107 results about "Field transformation" patented technology

The invention discloses a carrier roller fault diagnosis method based on multi-modal information fusion, and particularly relates to the field of carrier roller fault diagnosis. The method comprises the following steps: collecting various modal signal data of a carrier roller, and preprocessing original one-dimensional signal data; simplifying the one-dimensional time sequence data of each signal by using classification aggregation approximation; converting each signal into a two-dimensional feature image by using a Markov field and synthesizing the two-dimensional feature image into a three-channel image; a carrier roller fault diagnosis model based on a Swin Transform network architecture is established; the Swin Transform network architecture is trained and verified, if the network achieves the optimal effect, the next step is carried out, otherwise, the previous step is returned, and initial parameters are modified to continue training; and inputting the test set into the optimal Swin Transform network architecture to carry out carrier roller fault diagnosis, and finally obtaining a carrier roller fault classification result. According to the technical scheme, the problems that existing carrier roller fault diagnosis is based on single-mode information, and information is insufficient and inaccurate are solved, and a foundation is laid for follow-up control and maintenance implementation.

The invention provides a passive antenna array beam forming test method and system, and relates to the technical field of beam forming test. According to the method, channel baseline parameters are established in a silencing environment, frequency hopping and channel switching are combined to realize multi-frequency-point synchronous sampling, a reference source and self-polling data are utilized to calculate a mutual coupling correction matrix and a feeder line consistency factor, and weight projection and equivalent aperture field generation are completed without changing physical wiring conditions; further extracting a main lobe direction, a beam width, a side lobe level and a scanning consistency index through far field transformation, and unwrapping and smoothing a frequency-point-crossing phase result; and meanwhile, generating a unique traceability identifier, establishing full-link association between a test condition and a result, and outputting a criterion conclusion. According to the method, the accuracy, efficiency and traceability of passive array beam testing are improved.

Aspects of the present disclosure are directed to triggering object transitions between near-field and far-field regions in an artificial reality environment. For example, the near-field region can be within arm's length for the user, while the far-field region can be greater than arm's length from the user. Objects can be displayed within these regions such that the user can interact with the objects in the artificial reality environment. A field manager can cause a displayed object to transition from the near-field region to the far-field region and / or from the far-field region to the near-field region. The field manager can compare sensed user movements to one or more trigger criteria (e.g., distance threshold(s), velocity threshold(s), etc.) to trigger transition of an object between field regions.

The invention relates to the technical field of data processing, and discloses a cable joint arc multi-physical field simulation method and system. The method comprises the following steps: establishing a defect-containing geometric model, judging according to a temperature gradient change rate and a current rising rate to obtain an adaptive time step length, and carrying out bidirectional coupling iteration on an arc temperature field and a material dielectric parameter based on the time step length to obtain an accumulated damage factor; arranging a virtual temperature detection point to convert the simulation temperature field into a virtual monitoring signal, extracting temperature characteristic parameters and ultrasonic spectrum characteristic parameters, and projecting the temperature characteristic parameters and the ultrasonic spectrum characteristic parameters to a two-dimensional characteristic space to obtain a fault mode fingerprint spectrum. According to the method, adaptive adjustment of the time step length is realized by establishing an arc evolution phase discrimination mechanism, and the problems of low calculation efficiency and insufficient transient characteristic capture caused by a fixed time step length are solved.

The invention discloses a multi-target track identification method and system based on a Grubrum angle field and a CNN, and relates to a target identification and machine learning technology, and the method comprises the steps: obtaining the flight data of a target, and carrying out the preprocessing of the flight data; on the basis of the preprocessed flight data, calculating a Grubrum matrix, and converting the time sequence of the flight path into a two-dimensional image by utilizing angular field transformation, so as to represent an angle difference between a flight path point and a reference point at a corresponding moment in the two-dimensional image by utilizing any pixel value; and inputting the two-dimensional image into a pre-trained CNN (Convolutional Neural Network) model so as to output a target track identification result by using the pre-trained CNN model. According to the embodiment of the invention, the time sequence data is converted into the image form through the Grubrum angle field conversion, so that the CNN can conveniently carry out feature learning and pattern recognition, and the automatic recognition of the track is realized.

