963 results about "Weighting coefficient" patented technology

The invention relates to a remote sensing target detection method and device and a medium, and the method comprises the steps: inputting a feature map to a backbone network for feature extraction, inputting an extracted feature tensor into a multi-branch expansion convolution structure, and extracting multi-scale features through convolution kernels with different expansion rates. Then, multi-scale features are fused through a space and channel double-path attention mechanism, and enhanced features are generated; and the enhanced features are further input into a cascade pooling module to generate multi-level reconstruction features, and weight coefficients are calculated through a gating fusion network to carry out weighted fusion, so that multi-scale fusion features are obtained. Next, these features are input into an asymmetric decomposition convolutional layer for downsampling, and dynamic channel attention calibration is performed to generate channel enhanced features. And finally, inputting the feature map processed by the backbone network and the neck network into a detection head network, and outputting a target bounding box and category prediction. According to the method, high-precision detection of multi-scale rotating targets and high-density small targets is realized in a complex remote sensing scene.

The invention is suitable for the technical field of electrical equipment state monitoring, and provides an electrical equipment state monitoring method based on a non-contact leakage current sensor, and the method comprises the steps: collecting a leakage current signal of the surface of an insulating part of electrical equipment, and obtaining infrared thermal image data and an ultrasonic signal; variational mode decomposition is carried out on the leakage current signal to obtain a plurality of intrinsic mode functions; separating a leakage current effective component and an independent noise source based on a blind source separation algorithm; extracting a time-frequency characteristic of the effective component of the separated leakage current, extracting a local temperature gradient characteristic of the infrared thermal image data and a frequency spectrum energy characteristic of the ultrasonic signal, and generating a multi-modal characteristic vector; analyzing the space-time relevance of the multi-modal feature vector, dynamically distributing each modal weight coefficient for fusion, and generating a comprehensive fault feature; according to the method, the signal-to-noise ratio of the weak leakage current signal is improved, and the misjudgment rate is effectively reduced.

The invention relates to the technical field of power system stability control, in particular to a broadband oscillation suppression method and system for an offshore wind power flexible direct current grid-connected system and a medium, and the method comprises the steps: sequentially injecting positive and negative sequence perturbation sine harmonic signals at a grid-connected point of a wind turbine generator, constructing a multi-input multi-output impedance matrix, reducing the order of the multi-input and multi-output impedance matrix into single-input and single-output equivalent positive sequence impedance; the stability margin of the offshore wind power flexible direct current grid-connected system is obtained based on equivalent positive sequence impedance quantification; when the stability margin is lower than a preset margin threshold value, reconstructing the fan grid-side converter into a matrix control structure by cascading a weighting coefficient matrix between the inner and outer ring control loops of the fan grid-side converter; and iteratively optimizing the matrix type control structure based on impedance sensitivity, and synchronously connecting a low-pass filter in series in a voltage and current feedback channel to carry out impedance remodeling so as to finish broadband oscillation risk suppression. According to the invention, broadband oscillation suppression of the offshore wind power flexible direct current grid-connected system is realized through impedance modeling and a matrix type control strategy.

The invention relates to the technical field of communication, in particular to an airborne Tiantong phased-array antenna and an application method thereof. According to the method, phased-array antenna beams are driven to perform electronic scanning according to a geometric azimuth angle, beacon signal blind acquisition is performed in a preset airspace in a wide beam mode, and when an initial signal is detected, the mode is immediately switched to a narrow beam scanning mode; dynamically adjusting a phase weighting coefficient to realize pointing adjustment, so that a signal quality parameter is continuously converged in an optimal threshold interval; carrying aircraft maneuvering acceleration feedback is introduced, beam pointing offset is dynamically compensated based on a pre-constructed interference feature library, and signal fluctuation caused by a multipath effect is suppressed; and continuously checking a data validity flag bit of the main inertial navigation system, seamlessly switching to an antenna built-in inertial navigation data stream when a failure threshold is triggered, and activating a rapid recapture program to maintain zero interruption of a communication link. Through dual-source inertial navigation data fusion and intelligent beam control, rapid and accurate tracking of Tiantong satellites in an airborne environment is realized.

