178 results about "Maximum eigenvalue" patented technology

The invention provides an automatic early warning method for sudden weather in a target area, which belongs to the technical field of weather early warning, and comprises the following steps of: establishing a primary dense matrix by adopting adaptive filtering processing and a frequency domain signal separation technology, and generating a secondary dense matrix by applying a marine meteorological recognition model of a spiral progressive network structure; a dynamic statistical equation is used to calculate the physical coupling relationship of each parameter to establish a multi-scale weather process balance matrix, a maximum flow and minimum cut algorithm is used to optimize a weather system coupling relationship network to calculate a coupling degree matrix, and a dynamic threshold adjustment mechanism is established according to coupling strength parameters to adjust the early warning detection frequency. And based on a comparison result of the coupling degree moment order maximum characteristic value and a preset risk threshold value, establishing a grading early warning system and outputting a corresponding early warning signal to control an offshore oil and gas platform emergency response system. The technical problem that a multi-time scale weather process coupling relationship cannot be effectively processed is solved.

The invention provides a GCN-based honey situation map analysis method, and relates to the technical field of neural networks. Node information and side information are obtained, a honey array subgraph is constructed, and a # imgabs0 #-order Laplacian matrix is obtained; updating and recurring a characteristic value of the # imgabs1 #-order Laplacian matrix based on the maximum characteristic value of the honey array subgraph to obtain a normalized Laplacian matrix; performing random feature masking on each node to obtain a plurality of updated node feature matrixes; and obtaining node outputs of the plurality of honey array sub-graphs on the plurality of levels based on the updated node feature matrix, obtaining a prediction value, carrying out iteration on parameters of the graph neural network based on the prediction value, obtaining an updated graph neural network, and generating a prediction result. According to the invention, the convolution calculation is carried out on the graph signal, and the Laplacian matrix and the Chebyshev polynomial of the graph are combined, so that an efficient graph convolution kernel is constructed, and the accuracy and the performance of network situation graph analysis are improved.

The invention relates to a direct positioning method in the presence of a site error, belongs to the technical field of radiation source positioning, and solves the problems of high positioning delay and poor precision of a radiation source position in the prior art. The direct positioning method comprises the following steps: based on an observation station measurement position model, performing Taylor series first-order expansion on an observation station measurement signal model at the measurement position of each observation station to obtain an expanded observation station measurement signal model; constructing a cost function of a radiation source position according to the real-time error model, the measurement signal of each observation station and the real-time position signal model; and according to each possible position of the radiation source, determining the maximum characteristic value corresponding to each possible position in the cost function of the radiation source position by adopting a grid search method, and taking the possible position corresponding to the maximum characteristic value as the positioning position of the radiation source. And rapid and accurate positioning of the position of the radiation source is realized.

The invention relates to the field of reactor data processing, and particularly discloses a numerical calculation method for solving a neutron transport equation. The method comprises the following steps: 1, describing the geometry of a reactor core of a nuclear reactor by utilizing a construction entity geometric modeling method, and subdividing a curved surface grid along a curved surface structure; 2, directly expanding a neutron flux distribution function in each curved surface grid by using a multivariate polynomial; 3, deriving a neutron transport equation and an integral residual equation of a boundary condition under each curved surface grid; 4, solving geometric integral parameters of each curved surface grid; step 5, establishing an adaptive linear system; and step 6, iteratively calculating and solving a maximum characteristic value and an unknown expansion coefficient.

