57 results about "Variogram" patented technology

The invention relates to the technical field of mining planning and management, in particular to a mine geological resource reserve management method. The method comprises the following steps: collecting mine geological multi-source data, and carrying out multi-source data preprocessing to obtain fused geological data; geological semantic feature extraction is carried out on the fused geological data, semantic reconstruction is carried out, and semantic geological units are obtained; constructing a variation function by using the fused geological data; performing reserve probability calculation according to the variation function to obtain unit probability data; performing Bayesian inference according to the unit probability data, and performing unit reserve probability generation to obtain a unit reserve probability; mining unit construction is conducted on the semantic geological units, and mining units are obtained; and according to the mining unit and the unit reserve probability, constructing a mining optimization objective function to obtain a comprehensive objective function. The change of mine resources is accurately reflected through the probability reserve assessment method, the resource utilization rate is improved, and the production cost is reduced.

The invention provides a dynamic geologic model updating and predicting method based on tunnel face information, which constructs and dynamically updates a geologic model by combining face grid point data acquired in a tunneling process and adopting Kriging interpolation and variation function fitting technologies. The method comprises the steps of collecting tunnel face data, cleaning standard data, and extracting a starting end face and a terminating end face of a tunnel; generating an initial geologic model by using the three-dimensional variation function model and Kriging interpolation; newly added tunnel face data are integrated, model parameters are dynamically adjusted, and geological distribution is updated; and finally, displaying the dynamically updated geologic model through a three-dimensional visualization technology. The problem that a traditional static model is difficult to adapt to dynamic changes of geological conditions in the tunneling process is solved, the geological model can be optimized in real time, accurate and reliable geological information support is provided for tunnel construction, and construction efficiency and safety are remarkably improved.

The invention discloses a three-dimensional convolution method for decoding an imaginary language electroencephalogram topographic map. The method solves the problem that for a multi-rhythm electroencephalogram topographic map, an existing convolutional network method is insufficient in modeling capability, so that the decoding precision of the multi-rhythm electroencephalogram topographic map in an imaginary language is remarkably reduced. In the aspect of rectangular BEAM reconstruction, a frequency band specificity self-adaptive variation Kriging interpolation model is built, an optimal variation function is automatically selected according to spatial variation characteristics of electroencephalogram power of each frequency band, a high-resolution BEAM graph sequence under multiple frequency bands and multiple time steps is built, and spatial continuity and CNN structure adaptability are considered. A double-branch three-dimensional convolutional neural network is designed, high-dimensional feature extraction is performed on space-time and space-frequency band tensors, a multi-scale structure and dynamic expression ability of neural activity are mined, discrimination of two feature vector branches is dynamically weighted, and the recognition ability of a classifier to a language imagination electroencephalogram mode is improved.

The invention relates to the technical field of seismic inversion, and particularly discloses a variational reasoning full-waveform inversion method and device based on geological prior information, and the method comprises the steps: constructing a plurality of filters; acting the plurality of filters obtained by construction on a seismic migration section to calculate a correlation coefficient matrix; according to a logging information fitting theory variation function, performing low-rank approximation on the calculated correlation coefficient matrix to obtain a correlation coefficient matrix after low-rank approximation; randomly generating a plurality of speed samples based on the correlation coefficient matrix after low-rank approximation; calculating the gradient of the objective function according to the plurality of speed samples; according to the gradient of the objective function, calculating the gradient of posteriori distribution by using a Bayesian formula; and based on the gradient of the full waveform inversion posteriori distribution, utilizing a Stein variational gradient descent algorithm to carry out multiple times of iterative updating on the plurality of speed samples, and outputting a target speed model obtained by inversion of the plurality of speed samples. According to the scheme, the multiplicity of inversion is reduced, and the inversion efficiency is improved.

The invention relates to the field of environmental analysis, and provides a gas hydrate three-phase coexistence region modeling method and system. The method comprises the following steps: acquiring three-dimensional seismic data and original logging data, and preprocessing the original logging data to obtain corrected logging data; respectively performing logging interpretation and structural interpretation according to the corrected logging data and the three-dimensional seismic data to obtain interpretation results; converting a structure interpretation result in the interpretation results into a three-dimensional geological structure model, and performing grid division on the research area by adopting a corner grid system to obtain a structure grid model; and performing attribute modeling on the constructed grid model based on variation function analysis, and predicting porosity and saturation in a three-dimensional space through a Kriging interpolation algorithm to obtain a three-phase coexistence region model. According to the method, the influence of complex geological conditions on the accuracy of the model is reduced, and the precision of the three-phase coexistence region model is remarkably improved.