The invention relates to the technical field of data processing, in particular to a backtracking analysis model construction method based on an attack chain, which comprises the following steps that: a kernel layer security agent acquires process, file and network behavior characteristics in a hardware isolation environment, and generates an event tuple; the tensor network pipeline performs three-dimensional decoupling mapping on the tuple into a behavior fingerprint vector, an orthogonalization noise feature and an asymmetric adjacent tensor, and compresses the behavior fingerprint vector, the orthogonalization noise feature and the asymmetric adjacent tensor into a space-time topology tensor block; the reinforcement learning controller constructs a directed acyclic graph based on the tensor blocks, calculates connectivity loss and outputs an event risk score; the dynamic routing engine constructs a decision tree model according to the risk mark, the burst frequency and the correlation entropy, and implements three-level shunting and a multiple simulation system to generate an anti-interference index; and when the deviation between the physical trajectory and the digital model exceeds the tolerance, the closed-loop feedback weight coefficient updates the loss function parameter and adjusts the channel resource weight. And the problem of threat discovery delay caused by attack chain breakage under massive events is solved.

An adaptive variable step-size LMS filter based on a hyperbolic tangent function and its filtering method and computer equipment belong to the field of filter technology, and solve the problem that the LMS algorithm improved based on a nonlinear function requires too many manual adjustments of parameters and that the convergence speed and convergence accuracy of the traditional LMS algorithm cannot be satisfied simultaneously. The method of the present invention includes: determining the filter tap length and the number of iterations; initializing the filter parameters; obtaining the filter input signal vector; determining the instantaneous expected signal of the filter; obtaining the instantaneous output signal; obtaining the instantaneous error signal of the filter and the accumulated amount of the error signal based on the instantaneous output signal; obtaining the instantaneous step size using a step-size update formula based on a hyperbolic tangent function; and continuously iterating the filter tap weight coefficient vector to construct an LMS filter using the filter tap weight coefficient vector obtained after the number of iterations. The present invention is suitable for improving the performance of the LMS filter, and the filter of the present invention can be widely used in various signal processing fields.

The invention discloses an equipment state monitoring method based on multi-source information fusion, and the method comprises the steps: enabling a real-time sensor to collect the operation parameter data, vibration parameter data and environment parameter data of equipment, collecting the historical state data of the equipment with a recording label, and carrying out the preprocessing of the data; time domain features, frequency domain features and working condition features of the data are extracted according to a layering mode, and dynamic weight coefficients of all the features are set; calculating a reference threshold value by adopting a specific model; evaluating the health condition of the current equipment by adopting a depth measurement method, calculating an equipment health factor, and calculating a trend compensation item; and obtaining an equipment state monitoring dynamic threshold based on the reference threshold, the health correction item and the trend compensation item. The invention further discloses an equipment state monitoring device based on multi-source information fusion, corresponding equipment and a storage medium. According to the equipment state monitoring method based on multi-source information fusion provided by the embodiment of the invention, the accuracy, real-time performance and reliability of equipment state monitoring can be effectively improved.

The invention discloses an audio signal processing method and device based on MOE, a medium and a product, and relates to the field of audio signal processing. According to the technical scheme, firstly, the audio signal processing device decomposes the audio signal into the sub-band signals in different frequency intervals, and the complexity parameter of each sub-band signal is calculated; on the basis of the sub-band signals and complexity parameters corresponding to the sub-band signals, the gating network in the hybrid expert model can intelligently distribute the optimal expert network combination for each sub-band signal, the speciality of each expert network is fully utilized, the processing efficiency is improved, the optimal expert network combination is screened by adopting a processing weight coefficient threshold, and the processing efficiency is improved. And redundant processing is avoided. And finally, the audio signal processing device obtains a structured audio data stream through quantization coding and data packaging. According to the dividing and treating method, the quality and efficiency of audio signal processing are improved, the processing strategy can be flexibly adjusted according to the signal characteristics, and the adaptability is high.

The invention relates to the technical field of bearing fault diagnosis, in particular to a multi-bearing fault positioning and diagnosing method, which comprises the following steps of: acquiring bearing vibration signals of a plurality of sensing points, extracting peak time information to complete synchronous calibration, extracting high-frequency band energy and constructing a feature vector, screening deviation features and clustering and grouping, and identifying an energy abnormal channel, positioning a bearing position, extracting time domain and frequency domain features, comparing fault modes, and outputting a fault type diagnosis result. According to the method, the consistency of multi-channel data synchronization is realized by extracting the peak point of the vibration signal and calibrating the time offset, the high-frequency energy change rate is utilized to match the weight coefficient, the detection sensitivity of tiny fault features is enhanced, and the separability of a composite fault state is improved by combining feature relative deviation rate screening and density trend clustering. A frequency section energy proportion change recognition channel is adopted, accurate positioning of a fault bearing is achieved, and accurate judgment of a fault type is achieved in combination with a multi-feature comparison mode.