The invention belongs to the technical field of novel power systems, and particularly relates to a network construction type control transformation method, system and equipment and a medium. Aiming at the defects existing in the transformation capacity calculation process of the existing network construction type control transformation method, the invention adopts the following technical scheme: the network construction type control transformation method for realizing subsynchronous oscillation suppression comprises the following steps of: obtaining a network following type critical short-circuit ratio and a network construction type critical short-circuit ratio; calculating an initial generalized short-circuit ratio; calculating influence factors of the initial generalized short-circuit ratio with respect to the rated capacity of each converter and sorting the influence factors; selecting a station where the maximum influence factor is located to carry out equal-capacity network construction type transformation, calculating a generalized short-circuit ratio after system transformation, and determining whether to increase a transformation station or not; and after the transformation position is determined, solving to obtain the minimum transformation capacity by taking the condition that the maximum characteristic value of the expanded admittance matrix is smaller than the net-forming type critical short-circuit ratio as a constraint. The method has the advantages that the minimum transformation capacity is obtained based on the influence factors and the short-circuit ratios, and the economic cost is reduced to the maximum extent.

The invention discloses a spaceborne synthetic aperture radar scattering wave mutual interference suppression method and system based on matched filtering, a storage medium and electronic equipment, and the method comprises the following steps: estimating related parameters of mutual interference according to the prior information of an orbit and a time-frequency relation, and carrying out the range pulse compression of an interference signal in echo data; detecting whether the range pulse contains interference or not by adopting a maximum characteristic value sequence detection method; calculating a relative signal to interference ratio containing interference pulses; reconstructing and extracting interference signals under the conditions of high signal to interference plus noise ratio and low signal to interference plus noise ratio respectively by using a feature subspace projection method and a robust principal component analysis method; and filtering the echo data, and carrying out mutual interference inverse distance compression processing on the data so as to recover normal echoes without interference. According to the invention, the compression gain of mutual interference can be effectively utilized, mutual interference signals in synthetic aperture radar echoes can be extracted more effectively, and high-precision interference suppression is completed while useful signals are better protected.

The invention provides a time-varying frequency spectrum map compression surveying and mapping method based on dictionary learning and smooth regularization, and the method comprises the steps: building a sparse base psi through the path loss between two points in a spatial frequency domain, carrying out the sparse representation of a signal in the spatial frequency domain, and building a compression sensing model of a three-dimensional space through combining with a measurement matrix; a measurement matrix in the compressed sensing model is optimized, singular value decomposition (SVD) is used for decomposing psi, sparse bases psi m corresponding to the first m maximum eigenvalues are selected, and an optimized measurement matrix is obtained after m times of iteration; the optimized measurement matrix is substituted into a compressed sensing model, an O-KSVD dictionary is adopted to learn and update a sparse base psi, and spectrum data are recovered; finally, sparse regularization and weighted smoothing regularization are combined to achieve the minimum Euclidean distance between adjacent points in the reconstructed spectrum map, smoothing regularization is used for driving the adjacent points in the reconstructed spectrum map to be similar, sparse representation in the spectrum map is updated through iteration till convergence, and actual spectrum mapping is obtained.

The invention discloses a time-varying condition guided wave structure damage monitoring method, which comprises the following steps: arranging a piezoelectric patch damage monitoring path on a structural member, and respectively collecting response signals of different damage degrees at different temperatures; classifying according to different damage degrees, and extracting a plurality of characteristic parameters of the response signals one by one; the intra-class distance and the inter-class distance between the feature vectors are calculated through variance calculation and mahalanobis distance, when the intra-class distance is smaller than the inter-class distance, the separability of the vectors is good, and therefore appropriate feature parameters are screened; selecting a Gaussian kernel function to carry out kernel processing on the feature matrix, constructing a feature filtering model by using the first three feature vectors with the maximum feature values of the health kernel matrix, and then projecting the health kernel matrix and the damage kernel matrix onto the feature filtering model at the same time to obtain features with temperature factors filtered out; and finally, inputting the filtered sample into a CNN classification model for training, wherein the trained model can complete the recognition of the damage degree of the structural member. The feature filtering model and the damage degree recognition model are constructed by means of a machine learning algorithm, and structural damage monitoring under the time-varying condition is achieved.