The invention belongs to the technical field of urban meteorological data processing and spatial information calculation, and particularly relates to an urban meteorological gridding data construction method and device based on spatial interpolation. Comprising the following steps: clustering meteorological data: carrying out partition optimization by utilizing an objective function during clustering; partition interpolation: according to a meteorological data clustering result, carrying out spatial division and processing on regions, carrying out statistics on the number of base stations for each region, and determining region types based on the spatial distribution density of the base stations; determining a global variation function and a region variation function for each region, and calculating an estimated value of a to-be-interpolated point by adopting a corresponding Kriging interpolation method for different region types based on the variation functions; region junction point interpolation: for points near different region junctions, interpolating by using the meteorological data of the related region base stations, and weighting to obtain a final result; finally, urban meteorological gridding data construction based on spatial interpolation is realized.

The invention relates to the technical field of high arch dam deformation field feature analysis, in particular to an interpolation algorithm for high arch dam deformation field feature analysis. According to the technical scheme, the method comprises the steps of data collection and classification, variation function initial parameter setting, variation function fitting through a local weighted least square method, variation function parameter global optimization, Kriging interpolation calculation and consistency checking and parameter adjustment, in the data collection and classification process, monitoring data of the high arch dam under different load working conditions are collected, and the monitoring data of the high arch dam under different load working conditions are obtained. The load working conditions comprise a high water level in a flood period and a low temperature period in winter, and the deformation data of different parts and different time from the total station, the level gauge, the inclinometer and the strain gauge are associated with the corresponding load working conditions and monitoring point coordinate information. Through comprehensive data association, reasonable parameter setting and optimization, accurate interpolation calculation and a reliable result adjustment mechanism, more accurate deformation field analysis of the high arch dam under different load conditions is realized, and safe operation of the dam is guaranteed.

The invention discloses a tunnel geological information modeling and dynamic updating method and system. The method comprises the following steps: firstly, collecting drilling data before construction and stratum information exposed by a tunnel face in a construction process, defining a lithology category set after standardization processing, and converting the lithology category set into an indication variable; empirical semi-variance is calculated based on the standardized data set, and variation parameters are obtained through variation function fitting; determining a neighborhood radius according to the variation parameter, and constructing an initial geologic model by adopting an indicating Kriging method; in the tunneling process, newly revealed stratum information is fused in an incremental mode, variation parameters are adjusted in a self-adaptive mode, a geologic model is updated, and meanwhile uncertainty quantification is conducted; and finally, determining the most probable lithology category of each spatial position based on the updated geologic model, and generating a lithology distribution diagram. According to the method, lithology category probability prediction and geologic model dynamic updating are realized, tunnel geologic spatial features can be accurately described, prediction uncertainty is quantified, and prediction precision and reliability are improved.

The invention relates to the technical field of gold mine geological modeling exploration, in particular to a gold mine modeling and reserve estimation method under multi-source information constraint. Geological images and exploration data of the target mining area are collected; performing interpolation operation according to the geological image in combination with the elevation data and the drilling data to establish a geological model; establishing a fracture model by adopting a deterministic modeling method according to the geological image; constructing a lithologic model by adopting an envelope surface modeling method according to the drilling data and the geophysical exploration data; constructing a gold ore body model by adopting an envelope surface modeling method according to the elevation data and the geophysical exploration data; obtaining mineralization information to construct a variation function model, and establishing a grade model by adopting a Kriging difference method according to the variation function parameter value of each fault; and estimating the gold ore reserves of the target mining area based on the established model. According to the method, the model can be established in combination with multi-source information, and the ore body structure can be accurately represented, so that the estimation precision of the gold ore is effectively improved.