The invention provides an intelligent power distribution network high-resistance identification and fault accurate positioning method, medium and system, and the method comprises the steps: uploading the zero-sequence voltage and zero-sequence current signals of each node of an intelligent power distribution network to a main station, carrying out the mean interpolation, logarithmic transformation and normalization processing of abnormal data, and carrying out the recognition of the zero-sequence voltage and zero-sequence current signals of each node of the intelligent power distribution network; then, the sequential sequence signals are converted into images through the Gramer angle field transformation thought, spatial domain full information features are fused, a fault feature quantity space-time atlas sample library is established, convolution and activation function operation and pooling operation are conducted on input samples through a two-channel convolutional neural network model, back propagation is conducted on full connection layer output, and the fault feature quantity space-time atlas sample library is established; the method is used for calculating the gradient of a loss function relative to network parameters and updating the weight and bias of the network through the gradients so as to minimize the loss function, and through establishment of a two-channel convolutional neural network model, the method can perform accurate fault identification and accurate fault positioning on the high-resistance fault of the intelligent power distribution network, and has a good fault identification effect on the high-resistance fault of the intelligent power distribution network. And the influence of transition resistance is avoided.

The invention discloses a random field modeling method for quantifying an initial defect of a spherical shell structure. The method comprises the following steps: determining simulated spherical shell data corresponding to a target spherical shell; the simulated spherical shell data comprises spherical shell geometric information and spherical shell material information; determining a target correlation function and representative discrete points corresponding to a target spherical shell based on spherical shell geometric information and the spherical shell material information; determining a correlation coefficient matrix based on the target correlation function and the representative discrete points; performing eigenvalue decomposition on the correlation coefficient matrix to obtain a standard random field; and performing random field transformation on the standard random field based on a random field transformation formula to obtain a target random field corresponding to the target spherical shell. According to the method, the target correlation function and the representative discrete points are determined according to the simulated spherical shell data, the geodesic distance is introduced into the target correlation function, the spherical shell random field is established based on the eigenvalue decomposition method, the intrinsic correlation of the initial defect of the spherical shell can be accurately quantified, and then the accuracy of spherical shell performance analysis can be improved.

The invention provides a power distribution network single-phase earth fault line selection method and system, and particularly relates to the technical field of power system fault detection, and the method carries out the signal decomposition and noise screening of a power distribution network zero-sequence current signal, and obtains a denoised signal. And then converting the de-noised signal into a two-dimensional matrix image by using Grubrum angle difference field transformation. The image is input into a teacher network containing a wavelet convolution module to extract a fault feature vector. And knowledge distillation is carried out on the feature vectors through a student network containing a depth separable convolution head module, a fault line discrimination result is output, and fault line selection operation is carried out according to the result. According to the method, the technical problem of how to construct a single-phase earth fault line selection method considering high-precision identification and low calculation complexity in a power distribution network environment with high noise interference and weak characteristic difference is solved, the fault detection precision is remarkably improved, and the calculation complexity is reduced; and the performance bottleneck of a traditional method in a complex noise and high-resistance fault scene is effectively solved.

The invention provides a method and a system for improving the precision of an on-orbit SAR antenna pattern of a light and small satellite. The method comprises the following steps: S1, establishing an SAR antenna mechanical flatness field model; s2, establishing a thermal deformation field model of the SAR antenna; s3, establishing an SAR antenna T / R active channel amplitude-phase data field model; s4, SAR antenna machine, thermal and electrical errors are converted into a unified phase coordinate system; s5, establishing a satellite attitude pointing error model, and performing unified physical field conversion; and step S6: respectively carrying out sensitivity analysis on the direction of the antenna pattern aiming at the satellite attitude and the SAR antenna mechanical, thermal and electrical errors. According to the method, ground test and on-orbit data are adopted to compensate for SAR antenna mechanical, electrical and thermal phase errors and satellite platform attitude errors, so that the precision of an SAR antenna pattern model is further improved, and the on-orbit image quality is practically guaranteed.