The invention belongs to the technical field of geographic information systems and PS-InSAR, and discloses a stress-strain-considered InSAR building deformation risk assessment method, which comprises the following steps: carrying out PS-InSAR processing on time sequence SAR image data to obtain a time sequence deformation result and carrying out precision verification; adopting a PS point correction method based on a self-adaptive building contour to screen out PS points of the building; selecting five indexes of building strain, building cumulative inclination amount, building cumulative settlement amount, building age and building structure material type; building risk assessment modeling is carried out by applying an analytic hierarchy process, consistency check and weight calculation are carried out, and the maximum characteristic value of a judgment matrix established by the five risk assessment indexes is calculated; grading the indexes of the model, performing risk assessment on each index, and drawing; and verifying a building risk assessment result in combination with actual dangerous house data. According to the invention, accurate evaluation of the deformation risk level of the building is realized.

The invention discloses a comprehensive evaluation method and system for the frequency voltage supporting capability of a new energy station, and the method comprises the steps: determining an evaluation criterion according to the frequency voltage supporting characteristics of the new energy station; according to the evaluation criterion, establishing an evaluation index of the frequency voltage supporting capability of the new energy station; according to the actual measurement value and the simulation value of the new energy station, performing dimensionless processing on the evaluation index quantity to obtain an evaluation index quantized value; comparing the evaluation indexes in pairs by adopting an analytic hierarchy process, and constructing a difference matrix; obtaining a weight coefficient of the evaluation index by calculating a maximum characteristic value of the difference matrix and a corresponding characteristic vector; and substituting the evaluation index quantized value and the weight coefficient into a pre-constructed new energy station frequency voltage support capability evaluation model, and determining a new energy station frequency voltage support capability comprehensive score. The support capability of the new energy station can be accurately and rapidly evaluated, and the contribution of the new energy station to power grid frequency and voltage stability is quantitatively analyzed.

The invention provides a method, medium and system for estimating the orientation of a horizontal array in an uncertain shallow sea environment, and belongs to the technical field of geophysics, and the method comprises the steps: firstly, receiving a far-field narrow-band signal through a horizontal array, obtaining a time domain sampling snapshot, then calculating an array sampling covariance matrix, and constructing a signal subspace through the first J feature vectors; meanwhile, according to a standard shallow sea sound velocity profile and a parameter uncertainty range, a horizontal wave number range of a propagation mode is obtained, sampling is carried out in the range to construct a mode matrix, and feature vectors corresponding to the first K maximum feature values are extracted to construct a mode subspace. Next, constructing an expansion matrix by using base vectors in the signal subspace and the modal subspace, performing QR decomposition to obtain an upper triangular matrix, and extracting diagonal elements of the upper triangular matrix to construct a sound source orientation estimation function; and finally, the angle corresponding to the peak value of the sound source orientation estimation function is the sound source orientation estimation value.

The invention discloses a reactor transport critical simulation method and system based on a Krylov-Schur algorithm, and relates to the field of nuclear reactor core design and safety, in particular to the reactor transport critical simulation method and system based on the Krylov-Schur algorithm. The objective of the invention is to realize that the Krylov-Schur method is used for carrying out transport critical calculation to solve the maximum characteristic value, and the convergence precision of the maximum characteristic value and the characteristic vector is improved. The method is technically characterized by comprising the following steps of: discretizing a steady-state neutron transport equation and deforming in an operator form, performing multi-group approximation on an energy variable E in the steady-state neutron transport equation, performing discrete longitudinal standard discretization on an angle variable to obtain a coupled equation, expressing the coupled equation in the operator form, and giving a standard characteristic value problem; and solving the eigenvalue and eigenvector of the matrix P in the standard eigenvalue problem. The method is used for reactor transportation critical simulation.

