment for inversion of land water reserves based on a Slepian basis function combined with GNSS three-dimensional deformation, and the method comprises the steps: obtaining vertical and horizontal deformation observation values of GNSS through simulation; determining a Slepian primary function of the target area according to the boundary data and a preset maximum expansion order, and further determining a function relationship between a Slepian coefficient of GNSS three-dimensional deformation and a Slepian coefficient driven by mass load change; determining an observation equation according to the function relationship, and resolving the observation equation by adopting least square joint adjustment; and according to the Slepian coefficient driven by the mass load change, carrying out inversion on the regional land water reserves of the target region to obtain a regional land water reserve change result. The method has the beneficial effect that the precision and efficiency of regional land water reserve inversion are improved.

The invention relates to the technical field of pet food spectrum detection, and discloses a pet food spectrum detection device and a detection method. Sampling at equal intervals in a preset wavelength range, and eliminating noise interference through dark current and standard whiteboard calibration; taking a natural logarithm of the reflectivity to balance a dynamic range; constructing a linear spline basis in an equidistant and segmented manner, and adaptively calculating a regularization coefficient according to the energy ratio of the signal to the basis function; constructing an augmented matrix according to a basis function and a signal inner product, and solving a spline coefficient through Gaussian elimination; reconstructing a spectrum by using the coefficient and quantifying a residual error; establishing a reference model based on the coefficient mean value and the standard deviation of the multiple qualified samples; comparing the Euclidean distance between the to-be-detected sample coefficient and the model mean value with a threshold value, and generating a residual error, a distance, a threshold value and a judgment report; empirical parameters or manual adjustment and optimization are not needed in the whole process, online adaptive detection in batches can be achieved, and noise correction, feature enhancement, over-fitting suppression and traceable rapid and accurate detection are achieved.

The invention discloses a load curve decomposition method and system based on a curvature equipartition rational B-spline basis function, and the method comprises the steps: calculating the curvature of a historical load sequence, generating a continuous curvature function, determining a group of B-spline node sequences meeting the cumulative absolute curvature equipartition condition, and enabling the node distribution to be adaptive to the load fluctuation intensity; a B-spline primary function is generated based on the node sequence, and a rational B-spline primary function is constructed through weight optimization so as to finely describe a continuous time fluctuation mode; a fitting model is established in combination with time period virtual variables, and a load curve is decomposed into two parts, namely a time period characteristic load level representing different time period reference power consumption intensity and a time characteristic load curve representing a continuous change fluctuation form. The defect that a traditional uniform spline is poor in adaptability to a non-uniform load form is overcome, self-adaptive decomposition with definite physical significance is achieved, and a better-quality data basis is provided for load analysis and prediction.

The present application belongs to the technical field of structure optimization design, and discloses a macro-micro double-scale structure topology optimization method and equipment capable of controlling the minimum size of microstructure, comprising the following steps: (1) interpolating the elastic matrix and the corresponding density of the actual microstructure in the offline database by using radial basis function interpolation, and then constructing a mapping model; (2) constructing a compliance minimization optimization model, wherein the design variables of the optimization model are the cutting height of two virtual microstructures in each unit and the topology variable, and the constraint functions are the volume fraction of the overall structure and the topology variable constraint; (3) calculating the sensitivity of the objective function and the constraint function to the design variable, and then updating the design variable; (4) when the compliance error of the updated double-scale structure meets the optimization termination condition, the optimization is completed and the double-scale structure meeting the requirements is obtained; otherwise, go to step (3). The present application solves the problems of low calculation efficiency and poor connectivity.

The invention discloses a radial basis function boundary optimization and multivariate quadratic quasi-interpolation combined data fitting method, which comprises the following steps of: firstly, selecting m data points at two ends from a given data set, fitting boundary data by using Gaussian radial basis function interpolation, and adjusting function characteristics through shape parameters; and fitting all data including the boundary data by using a multivariate quadratic quasi-interpolation method, calculating a weight function of the data and the corrected data, and combining a boundary fitting function to obtain a final result. The method provided by the invention solves the problems that the traditional interpolation calculation is complex, the numerical value is unstable and the boundary processing precision is low, the calculation complexity is low, and the numerical value is stable and reliable; the application effect in the fields of mechanical design, fluid dynamics simulation and the like is remarkable, compared with traditional polynomial fitting, the boundary fitting error is reduced by at least two orders of magnitude, and the calculation efficiency is greatly improved.

The invention discloses a rapid prediction method for a particle thermal-mechanical coupling process, and the method comprises the steps: scanning energy carrier particles based on a microfocus computed tomography technology, and generating a three-dimensional digital model; importing the three-dimensional digital model into a multi-physics field simulation platform, and establishing a transient full-order model; based on the transient full-order model, generating a training data set and a test data set under different working conditions and corresponding temperature fields and stress fields; performing decomposition calculation on the training data set by using an intrinsic orthogonal decomposition POD reduced-order model to obtain a POD primary function and a spectral coefficient; establishing a mapping model from an input working condition to an output spectral coefficient based on a BP neural network; inputting the target working condition into the mapping model, and outputting a POD spectral coefficient; and on the basis of the calculation result and the POD spectral coefficient, reconstructing full-field physical quantity distribution at any moment under the target working condition through linear combination. The method solves the problems that traditional numerical simulation is low in calculation efficiency and large in resource consumption.

The invention provides a numerical feature embedding method based on a scale perception radial basis function and a situation awareness method for the health degree of power equipment, and the method comprises the steps: obtaining a logarithm of to-be-coded numerical data x, and obtaining a base number D and an index L; performing RBF (Radial Basis Function) expansion on the base number D to obtain the expression of the base number D; carrying out soft sub-bucket distribution on the index L to obtain a distance from the index L to each soft sub-bucket, taking the distance as a coefficient of the soft sub-bucket, and weighting each soft sub-bucket according to the coefficient of the soft sub-bucket to obtain representation of the index L; and converting the representation of the index L into two numbers by adopting a gated linear layer network, and scaling the representation of the base number D to obtain the coded representation of the numerical data x.