The invention discloses a non-invasive road multi-component cover plate fault intelligent detection method and device. The method comprises the following steps: collecting vibration data of a cover plate in three directions; performing noise reduction preprocessing on the vibration data in the three directions; respectively carrying out Grubm angle field transformation on the denoised data to obtain grey-scale maps, and fusing the grey-scale maps in three directions according to an RGB form to obtain a GADF color map; building a lightweight twin CNN-BiMGRU-MHAM model, and inputting a color image into a network for training; and performing intelligent fault detection on the road multi-component cover plate by using the trained model, and displaying a fault detection result on a terminal in a data and / or image form. According to the method, information fusion and deep learning technologies are combined, the efficiency and accuracy of fault detection of the coverage board are effectively improved, the detection and maintenance cost is reduced, driving safety is not affected, and a management department can master the health condition of the coverage board in time.

The invention discloses an electromagnetic simulation method and system based on a variational finite element method. A calculation area is set; defining two auxiliary vector fields on an artificial truncation boundary; performing projection on the two auxiliary vector fields in each far-field direction by using a finite element basis function, and calculating an expansion coefficient in each far-field direction; constructing a finite element equation stiffness matrix and a boundary matrix; and according to the near-field electric field, the stiffness matrix, the boundary matrix and the expansion coefficients of the two auxiliary vector fields in the current far-field direction, calculating an electromagnetic far-field component in each far-field direction. By introducing a variational finite element method, precision loss caused by geometric boundary complexity or medium heterogeneity in a traditional integration method is eliminated. A phase factor is directly embedded into the auxiliary vector field, electromagnetic wave propagation characteristics are accurately represented, calculation reliability is ensured, and electromagnetic simulation requirements of a large-scale simulation domain can be met by improving the conversion process from a near field to a far field.

The invention discloses a power distribution network fault type diagnosis and section positioning method and equipment, and the method comprises the steps: selecting eight-channel waveforms of a single measurement point of a fault line: three-phase zero-sequence voltage, zero-sequence voltage, three-phase current and zero-sequence current as research and judgment waveforms; based on a fault waveform data set and a batch simulation waveform data set collected by an actual line, a waveform data set with high quality, wide coverage scene and balanced type distribution is generated through a conditional generative adversarial network; each channel of waveform of the fault waveform is subjected to classification complete set normalization, feature region interception and field transformation dimension expansion to form a two-dimensional field matrix, eight groups of two-dimensional matrixes are preprocessed to form eight-channel two-dimensional matrix tensors, and a model training set is formed in combination with fault types and fault section labels; according to the invention, classified identification of the fault type and the fault section of the fault waveform is realized; according to the method, troubleshooting can be accelerated, the power supply reliability of the power distribution network is improved, and technical support is provided for grounding fault diagnosis completely depending on data driving.

The invention discloses a phase-free antenna measurement method, a phase-free antenna measurement system and a storage medium. The phase-free antenna measurement method comprises the following steps: constructing a measurement system comprising an antenna to be measured, a measurement probe, a fixed auxiliary reference antenna and software defined radio; the probe is controlled to move and sample on the scanning plane, and the SDR synchronously collects receiving signals of the probe and the reference antenna; based on the frequency domain transformation result of the two paths of signals, the radiation field amplitude of each sampling point and the relative phase relative to the reference antenna are directly calculated through non-iterative operation, so that complete complex electric field distribution on the scanning plane is reconstructed; and finally, obtaining a far-field directional diagram through near-field-far-field transformation. The fixed reference antenna is introduced to serve as a phase reference, synchronous acquisition and direct calculation are achieved in combination with the SDR, the antenna radiation field can be rapidly and accurately recovered without direct phase measurement or complex iteration in the single scanning plane and single scanning process, and the method has the advantages of being simple in system, high in measurement efficiency and high in signal drift resistance.