The invention discloses a flexible control method for new energy grid connection, and relates to the technical field of new energy grid connection control. The method comprises the steps of obtaining grid-connected point electrical parameters through a self-measurement mode or a communication interface mode, performing harmonic feature extraction and compensation based on a deep learning network, and adopting a state estimation method to fuse multi-source data to estimate power grid parameters; and constructing a multi-objective optimization function including voltage quality, a power factor and harmonic suppression, dynamically adjusting a weight coefficient of the optimization function by using a swarm intelligence algorithm, and realizing inverter switching sequence optimization through multi-objective model prediction control. Through a dual-mode communication redundancy mechanism and local intelligent control, while a communication function of a traditional main link is reserved, a standby wireless communication link is introduced to realize real-time backup transmission of a control instruction, and through a local autonomous decision-making capability, the problem of control failure caused by a communication problem in a traditional scheme is fundamentally solved.

The invention relates to the technical field of numerical control machine tool precision detection and compensation, in particular to a coordinated motion planning control method and related products in the five-axis machining process, and the method comprises the steps: building a machine tool digital twin model, and combining constraint conditions; calculating a dynamic load rate and adjusting an error weight according to the load rate; an optimal cooperative motion instruction is obtained through multi-objective optimization; obtaining a feed-forward compensation amount and a feedback compensation amount; fusing the feedforward compensation amount and the feedback compensation amount to obtain a total compensation amount, and correcting the optimal cooperative motion instruction in real time to generate a final control signal; according to the method, dynamic constraints of different dimensions of a shaft are unified, the real-time dynamic load rate of the shaft is calculated, and weight coefficients of a translation shaft error and a rotation shaft error in an objective function are dynamically adjusted and optimized according to the difference; and constructing a multi-objective optimization function, solving an optimal cooperative motion instruction according to the dynamic weight, and integrating feed-forward compensation and feedback compensation to generate a final control signal so as to realize accurate control.

The invention relates to the technical field of telecommunication, in particular to a dual-mode communication link transmission method based on dynamic channel perception, which comprises the following steps of: firstly, detecting channel states of two links in parallel in a data transmission gap or according to a preset period to obtain a signal-to-noise ratio and a channel load rate; determining a service quality grade according to a service type of a transmission data packet, selecting an adaptive signal-to-noise ratio and a channel load rate weight coefficient, and further calculating a comprehensive quality score of the two links; dynamically calculating a power line link switching threshold based on a network average time delay requirement or a power line link historical quality stability parameter; comparing the power line link score with a threshold value and a wireless link score, and determining whether to switch or not; and finally, synchronizing the data transmission serial number and the protocol context through signaling to realize seamless switching. According to the invention, the real-time performance of channel sensing and the accuracy of quality evaluation are improved, the reasonable switching decision and the coherent data transmission are ensured, and the reliability of a communication system is enhanced.

The invention discloses a parallel unit control system applied to large and medium-sized refrigeration houses. The parallel unit control system comprises an acquisition module, an analysis module, a decision module, an execution module and a coordination module, wherein the acquisition module integrates a multi-modal data set and transmits the multi-modal data set to the analysis module; the analysis module extracts high-frequency abnormal components and current harmonic distortion features in the vibration signals by using a time sequence convolutional network, and matches the features with a fault feature library to generate fault type identification signals; the decision module generates a compressor start-stop sequence and a load distribution scheme; the execution module generates an execution effect feedback signal; and the coordination module dynamically adjusts an optimization algorithm weight coefficient through a priority arbitration mechanism, integrates an optimization instruction and self-healing feedback to generate a final control instruction, and constructs a closed-loop link to return unit operation parameters to the fault feature library to realize continuous optimization. The parallel unit control system applied to the large and medium-sized refrigeration houses can solve the problems of low energy efficiency, fault response lag and uneven equipment service life in traditional refrigeration house parallel unit control.