The invention particularly relates to a method for realizing scenarized wireless network quality evaluation based on an analytic hierarchy process. According to the method for realizing the scene wireless network quality evaluation based on the analytic hierarchy process, a scene health degree evaluation index system is constructed, and a judgment matrix is constructed for each layer of indexes by performing assignment on the relative importance of each layer of indexes; calculating a maximum characteristic value lambda max of the judgment matrix and a corresponding characteristic vector thereof to obtain a weight vector W; after the judgment matrix passes the consistency verification, calculating the comprehensive weight of the final index; scoring each index according to the formulated rule, obtaining the final score of each index in combination with the index weight, adding to obtain the health degree coefficient of the whole scene, and further obtaining the health degree level of the whole scene. According to the method for realizing the scenarized wireless network quality evaluation based on the analytic hierarchy process, comprehensive monitoring on the wireless network and comprehensive evaluation on the scenarized wireless network are realized.

The invention relates to the technical field of weight assignment, in particular to a fixed-point industry development effect evaluation system-based weight assignment calculation method, which comprises hierarchical model establishment, index importance evaluation, hierarchical analysis, entropy analysis and index weight assignment. According to the weight assignment calculation method based on the fixed-point industrial development effect evaluation system, an original judgment matrix is constructed according to relative importance data of indexes of all layers through an analytic hierarchy process, and a maximum feature root and a maximum feature vector are calculated to serve as bases to perform consistency check on the judgment matrix; then, the relative importance data of the indexes are subjected to forward processing through an entropy evaluation method, so that the index proportion and the entropy and information entropy redundancy of each index are calculated, the index weights obtained through the analytic hierarchy process and the entropy evaluation method are subjected to weight combination, weight assignment of each index is obtained through weighted arithmetic average processing, and the weight assignment of each index is obtained; the subjective influence and the influence on weight fluctuation during data change can be effectively reduced.

The invention discloses an image classification method based on multi-manifold joint metric learning, and the method comprises the steps: 1) data preprocessing: constructing Grassmann features and SPD features; 2) constructing a double-flow kernel matrix, including constructing a Grassmann manifold kernel matrix and an SPD manifold kernel matrix, and respectively calculating a Grassmann test kernel matrix and an SPD test kernel matrix; 3) implementing multi-manifold joint metric learning, and extracting kernel feature vectors from the kernel matrix to calculate an intra-class scatter matrix and an inter-class scatter matrix; solving the projection matrix, sorting the feature vectors according to the sizes of the feature values, and selecting the feature vectors corresponding to the first plurality of maximum feature values to form the projection matrix; the method comprises the steps of (1) extracting different ground features, (2) carrying out feature fusion, completing extraction and implementing sample mapping, and (3) constructing a sample-level fusion convolutional neural network and completing classification of different ground features of an original image.The method belongs to the technical field of image processing and machine learning, and the classification accuracy can be remarkably improved.

The invention discloses a non-cooperative information source DOA (direction of arrival) estimation method based on characteristic vectors, and aims to solve the problem of non-cooperative information source DOA estimation under the condition that the number of information sources is unknown, the condition of coherent information sources is unknown and array aperture sacrifice is avoided. According to the method, eigenvalue decomposition is carried out on a covariance matrix of an array receiving signal, so that an eigenvector corresponding to the maximum eigenvalue of the covariance matrix is obtained, then a linear model based on the eigenvector and a guiding matrix is established, and DOA estimation of an information source is achieved by solving the linear model. Theoretical analysis proves that under the ideal white noise background, the characteristic vector corresponding to the maximum characteristic value of the array receiving signal covariance matrix is a linear combination of all information source steering vectors. A numerical simulation result shows that the scheme can realize information source DOA estimation under the condition of lacking information source prior information, and a simpler, more convenient and more reliable solution can be provided for array signal processing.

Provided is a target sound extraction technique based on a steering vector generation method enabling instability in a calculation to be prevented when a neural network is trained by using an error back propagation method to reduce an estimation error of a beamformer. A target sound signal generation apparatus generates a target sound signal yt,f corresponding to a target sound included in an observed sound from an observed signal vector xt,f corresponding to the observed sound collected by using a plurality of microphones. The target sound signal generation apparatus includes a mask generation unit, a steering vector generation unit, a beamformer vector generation unit, and a target sound signal generation unit. The mask generation unit is configured as a neural network trained by using an error back propagation method. The steering vector generation unit generates a steering vector hf by determining an eigenvector corresponding to a maximum eigenvalue of a predetermined matrix generated from the observed signal vector xt,f and a mask γt,f by using a power method.