The invention discloses a smart watch arrhythmia early warning method and system based on deep learning. The method comprises the steps that firstly, PPG signals are collected through a photoelectric volume pulse wave sensor, and denoising and standardization processing are conducted through an adaptive filtering algorithm; then, the heart beat position is recognized through a wavelet transform peak value detection algorithm, and a heart rate variability feature sequence is obtained; converting the one-dimensional heart rate variability feature sequence into a two-dimensional heart rhythm image by using an improved Gramian angular field transformation algorithm; and training an arrhythmia classification model by adopting a ResNet-50 convolutional neural network to realize automatic identification of different types of arrhythmia. And finally, continuous monitoring and timely alarming are realized through a sliding window technology and an early warning mechanism. The key technical problems of low detection precision, poor real-time performance, high equipment cost, poor user experience and the like in the existing arrhythmia detection technology are effectively solved, and an innovative solution is provided for heart rhythm health monitoring of the intelligent wearable equipment.

Disclosed in the present invention are a conversion core member, a double-lens mode field conversion device and a manufacturing method therefor. The conversion core member comprises a solid-core optical fiber part, a hollow-core optical fiber part and a calibration sleeve, wherein the solid-core optical fiber part comprises a solid-core-end lens and a solid-core single-mode optical fiber having a first mode field diameter, and the solid-core end lens is located at the tail end of the solid-core optical fiber part; the hollow-core optical fiber part comprises a hollow-core-end lens and a hollow-core optical fiber having a second mode field diameter; the calibration sleeve has a hollow cavity I in the axial direction thereof; the solid-core optical fiber part and the hollow-core optical fiber part respectively extend into the cavity I of the calibration sleeve via two open ends of the cavity I; and the solid-core-end lens and the hollow-core-end lens are arranged opposite each other. In the device, the conversion of different mode field diameters between the single-mode optical fiber and the hollow-core optical fiber is implemented by means of two lenses, such that a reliable and stable connection between the single-mode optical fiber and the hollow-core optical fiber is achieved.

A measuring system and a measuring method of an antenna pattern based on near field to far field transformation (NFTF) are provided. The measuring system includes a probe antenna, a reference antenna, and a control system. The probe antenna measures an electric field radiated by an antenna under test to obtain electric field information. The reference antenna measures the electric field to obtain a reference phase. The control system is coupled to the antenna under test, the probe antenna, and the reference antenna, wherein the control system applies near field focusing to the reference antenna to configure a focus point of the reference antenna on the antenna under test, and the control system performs the NFTF according to the electric field information and the reference phase to output far field patterns.

This invention provides a method for rapid planar near-field measurement of large-aperture antennas, relating to the field of antenna technology. The method obtains sparse field amplitude and phase information of several sampling surfaces in the near-field radiation region of the antenna under test through large-step uniform sampling. Simultaneously, it samples a finite number of precise field amplitude and phase information points for each sampling surface and obtains the field distribution characteristics by calculating the fluctuation of the amplitude at the sampling points. Based on the sparse field amplitude and phase information and field distribution characteristics of the sampling surfaces, it uses partitioned interpolation and field distribution calibration methods to obtain the initial electric field of each sampling surface. Based on planar wave spectrum theory, iterative numerical calculations are performed on the initial electric field to recover the near-field distribution of each sampling surface. Finally, the far-field radiation characteristics of the antenna are obtained through near-field-to-far-field transformation. This invention can significantly shorten the testing time while obtaining key indicators of the antenna's far-field characteristics, including the normalized radiation pattern main lobe and first sidelobe, 3dB beamwidth, sidelobe level, and axial ratio.