The invention discloses an intelligent flow sensor test analysis method and system, and relates to the technical field of sensor testing, and the method comprises the steps: applying a stepped flow excitation signal to a tested sensor through a standard flow generation device, and synchronously collecting the dynamic response data of the sensor; extracting a feature parameter set from the response data; performing similarity matching calculation to obtain a matching degree evaluation value of each error type; quantifying the induction relation strength among different error types; generating a composite weight coefficient in combination with the matching degree evaluation value and the evoked relationship strength; and making a multi-stage compensation strategy based on the composite weight coefficient. The method has the advantages that by combining error correlation network modeling driven by historical data, induction mechanisms and interaction relations of different error types can be analyzed, the pertinence of a compensation strategy is greatly improved, a multi-stage error suppression mechanism is formed, stable control over comprehensive errors is achieved under complex working conditions, and meanwhile the system maintenance requirement is lowered.

The invention relates to the technical field of network monitoring, in particular to a radio spectrum measurement system based on mobile crowdsourcing, which comprises a crowdsourcing acquisition module, a time sequence processing module, a space optimization module and a dynamic fusion module. According to the method, periodic scanning is combined with a multi-source heterogeneous data encapsulation protocol, timestamps, geographic coordinates, a signal-to-noise ratio and equipment voltage are integrated, multi-dimensional data standardized acquisition and dynamic association are achieved, time sequence weight parameters are calculated through a time sequence alignment algorithm, high and low frequency event distinction degree is enhanced, time noise is restrained, space weight coefficients are calculated through a DBSCAN algorithm, and multi-dimensional data acquisition and dynamic association are achieved. The method has the advantages that comprehensive credibility is generated through linear weighting, space-time coupling precision and evidence theory sorting credibility are improved, correlation between the credibility and the signal-to-noise ratio is analyzed, Kalman filtering is used for compensating voltage fluctuation, multi-source data fusion and error suppression are achieved, spectrum feature anti-interference capacity is enhanced, positioning deviation is reduced, and spectrum allocation efficiency and monitoring system stability are improved.

The invention provides an audio call quality detection and evaluation method and system, and belongs to the technical field of audio call detection.The audio call quality detection and evaluation method comprises the steps that an evaluation system which comprehensively covers multiple voice line states such as idle, recording and disconnection and has a multi-dimensional index collection function is constructed, and an adaptive mechanism for dynamically distributing evaluation indexes and adjusting weight coefficients is combined; accurate evaluation and comprehensive weighted analysis of call quality of each state are realized, reports are periodically collected to iteratively optimize weight coefficients by using a machine learning model, and comprehensiveness, accuracy and dynamic adaptability of evaluation results are ensured. And a final call quality report containing the comprehensive score, each state score and evaluation index data is generated, so that the practicability and the flexibility of the evaluation method are remarkably enhanced, and powerful and accurate support is provided for continuous optimization of the call quality.

The invention discloses an adaptive equalization method for combined compensation of carrier frequency offset and channel linear damage. The adaptive equalization method specifically comprises the following steps: taking data subjected to matched filtering, clock synchronization, orthogonality and normalization as input signals; for an input signal, using the filter linear damage compensation weight vector and the frequency offset compensation weight coefficient to obtain output data of the filter; obtaining an expected value of filter output data, and calculating to obtain a theoretical error value of adaptive equalization; and obtaining a filter weight vector used for next adaptive equalization according to the theoretical error value. The adaptive equalization method for combined compensation of carrier frequency offset and channel linear impairment is suitable for a coherent optical transmission system, uses an adaptive equalizer to compensate signal frequency offset and channel linear impairment at the same time, does not need an additional frequency offset estimation algorithm, meets the advantages of multiple rates and multiple modulation formats, and is high in convergence speed and high in robustness. And the calculation complexity is low.

The invention provides a data processing priority scheduling method for high-speed satellite communication and related equipment, which can effectively improve the transmission performance of a high-speed satellite communication system so as to adapt to differentiated requirements of various communication scenes. The method comprises the following steps: acquiring cross-layer state information; calling a strategy matrix, wherein the strategy matrix is used for defining weight coefficients of different data types under different network states and channel qualities; calculating a target priority score for the to-be-transmitted data stream according to the cross-layer state information and the strategy matrix; sequencing the target priority scores from high to low to obtain a sequencing result; scheduling transmission queues of the data streams for transmission according to a sorting result, and allocating cache transmission queues for the data streams with high priority scores; when it is monitored that the priority score of the new data stream exceeds a preset emergency threshold value, the new data stream is inserted into a cache transmission queue of the transmission queue, and back pressure control is carried out on data streams of other priorities.