The invention provides a production line parameter real-time scheduling method based on reinforcement learning, and belongs to the technical field of production lines, and the method comprises the steps: collecting sensor data to form an original state vector, reducing the dimension of the original state vector into a low-dimensional feature state vector through a state compression encoder, inputting the low-dimensional feature state vector corresponding to a microcosmic scheduling unit into a parameter decision model, and carrying out the real-time scheduling of the microcosmic scheduling unit; the attention weight coefficient of the model is determined by the product of the historical scheduling success rate, the element value of the difference degree matrix and the maximum characteristic value of the inter-stage sensitivity matrix, after a scheduling instruction is output, simulation evaluation is performed in a digital twin platform, and after the scheduling instruction passes, a real production line is issued for execution and multi-time-scale deviation indexes are collected; and calculating a comprehensive reward value according to the indexes, and storing an experience sample to an experience playback buffer area for model online update training, thereby solving the technical problem that production line parameter scheduling is difficult to consider multi-level state feature recognition and dynamic decision weight optimization at the same time.

The invention provides a quantization method and a quantization device, which can perform singular value decomposition on a first matrix formed by word embedding vectors to obtain the maximum characteristic value of each block matrix in M block matrixes, and can determine the quantity of quantization bits of each block matrix according to the maximum characteristic value of each block matrix. The method comprises the following steps of: dividing each block matrix into a plurality of block matrixes, quantifying each block matrix to obtain each quantized block matrix, combining each quantized block matrix to obtain a quantized first matrix, and representing the amount of semantic information carried in each block matrix by a maximum characteristic value of each block matrix to a certain extent, according to the embodiment of the invention, the quantized bit number can be allocated to each block matrix according to the amount of semantic information carried by each block matrix, so that different bit numbers can be allocated to the block matrixes carrying different semantic information, differential bit number allocation can be realized, semantic information loss can be reduced, and user experience can be improved.

The invention discloses a rapid explosion suppression and flame prevention method for a transformer, and belongs to the technical field of power equipment safety, and the method specifically comprises the steps: S1, collecting multi-physical field data in real time through a sensor network disposed in the transformer; s2, based on the multi-physical field data, determining an electrothermal instability index, a fractal dimension of fire nucleus growth, a maximum characteristic value of an irradiance Hessian matrix and a spectral coherence length attenuation rate; s3, combining the electrothermal instability index, the fractal dimension of fire core growth, the maximum characteristic value and the spectral coherence length attenuation rate to construct a characteristic vector; and S4, identifying the initial fault state based on the feature vector, correcting the initial fault state by using a preset physical model to obtain a final fault state, and triggering a grading response strategy according to the final fault state. And a solid data basis is provided for subsequent accurate diagnosis.

The invention provides an operation and maintenance effect evaluation method, system and device and a storage medium, and relates to the technical field of sewage treatment plant operation and maintenance effect evaluation. The method comprises the steps of determining evaluation indexes; key indexes related to operation and maintenance of the sewage treatment plant are selected; a weight is determined by adopting an analytic hierarchy process, and for each pair of indexes, a score is given according to relative importance, a judgment matrix is formed, and a feature vector and a maximum feature value are calculated; verifying the consistency of the judgment matrix, and ensuring the rationality of evaluation; a fuzzy membership function is determined based on fuzzy mathematics, a fuzzy evaluation matrix is calculated, and then comprehensive evaluation and grade determination are carried out. According to the embodiment of the invention, the index weight is determined through the analytic hierarchy process, and the evaluation data is processed in combination with fuzzy mathematics, so that comprehensive evaluation is carried out, the influence of qualitative and quantitative indexes is comprehensively considered, the method adapts to uncertainty in actual operation, and the overall economic and social benefit capability, the sustainable development capability and the replicability are improved.