The invention provides a wind wheel device for reducing wind load and wind-induced vibration of a photovoltaic module, and relates to the technical field of structural vibration control, the wind wheel device comprises wind wheel bases arranged on two sides of a photovoltaic panel, the left end and the right end of each wind wheel base are respectively provided with two circular hole grooves, and wind wheel bearings are arranged in the circular hole grooves; the two ends of a wind wheel rotating shaft penetrate through the wind wheel bearings on the two sides respectively, and a wind wheel is arranged on the wind wheel rotating shaft. According to the wind wheel device, flow field transformation caused by passive rotation of the light wind wheel along with wind acts on the surface of the photovoltaic panel, the pressure difference between the upper surface and the lower surface of the photovoltaic panel can be effectively improved, and the vibration phenomenon generated when the photovoltaic panel is subjected to wind loads is reduced; the wind wheel devices are installed on the front edge and the rear edge of the windward side of the photovoltaic panel and passively rotate along with wind through light wind wheels. Flow separation at the windward front edge can be effectively reduced by mounting the front edge, shedding of trailing edge trailing vortex can be effectively reduced by mounting the front edge on the trailing edge, and the influence of flow-induced vibration is further improved.

This invention relates to the field of distribution network fault diagnosis technology, specifically to a method and device for fault line selection in distribution networks based on a multimodal enhancement and hybrid deep learning model. The method includes the following steps: acquiring zero-sequence current and voltage transient waveform data of each line in the distribution network at the time of a fault; performing Gram angle field transformation on the one-dimensional waveform data using a penalized inner product to convert it into two-dimensional GASF and GADF images, respectively, and stitching them together to generate a four-channel E-GAF image; inputting the E-GAF image into a Trans-DenseNet model for fault discrimination, and fusing local-global features by embedding a Transformer encoder layer in the DenseNet to output the final fault line selection result. Compared with existing technologies, this invention can adapt to complex faults such as high resistance and arcing, achieving high-precision fault location and providing reliable support for intelligent fault self-healing in new distribution networks.

The application discloses a kind of deep neural network gas identification method and system based on internet of things electronic nose, gas sample response signal obtained by gas sensor array collection is sent to cloud data center by terminal equipment with sensor array and internet of things module.In the pre-processing of data to microcontroller MCU, gram angle field transformation is carried out, so that the two-dimensional sensor response data received by host computer becomes three-dimensional data that can be input into convolutional neural network after dimensionality increasing, and the gas label is output through classifier, which can represent gas category and concentration grade, realizes the real-time, anti-interference and high detection accuracy of remote monitoring.Gives play to the advantages of strong feature extraction ability, fast model convergence and high recognition accuracy of convolutional neural network.It can be widely applied in environmental detection, industrial production, medical treatment and safety fields.

The present invention discloses a video flow measurement method, device, storage medium and equipment based on an improved recursive full-field transformation optical flow model, belonging to the field of hydrological flow measurement technology. The method comprises: obtaining a river video and performing frame cutting processing to obtain video frames; calculating a transformation matrix based on the river video and pre-acquired survey results, transforming the on-map coordinates of each speed measurement point in each video frame in the river video into world coordinates through the transformation matrix, and obtaining the cross-sectional area between each speed measurement point; inputting the video frame into a trained improved recursive full-field transformation optical flow model, and outputting the pixel displacement between two adjacent frames of each speed measurement point; calculating the river surface flow velocity according to the pixel displacement, the world coordinates and the cross-sectional area; the present invention introduces a CBAM attention mechanism and deformable convolution, which can better detect complex river changes and subtle water surface movements, thereby improving the accuracy of the algorithm.