The invention provides a satellite compass installation error adaptive correction method, device and equipment and a storage medium, and the method comprises the steps: obtaining the operation data of a ship, and carrying out the preprocessing of the operation data; then, whether the navigation state meets a correction condition or not is evaluated according to the preprocessed operation data; when it is judged that the sailing state meets the correction condition, a multi-dimensional reliability factor is calculated according to the ship speed, the course change rate and the attitude change degree, a time weighting coefficient is calculated according to the accumulated operation time of the system, and the observation noise variance of a Kalman filter is jointly adjusted based on the multi-dimensional reliability factor and the time weighting coefficient; and performing Kalman filtering adaptive state prediction and updating based on the observation noise variance so as to extract an estimated value of the installation error of the satellite compass, and applying the estimated value of the installation error of the satellite compass to course data output by the satellite compass by adopting a progressive strategy. The installation error of the satellite compass is automatically corrected, so that the navigation requirement of a ship in long-term navigation is met.

The invention discloses a duplex-screen intelligent cabin interaction method and system, and relates to the technical field of intelligent cabin interaction, and the method comprises the steps: collecting user operation behavior data, vehicle state data and environment perception data in real time through a vehicle-mounted sensor group, a man-machine interaction interface and a vehicle bus; establishing a multi-dimensional user data model containing an interaction feature vector, a driving mode matrix and a preference configuration tree; the method comprises the following steps: analyzing user historical interaction data based on an adaptive weighting algorithm, and calculating use frequency weight coefficients of different functional modules; according to the duplex-screen intelligent cockpit interaction method and system provided by the invention, the precision of a user portrait is remarkably improved by constructing a multi-dimensional user data model, personalized interaction experience is realized, and the interface layout and interaction logic can be dynamically adjusted according to a real-time driving scene by adopting a self-adaptive weighting algorithm and a random forest classifier, so that the user experience is improved. The real-time performance and the adaptability of the system are enhanced, and the interaction experience is optimized.

The invention relates to the technical field of information, and discloses a radar dynamic anti-interference method and system based on interference source positioning. The method comprises the following steps: acquiring current signal data and a historical signal sequence, and determining an initial deviation value; extracting a historical deviation sequence according to the initial deviation value to obtain a deviation change trend vector; calculating a trend slope and a fluctuation amplitude, and determining an adjustment trigger signal; extracting feature interaction influence according to the adjustment trigger signal, and determining a weight coefficient update increment; iterative optimization is carried out in combination with the convergence rate parameter, and an optimized convergence rate value is obtained; adjusting deviation compensation model parameters according to the convergence speed value, and outputting a deviation compensation result when the compensation residual error is lower than a preset threshold value; and carrying out positioning calculation according to a deviation compensation result, and determining a corrected positioning coordinate. The model is dynamically optimized through technologies such as a sliding window and gradient descent, the problem of positioning deviation caused by complex environment signal interference is solved, and the positioning precision and the response speed are improved.

The invention discloses a cable insulation aging state evaluation method, which realizes aging state evaluation of a running cable system by fusing data acquired by multiple sensors and combining an intelligent algorithm. The method comprises the following steps: 1, collecting cable operation data of different years, and constructing a preliminary data set; 2, generating simulation data to supplement the insufficiency of actually acquired data, and constructing a complete data set; 3, performing feature extraction on the signals in the complete data set; 4, carrying out weight analysis on the extracted features; 5, arranging a sensor, and collecting multi-source data in real time; 6, an alarm threshold value is set, early warning is carried out on the cable section with the aging score exceeding the threshold value, and the remaining life is evaluated; 7, synchronizing the early warning information to a maintenance management platform, and generating a detailed early warning report; and 8, checking an alarmed section, evaluating the alarm accuracy, and if the alarm is unreasonable, correcting the weight coefficient and the scoring criterion in time. By fusing data acquired by multiple sensors and combining an intelligent algorithm, the aging state of the operating system is evaluated.