The invention discloses a penalty term-based multicast and unicast communication sensing integrated secure beam forming method, which comprises the following steps of: sending a multicast and unicast mixed signal to a user by a base station, decoding a multicast signal by a communication user by using NOMA (Non-Orthogonal Multiple Access), eliminating the interference of the multicast signal on a unicast signal, acquiring the reachable rate of multicast and unicast communication at the communication user, and transmitting the unicast and unicast communication to the base station; obtaining a multicast communication reachable rate at the radar user; the base station perceives the radar user and the target and obtains beam gains of the radar user and the target; a target eavesdropping unicast signal serving as an eavesdropper in the sensing process is obtained, and the unicast signal eavesdropping rate at the target position is obtained; establishing a hybrid multicast and unicast secure communication beam optimization problem, converting a relation between a trace of a positive semi-definite matrix in the problem and a maximum characteristic value into a penalty term, multiplying the penalty term by a penalty factor, putting the multiplied penalty term into a target function for constraint, and obtaining a convex optimization problem; solving the convex optimization problem through iteration to obtain multicast and unicast beams; the system has the advantages of good robustness and good performance.

The invention provides a financial risk early warning analysis method and system, a storage medium and a device, relates to the field of enterprise risk management, and solves the technical problem that the existing financial risk early warning technology is not timely and inaccurate. The method comprises the following steps: acquiring financial data, operation data and external environment data of an enterprise to obtain a plurality of index data; classifying the plurality of index data into a plurality of risk domains, calculating the risk field intensity of each risk domain based on the index data in the risk domains, and forming a risk vector; obtaining the risk field intensity of each risk domain in a preset period of the enterprise, inputting the risk field intensity of each risk domain in the preset period into a first network based on a graph neural network, and outputting a cross-risk domain infection probability; constructing an adjacent matrix according to the cross-risk domain infection probability; and calculating the maximum eigenvalue of the adjacent matrix and the risk vector, and triggering graded early warning based on the maximum eigenvalue and the derivative of the maximum eigenvalue. The method and the device are used in the whole-process financial risk management process of enterprises.

The invention provides a power system harmonic resonance analysis method and device, equipment and a storage medium. The method comprises the steps of solving a maximum characteristic value of a node impedance matrix by using a power iteration method, reducing the calculation amount and improving the speed. And obtaining a maximum characteristic value set of multiple frequency points through iterative calculation, and constructing modal impedance as an analysis parameter. And determining the resonant frequency and the interval by adopting the dual-frequency step length. Meanwhile, a half-power bandwidth method is introduced to calculate the damping ratio, and an accurate harmonic resonance index is constructed in combination with modal impedance, so that the analysis efficiency and accuracy are improved.

The invention provides a muscle fiber disorder imaging method, a muscle fiber disorder detection method, a muscle fiber disorder detection system and a storage medium, and belongs to the technical field of biomedical detection software. The method is technically characterized by comprising the following steps: S100, acquiring a boundary voltage difference matrix of muscle to be detected; s200, carrying out normalization processing on the delta V; s300, the contour line of the muscle to be detected is extracted, the area in the contour line is an EIT simulation model area, and then a network node coordinate matrix of the muscle simulation model area to be detected is obtained; s400, normalizing the delta V and inputting the network node coordinate matrix into a pre-trained maximum eigenvalue matrix solving network to solve a maximum eigenvalue matrix delta lambda max; and S500, imaging is carried out. By means of the technical scheme, muscle changes can be visually displayed from the image in the mode of the three-dimensional image and the global muscle fiber disorder coefficient, and the muscle fiber disorder degree is quantified from the numerical value.