The invention discloses a multi-period non-stationary hydrological variable valuation method, and relates to the technical field of hydrological prediction. Comprising the steps of obtaining hydrological variables of a plurality of observation stations in a drainage basin at different time points as hydrological time sequences of the observation stations; constructing an experimental variation function according to the hydrological time sequence of each observation station; selecting a drift model through a trend surface analysis method and model inspection; constructing a multi-period non-stationary variable equation set comprising a Lagrange multiplier and a hydrological variable weight based on the experimental variation function and the drift model; expressing the multi-period non-stationary variable equation set as a matrix form, and calculating a Lagrange multiplier and a hydrological variable weight through a matrix inversion method; and constructing an estimation formula according to the calculation result of the Lagrange multiplier and the hydrological variable weight, and performing multi-period spatial estimation on the hydrological variable of the to-be-measured drainage basin through the estimation formula. Dependence on the number of observation stations can be reduced, and accurate estimation can be achieved only through a small amount of observation station data.

The invention belongs to the technical field of data fusion and processing, and particularly relates to a multi-source data fusion analysis method and system for geological prospecting, and the method comprises the steps: 1, setting a plurality of detection points at the superficial part of a target geological prospecting region, marking a plurality of unknown points at the depth part of the target geological prospecting region, calibrating a position parameter of each detection point; collecting multi-source geological data of each detection point; 2, performing data fusion on the multi-source geological data of each detection point; 3, according to the position parameter of each detection point, constructing a multi-source composite covariance matrix of all the detection points in combination with a variation function; and 4, predicting multi-source geological parameters of the unknown point according to the multi-source composite covariance matrix. According to the method, the precision of geological parameter prediction is improved, the heterogeneity of the reservoir is effectively dealt with, and the stability and the physical reasonability of interpolation calculation are enhanced.

The invention discloses an intelligent spot check method, system and device for farmland construction quality and a storage medium, and relates to the technical field of agricultural monitoring. The method comprises the following steps: acquiring remote sensing environment data of a target area; the remote sensing environment data comprises a multi-temporal remote sensing image and ground environment data; inputting the remote sensing environment data into a high-standard farmland time sequence collaborative sampling model, and determining a spot check prediction result; the prediction result comprises a risk thermodynamic diagram and an uncertainty quantization map; the high-standard farmland time sequence collaborative sampling model is constructed based on a variation function model, a time sequence triggering mechanism and a Monte Carlo technology, and is used for dynamically optimizing a sampling strategy; and performing uncertainty analysis based on the spot check prediction result to generate a final spot check area. According to the invention, global precise monitoring and dynamic optimization regulation and control of high-standard farmland construction quality can be realized, and technical support is provided for long-term guarantee of farmland quality in black soil areas.

This invention discloses a method, apparatus, equipment, and storage medium for estimating geostress parameters in the longitudinal profile of a deep-buried, long tunnel, relating to the field of geological surveying technology. The method treats the tunnel geostress field as a random field and the measured geostress as several realizations of the random field. After obtaining the measured values ​​of geostress parameters from multiple points on the tunnel's longitudinal profile, the method first uses these measured data to fit multiple experimental variogram models to obtain model coefficients. Then, through error analysis, the optimal model is selected. Finally, using this optimal model, spatial interpolation calculations of geostress are performed to obtain the geostress results along the tunnel axis. This method eliminates the need for detailed and accurate geological data and soil parameters, and can quickly and efficiently obtain the geostress distribution results along the tunnel axis that meet the requirements and accuracy of tunnel engineering demonstration and design stages, using only a small amount of measured data from a few points. This shortens the required time and has significant theoretical and practical value.

Embodiments of the present application disclose a well-seismic joint reservoir prediction method, device and electronic equipment. The well-seismic joint reservoir prediction method comprises: selecting seismic attributes and performing standardization processing on the selected seismic attributes; obtaining a seismic attribute variogram plan based on the standardization-processed seismic attributes; establishing a relationship between the standardization-processed seismic attributes and reservoir parameters based on the seismic attribute variogram plan; and predicting the reservoir based on the relationship. The selected seismic attributes are used to obtain a seismic attribute variogram plan, thereby establishing a relationship between the standardization-processed seismic attributes and reservoir parameters, and the prediction of the reservoir is realized based on the established relationship between the standardization-processed seismic attributes and reservoir parameters. The seismic attributes are integrated into the prediction of the reservoir parameters, thereby improving the accuracy of the prediction results.