The present invention discloses a flat-field conversion system for a lobster-eye lens, which belongs to the field of detection. The present invention comprises a light source, a lobster-eye lens, a curved optical fiber bundle, a relay lens group, a medium-wave infrared detector, and a base. The lobster-eye lens is used to receive light projected by a detection object within the field of view. The present invention places the light source at the edge of the field of view of the lobster-eye lens, and the lobster-eye lens, the curved optical fiber bundle, the relay lens group, and the medium-wave infrared detector are coaxially fixed on the base in sequence from left to right. The lobster-eye lens detects objects in the edge of the field of view, and the medium-wave infrared light emitted by the object is converged on the front end surface of the curved optical fiber bundle through the lobster-eye lens and forms an image. The curved optical fiber bundle converts the curved image plane into a flat field and transmits it to the object plane of the relay lens group. The relay lens group transmits the image plane and performs cold stop matching with the medium-wave infrared detector, thereby resolving the contradiction between the large field of view and high resolution of the lobster-eye lens, and realizing high-precision detection of targets in the edge of the field of view by the lobster-eye system.

The embodiment of the invention discloses an acceleration test method for power grid equipment after on-site reconstruction, and the method comprises the steps: carrying out the in-situ dynamic test of the equipment through employing an acceleration sensor before and after carrying out the seismic mitigation and isolation measure installation of the equipment, monitoring the fundamental frequency change of the equipment, and carrying out the in-situ dynamic test of the equipment; the characteristics of the equipment are changed after the seismic mitigation and isolation measures are added; arranging an IBIS-FS system in the side direction of the equipment, and carrying out in-plane vibration measurement to obtain in-plane vibration measurement data; the IBIS-FS system is arranged in the transverse direction of the equipment, out-of-plane vibration measurement is carried out, out-of-plane vibration measurement data are obtained, comprehensive acceleration testing and displacement after transformation can be carried out on the transformation condition of seismic reinforcement of the electrical equipment more rapidly and accurately, and therefore feedback of the transformation effect can be obtained in real time, and the improvement effect of the electrical equipment is improved. Safety risks and extra cost possibly caused by equipment movement are avoided, and the test period is shortened.

The invention belongs to the technical field of spectrum technology and deep learning, and provides an aluminum alloy mark recognition method based on a lightweight spectrum attention network, which is used for improving the accuracy and robustness of aluminum alloy mark recognition. The method comprises the following steps: firstly, converting a one-dimensional XRF spectrum into a two-dimensional spectrogram through Grubby field transformation so as to enrich spatial information and amplify spectrum detail differences; then, a lightweight spectrum attention module is provided, and the module reduces the calculation amount through simple adaptive calculation, pays attention to important information of a spectrogram and inhibits redundant information; and finally, on the basis of a lightweight spatial attention module, a ShuffleNet structure is combined to design a lightweight spatial self-attention network, accurate recognition of the aluminum alloy mark is realized, and the accuracy and robustness of aluminum alloy mark recognition are effectively improved. Moreover, the aluminum alloy mark identification method is used as an end-to-end deep learning model, and the mark of the alloy can be accurately identified without complex preprocessing of a sample.

This invention discloses a method and device for fault type diagnosis and fault segment location in distribution networks. The method includes selecting eight-channel waveforms from a single measurement point on a faulty line: three-phase zero-sequence voltage, zero-sequence voltage, three-phase current, and zero-sequence current as the judgment waveforms. Based on fault waveform datasets collected from actual lines and batch simulation waveform datasets, a high-quality waveform dataset with broad coverage and balanced type distribution is generated through a conditional adversarial generative network. Each channel of the fault waveform undergoes classification and normalization, feature regions are extracted, and then the dimensions are expanded into a two-dimensional field matrix through field transformation. The eight sets of two-dimensional matrices are preprocessed into eight-channel two-dimensional matrix tensors. These tensors, combined with fault type and fault segment labels, form a model training set. This invention enables the classification and identification of fault types and fault segments in fault waveforms. This invention can accelerate fault investigation, improve the power supply reliability of distribution networks, and provide technical support for ground fault diagnosis that relies entirely on data-driven methods.