The invention discloses a minimum inverse eigenvalue radar signal detection method adopting stochastic resonance preprocessing, and the method comprises the steps: carrying out the secondary sampling of a received signal, compressing the frequency of the signal to match a stochastic resonance system, and expanding the adaptability of the stochastic resonance system to a higher-frequency signal. Then performing stochastic resonance preprocessing on the signals, enhancing signal energy under a low signal-to-noise ratio to obtain processed signal information, sending the processed signals to a CFAR signal detector, constructing a covariance matrix of the signals of the reference unit and the detection unit, performing eigenvalue decomposition on the covariance matrix to obtain a minimum inverse eigenvalue, and finally obtaining a minimum inverse eigenvalue; the minimum inverse feature value of the detection unit is compared with the average minimum inverse feature value of the reference unit to determine the existence problem of the signal. According to the method, the problems that a matrix constant false alarm detector based on the maximum characteristic value is easily influenced by noise uncertainty and a constant false alarm detector based on information geometry has relatively high calculation complexity are solved.

The present invention belongs to the technical field of new power systems, and specifically relates to a grid-type control transformation method, system, equipment and medium. In view of the deficiencies in the calculation of the transformation capacity by the existing grid-type control transformation method, the present invention adopts the following technical scheme: the grid-type control transformation method for realizing subsynchronous oscillation suppression includes: obtaining the critical short-circuit ratio of the following grid type and the critical short-circuit ratio of the grid type; calculating the initial generalized short-circuit ratio; calculating the influence factor of the initial generalized short-circuit ratio on the rated capacity of each converter and sorting it; selecting the station with the largest influence factor for equal-capacity grid-type transformation, calculating the generalized short-circuit ratio after the system transformation and deciding whether to add the transformation station; after determining the transformation position, the maximum eigenvalue of the extended admittance matrix is ​​less than the critical short-circuit ratio of the grid type as a constraint, and solving to obtain the minimum transformation capacity. The beneficial effects of the present invention are: based on the influence factors and each short-circuit ratio, the minimum transformation capacity is obtained, and the economic cost is minimized.

The invention discloses a neural network prediction-based LMS (Least Mean Square) power inversion adaptive step length adjustment method, which relates to the technical field of satellite navigation and signal processing, and comprises the following steps of: constructing and training a neural network model, inputting a current frame input signal power and a maximum characteristic value, and outputting a current frame prediction step length; compared with an LMS power inversion algorithm with a fixed step length, the method introduces a neural network to dynamically predict the optimal step length, solves the problem that traditional empirical step length adjustment is not adaptive, reflects interference intensity in combination with a covariance matrix and a maximum characteristic value of an input signal, realizes interference sensing adjustment, can increase the step length under the condition of relatively strong interference, and improves the accuracy of interference sensing adjustment. Convergence is accelerated and the tracking capability is enhanced; and under the condition of weak interference, the step length is reduced, so that steady state imbalance is reduced, and the anti-interference performance is improved. The method can solve the problem that in an existing LMS power inversion algorithm, step length parameters are difficult to adjust in a self-adaptive mode, and especially the problem that convergence is slow or system divergence is caused by improper step length setting in a multi-interference channel or dynamic environment.

The invention provides a wheat disease identification method and device, and is applied to the technical field of crop disease identification. The method comprises the steps of obtaining a training sample set; determining mass-to-charge ratio signal intensity data of volatile organic compounds of the wheat leaf samples in the training sample set, determining a maximum characteristic value of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data, mapping the maximum characteristic values of all the mass-to-charge ratio channels into pixel points, and obtaining a two-dimensional pixel matrix of the wheat leaf samples; extracting parent ions and pixel points where fragment ions corresponding to the parent ions are located from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixel points in the two-dimensional pixel matrix to obtain a recombined feature map; and inputting the recombinant feature map into a wheat disease recognition model for network training, wherein the wheat disease recognition model is used for recognizing the disease type and infection degree of wheat according to the mass-to-charge ratio signal features in the recombinant feature map.