The invention discloses a multi-period non-stationary multi-hydrological variable spatial valuation method, and relates to the technical field of hydrogeology. Comprising the steps of obtaining a univariate time sequence; combining the univariate time sequences of all observation stations in the drainage basin to obtain a multivariate random time sequence; calculating a long-term average level of each observation station through a space-time random function to obtain a long-term trend; subtracting the multivariable random time sequence of each observation station from the corresponding long-term trend to obtain a random residual error; according to the random residual error, determining a covariance function and a variation function between any two observation stations in the drainage basin; and on the basis, a multi-period non-stationary multivariable space valuation model is constructed by combining a multivariable Kriging model theory, and multi-hydrological variable valuation is performed on the drainage basin to be measured. According to the invention, internal correlation between variables can be revealed, and the prediction precision and reliability can be effectively improved.

The invention belongs to the field of oil and gas field exploration and development, and discloses a numerical reservoir model optimization and history fitting method and related device.The method comprises the steps that firstly, geological exploration data are utilized, a spatial clustering algorithm is adopted to automatically determine the local updating range of a structural model, and structural layers and stratums are optimized through spatial interpolation; outputting an optimized construction model and uncertainty measurement; based on the model and logging lithofacies attribute data, adaptively identifying a lithofacies updating key area and variation function parameters by applying a machine learning algorithm, and performing lithofacies simulation to output an optimized lithofacies model; combining the porosity, permeability and saturation attribute data, adaptively determining an attribute optimization range and variation function parameters by adopting a probabilistic neural network, and outputting an optimized attribute model through phase control attribute modeling; a three-dimensional oil-water two-phase black oil seepage model is constructed, a global optimization algorithm is adopted to automatically adjust parameters based on a target function, simulation data is calculated until convergence, optimal parameters are output to complete historical fitting, and efficient and reliable intelligent support is provided for complex oil reservoir development.

The invention discloses a millimeter wave radar point cloud generation method with freezing mechanism static filtering and local interpolation angle measurement. According to the method, exponential moving average (EMA) static clutter suppression with a freezing mechanism is carried out on a distance-Doppler unit, and a smoothing coefficient is temporarily improved to freeze background update when a suspected human body scattering target is detected, so that a slow target is prevented from being filtered out; and calculating a coarse grid spatial spectrum for the candidate units by adopting adaptive diagonal loading two-dimensional minimum variance undistorted response (MVDR), reconstructing a fine grid spatial spectrum in a peak neighborhood based on local Kriging interpolation of a Gaussian variation function so as to obtain a high-resolution azimuth angle and a pitch angle, and finally generating a three-dimensional point cloud in combination with the distance. Compared with the prior art, the method has the advantages that clutter suppression robustness is improved and false alarm and leak detection are reduced under the complex indoor slowly-changing background, and meanwhile, the angle measurement resolution is improved on the premise that the calculated amount is controlled through local interpolation.

The application provides a differential pressure transmitter comprehensive compensation method based on Kriging interpolation and cubic spline interpolation, which comprises the following steps: measuring the calibration output voltage of a differential pressure transmitter at multiple temperature calibration points, static pressure calibration points and differential pressure calibration points to obtain data calibration points; fitting a variogram function theoretical model by using Kriging interpolation and obtaining a compensation matrix; establishing an output voltage model by using cubic spline interpolation; when the measured value of the ambient temperature is a temperature calibration point, the measured value of the static pressure is directly brought into the output voltage model to obtain a theoretical voltage, three calibration output voltages adjacent to the theoretical voltage are selected, a differential pressure function relationship is established by using cubic spline interpolation, and the measured output voltage is brought into the differential pressure function relationship to obtain a to-be-measured differential pressure; when the measured value of the ambient temperature is not a temperature calibration point, a compensation coefficient is determined by using the variogram function theoretical model, and the to-be-measured differential pressure is calculated by using the compensation coefficient. The application adopts cubic spline interpolation at the modeling temperature and adopts Kriging interpolation at the non-modeling temperature, so that the algorithm with large calculation amount can be used to improve the compensation precision, the repeated interpolation at each temperature and static pressure interval is avoided, the workload is reduced, and the real-time performance of the compensation is improved.

The invention belongs to the field of intelligent monitoring, and discloses an active and passive fusion clean room environment intelligent monitoring method, which constructs an active and passive fusion monitoring system based on covariable statistics. A two-parameter variation function is established by introducing real-time wind field difference as a key covariable, and the defect that a traditional model ignores flow field dynamics is overcome; a comprehensive blind area guiding index is constructed by using a deviation between a basic information field generated by a fixed sensor and a correction information field formed by robot fusion data and combining an inherent coverage risk and a joint space disparity degree, and an absolute monitoring blind area is accurately identified; therefore, the control mode of fixed route inspection to blind area driving type active detection is converted. According to the method, on the premise that the hardware cost is not increased, the full-domain monitoring coverage rate, precision and sudden pollution response speed of the clean room are remarkably improved.