The invention provides a CSI behavior recognition training method based on meta-learning. The method comprises the following steps: establishing a CSI data acquisition system to obtain a data set; carrying out linear interpolation completion and Hampel filtering noise reduction on the data, and generating a two-dimensional feature map through Grubby angle difference field transformation; based on a task division strategy, dividing the CSI data into a support set and a query set, and constructing a cross-domain small sample learning task; in the model optimization stage, different meta-learning strategies are adopted for cross-domain differences, one meta-learning strategy is Mixup-MAML, Mixup data is introduced in an outer cycle of the MAML to enhance and generate a virtual query set so as to improve rapid adaptation and cross-domain generalization ability, the other meta-learning strategy is MixMeta-SGD, model parameters, learning rates and updating directions are jointly updated under a Meta-SGD framework, and a virtual sample is generated in combination with Mixup so as to enhance robustness; and finally, small sample fine tuning is carried out on the optimized model, final model parameters adapted to the target task are obtained, and the result can be used as input of a subsequent CSI behavior recognition module or a deep network, so that generalization and stability of the system are improved.

The invention discloses a display anti-shake method and system for intelligent riding glasses, and the method comprises the steps: carrying out the time alignment of an angular velocity and an accelerated velocity collected by an inertial measurement unit, and estimating and compensating an offset and scale factor; performing dual-channel adaptive filtering on the inertial data stream; detecting and identifying a glancing or turning intention of the wearer based on an angular velocity threshold value, acceleration abrupt change and short-time pause, and generating a gating coefficient for continuously adjusting the image stabilization intensity; according to the low-frequency attitude component and the gating coefficient, view field transformation and viewport offset used for stabilizing the superposed content are solved; and according to the angular motion state, performing prediction compensation to the display moment on the view field transformation to offset the rendering and display link time delay, taking the relative displacement of the key content of the adjacent frames as stability feedback, and updating offset and parameters on line to realize long-term self-correction. According to the technical scheme of the invention, the readability and security of the superposed information can be integrally improved.

The invention provides a method for identifying brain function diseases by using a neural network based on near-infrared data and a processing device. The method comprises the following steps: on the basis of first near-infrared data of a plurality of acquisition channels, utilizing Grubrum angle field transformation to obtain Grubrum angle field image data; and performing convolution and pooling operation by using the neural network based on the Grubrum angle field image data so as to convert the Grubrum angle field image data into a first feature vector below two dimensions. And obtaining a fusion feature vector based on the first feature vector and a second feature vector with the same dimension, wherein the second feature vector is obtained based on features of other dimensions of the target brain. And based on the fusion feature vector, determining an identification result of the brain function disease of the subject. Therefore, the second feature vectors of different dimensions can be conveniently fused as supplementary information, the complex difference between the brain function disease patients, especially the patients with mild brain function diseases or early brain function diseases and healthy people is comprehensively captured, and the classification precision is improved.

The invention relates to the technical field of digital signal processing and communication data links, in particular to a data link waveform data preprocessing method based on digital operation, which comprises the following steps: acquiring high-speed continuous discrete time sequence data acquired by a data source as original waveform data; a sign bit is extracted, an absolute value is taken, numerical field transformation is carried out by utilizing leading zero detection logic, and linear field data is mapped into approximate logarithm field representation data comprising an order code and a mantissa; dividing the mantissa into a reserved high order and a discarded low order; inputting the approximate logarithm domain representation data into a main addition path of a logarithm domain addition filtering topology to carry out additive operation, and carrying out error compensation on the discarded low bits by utilizing an error compensation lookup table to generate logarithm domain preprocessing data; in combination with sign bits, converting back to a linear domain through anti-logarithm mapping logic, and outputting preprocessed waveform data; according to the invention, dynamic optimal balance between cross-environment processing quality and system resource allocation is realized.