The invention discloses a soil heavy metal pollution interpolation prediction method of a variation coupling graph attention mechanism, and relates to the field of soil heavy metal detection, and the method comprises the following steps: collecting a target area soil sample, and calculating the Euclidean distance between every two sampling points; dividing a plurality of distance intervals; carrying out empirical mode decomposition on the heavy metal concentration and verifying an intrinsic part; calculating semi-variation values of the heavy metal concentration intrinsic part under different distances; fitting a theoretical variation function model; constructing a graph structure; calculating an attention weight coefficient between the nodes; obtaining node-level mapping features; and taking the node-level mapping characteristics output by the last layer of GAT structure as the input of a pre-trained multi-layer perceptron, and outputting a soil heavy metal pollution interpolation prediction result. According to the method, the variation function and the distance are integrated into the GAT model, the understanding of the soil heavy metal pollution mode is enhanced, the method can adapt to different geographical conditions and pollution scenes, and a more accurate and reliable prediction result is provided.

Embodiments of the present application disclose a method and device for establishing an ultra-shallow velocity model, electronic equipment and a storage medium. The method comprises: obtaining micro-logging data of a target area, wherein the micro-logging data comprises at least one measuring point and a corresponding velocity value of each measuring point, and the velocity value is obtained by measuring the velocity at the corresponding measuring point; for each analysis direction in at least one preset analysis direction, analyzing the micro-logging data in the analysis direction to obtain a function parameter, and determining a variogram corresponding to the analysis direction according to the function parameter; performing interpolation processing on the micro-logging data according to the variogram corresponding to each analysis direction, to obtain an interpolation result, and establishing an ultra-shallow velocity model of the target area according to the interpolation result. The technical scheme of the embodiments of the present application can be used to establish an accurate ultra-shallow velocity model.

The invention discloses a method and a device for constructing a lithofacies model in a riverway region under the condition of few wells. The method comprises the following steps: determining a plane layout map corresponding to a target layer in a target area, determining a target river channel area corresponding to the plane layout map, and obtaining multiple groups of sampling data corresponding to the target river channel area; determining a main variable range value corresponding to the target river channel region based on the multiple groups of sampling data corresponding to the target river channel region and the variation function, and determining a secondary variable range value corresponding to the target river channel region based on the main variable range value and the size information of the central beach in the target river channel region; and determining a layer structure model and a variable range direction corresponding to the target region, and determining a riverway region lithofacies model corresponding to the target layer based on the layer structure model, the main range value, the secondary range value and the variable range direction. According to the method, the accurate construction of the lithofacies model of the river region under the condition of few wells can be realized, and the construction, operability and flexibility of the lithofacies model of the river region are greatly improved.

The present invention discloses a GRNN smoothing factor setting method and system based on a variogram model, comprising: preprocessing sampled data; obtaining an omnidirectional experimental variogram of the preprocessed sampled data, and fitting the omnidirectional experimental variogram to obtain a range value; using the range value as a smoothing factor to perform GRNN prediction; constructing a GRNN model based on the smoothing factor that satisfies the prediction result and the preprocessed sampled data, and predicting the grade value at an unknown location using the GRNN model. The present invention makes it easier to obtain the optimal smoothing factor, improves the efficiency of parameter search, and has greater adaptability.

The application provides a tunnel face information-based dynamic geological model updating and predicting method, which combines the tunnel face grid point data obtained in the tunnel excavation process, adopts Kriging interpolation and variogram fitting technology, and constructs and dynamically updates the geological model. The method comprises the following steps: collecting the face data, cleaning and standardizing the data, and extracting the tunnel starting end face and the ending end face; generating an initial geological model by using a three-dimensional variogram model and Kriging interpolation; integrating the newly added face data, dynamically adjusting the model parameters, and updating the geological distribution; and finally displaying the dynamically updated geological model through three-dimensional visualization technology. The application solves the problem that the traditional static model is difficult to adapt to the dynamic changes of geological conditions in the excavation process, can optimize the geological model in real time, provides accurate and reliable geological information support for tunnel construction, and significantly improves the construction efficiency and safety.

The invention relates to the technical field of geological exploration, and discloses a geological exploration drill hole number optimization method and device, equipment and a storage medium, and the method comprises the steps: obtaining the drill hole data of an existing drill hole in a target exploration region, including space coordinate data and geological attribute data, and constructing a variation function model representing the geological attribute space continuity; according to the method, the number of the drill holes is optimized, interpolation is carried out based on the model, the standard deviation of each interpolation point is calculated, the number of the drill holes is optimized according to the distribution condition of the standard deviation, the number of unnecessary drill holes can be reasonably reduced, or the drill holes are added for an area with high data uncertainty, and optimal configuration of the number of the drill holes is achieved. By means of the optimal configuration, exploration work is more efficient, and time and cost consumption caused by blind drilling is avoided.

The application discloses a tunnel geological information modeling and dynamic updating method and system. The method first collects drilling data before construction and stratum information exposed in a working face during construction, defines a lithology category set and converts the lithology category set into an indicator variable after standardization processing; calculates an empirical semi-variance based on the standardized data set, obtains a variation parameter through variogram fitting; determines a neighborhood radius according to the variation parameter, and constructs an initial geological model by using an indicator Kriging method; in the tunneling process, newly exposed stratum information is incrementally integrated, the variation parameter is adaptively adjusted, the geological model is updated, and uncertainty quantification is simultaneously performed; finally, the most possible lithology category of each spatial position is determined based on the updated geological model, and a lithology distribution map is generated. The application realizes lithology category probability prediction and geological model dynamic updating, can accurately depict tunnel geological spatial characteristics, quantitatively predict uncertainty, and improves prediction accuracy and reliability.

This invention discloses an intelligent monitoring method for vegetation in spoil heaps based on optimal layout, belonging to the field of open-pit mine ecological restoration monitoring technology. This invention employs spatial variogram and simulated annealing algorithms to achieve optimal monitoring point layout; it automatically acquires multi-angle images and videos using adjustable supports, and performs keyframe extraction, multi-angle registration, and adaptive denoising preprocessing; the images are input into a lightweight dual-stream feature fusion network, and vegetation areas are accurately segmented through RGB and NDVI dual-stream attention fusion and combined loss function optimization; based on the segmentation results and environmental factors, vegetation coverage, plant height, and tiller number are automatically calculated; the time series is decomposed using STL, and the logistic growth model is integrated into the LSTM network as a physical constraint to output a growth trend prediction that conforms to biological laws. This invention achieves fully automated monitoring from layout to prediction, significantly improving the accuracy and efficiency of monitoring reclaimed grassland in spoil heaps.

The present invention discloses a quantitative evaluation method for the stationarity of training images based on a variogram, which calculates the strength of the spatial stationarity of the training image through the variogram, thereby providing an objective quantitative index for judging whether the training image is suitable for modeling. First, the variogram is used to quantify the variability differences between different spatial anchor points and other areas of the training image to obtain an anchor point distance model. Then, the stationarity coefficient of the anchor point is calculated through the anchor point distance model, and finally, the stationarity coefficients of all distance models are statistically analyzed to form a quantitative index for the stationarity of the training image. The present invention aims to solve the limitations of the quantification of the stationarity of training images, greatly improve the accuracy and efficiency of image analysis, systematically evaluate the stationarity of training images, and provide a more accurate and reliable spatial variability analysis method for geological modeling.

A three-dimensional convolution method for imagined language electroencephalography (EEG) topography decoding. It solves the problem of the deficiency of the existing convolutional neural network method in modeling ability for multi-rhythm EEG topography, which leads to a significant decline in decoding accuracy in imagined language. In the aspect of rectangular BEAM reconstruction, a frequency-specific adaptive variogram kriging interpolation model is built, which automatically selects the optimal variogram function according to the spatial variation characteristics of the EEG power of each frequency band, and constructs a high-resolution BEAM map sequence under multi-frequency and multi-time steps, taking into account the spatial continuity and the adaptability of the CNN structure. A double-branch three-dimensional convolutional neural network is designed to extract high-dimensional features from spatial-time and spatial-frequency tensors, respectively, to explore the multi-scale structure and dynamic expression ability of neural activity. The discriminative ability of the two feature vector branches is dynamically weighted to improve the recognition ability of the classifier for the EEG pattern of language imagination.