The invention provides a novel multi-source information and damage coupled coal mine rock burst prediction method. The method comprises the following steps that rock mass fracture signals, a three-dimensional stress field and acoustic emission parameters are collected in real time through a micro-seismic monitoring array, an optical fiber stress sensor and a rock sound probe; an LSTM-attention fusion module is adopted to carry out spatio-temporal feature extraction on the collected multi-source data, an attention weight coefficient is generated, and a weighted fusion feature vector is output; calculating the rock mass damage degree based on a statistical damage mechanics Weibull distribution theory; based on the law of conservation of energy and the damage mechanics theory, a dynamic energy threshold function is constructed, the mining time effect and the rock mass damage state are incorporated into a quantitative evaluation system, and a dynamic energy accumulation rate threshold function is established by introducing a time accumulation effect and a damage coupling term; when the real-time energy accumulation rate is larger than or equal to the energy accumulation rate threshold value and the time derivative of the energy accumulation rate is larger than the sudden change threshold value, third-level early warning is triggered. According to the invention, advanced accurate early warning of the rock burst can be realized.

The invention relates to the technical field of sensor data fusion, in particular to a real-time dynamic attitude stability control method and system based on multi-sensor fusion, and the method comprises the following steps: collecting multi-source sensor data, and correcting to generate a synchronous sensor data set; calculating a stability coefficient, a signal quality coefficient and a consistency coefficient, weighting the signal quality coefficient and the consistency coefficient to generate a dynamic fusion weight coefficient, executing weighted fusion based on the weight, outputting a fusion attitude state estimation result through adaptive extended Kalman filtering, and generating an optimization control instruction sequence by combining with dynamic constraint model rolling optimization; according to the method, a dynamic weight mechanism based on noise and signal strength is introduced to achieve self-adaptive fusion, the synchronization precision and reliability are improved by combining consistency verification, abnormal signals are corrected through self-adaptive filtering, the attitude estimation stability is enhanced, and the performance of the system is improved. And a control instruction is generated through rolling optimization, so that rapid and accurate adjustment is realized, and attitude keeping and control consistency in a complex environment is improved.

A channel state information determination method and device, in the method, a network device determines N rows of base vectors corresponding to N time units based on mobile state information and a channel map of a terminal device, each row of the N rows of base vectors comprises X base vectors, and the channel map is used for indicating channel characteristics based on a location area; the network equipment sends first indication information, wherein the first indication information is used for indicating N rows of base vectors; the network equipment respectively sends pilot signals in N time units; the network equipment receives second indication information, the second indication information is used for indicating weighting coefficients of Y base vectors in the X base vectors, and the weighting coefficients of the Y base vectors are determined based on the pilot signal; and the network equipment determines channel state information corresponding to M time units based on the weighting coefficients of the Y base vectors and the N rows of base vectors, and the M time units comprise N time units. The method can improve the accuracy of the channel state information.

The invention discloses a super capacitor energy storage life evaluation method under a fire storage frequency modulation working condition, relates to the technical field of power system energy storage, and can at least partially solve the problems that an existing super capacitor life evaluation method is poor in adaptability to the frequency modulation working condition, insufficient in multi-factor coupling quantization and lack of engineering criteria. The method comprises the steps of simulating a fire storage frequency modulation working condition and collecting multi-dimensional operation data of a super capacitor; preprocessing the collected data and extracting life characteristic parameters such as capacity attenuation rate, equivalent C rate, average temperature rise, frequency modulation load factor and the like; key influence factors are screened through correlation analysis, weight coefficients are trained through a machine learning algorithm, and a coupling influence coefficient model is constructed; establishing a capacity attenuation equation, an equivalent series resistance growth equation and a response delay growth equation, and fusing three criteria to determine a life end point; and dynamically correcting model parameters based on the real-time frequency modulation data and calculating the residual life. Accurate evaluation and dynamic prediction of the service life of the super capacitor under the fire storage frequency modulation working condition are achieved.

The invention discloses a power system frequency risk quantification and coordination control method and system, and belongs to the technical field of power systems and automation thereof. The method comprises the following steps: collecting power grid online operation data and system working conditions; frequency response curves of the system after different faults are obtained through fault simulation; dividing the frequency into different frequency bands according to a binary criterion formed by a plurality of frequency threshold values and corresponding duration time, and distributing an initial weight coefficient for each frequency band; obtaining a dynamic weight coefficient of each frequency band by combining the frequency change rate in the frequency band and a preset weight gain coefficient; based on the dynamic weight coefficient of each frequency band, performing time integration on the frequency deviation value in each frequency band to obtain a transient frequency risk assessment value; and judging the system stability and the risk level, and giving a corresponding frequency control strategy according to the risk level so as to coordinate and control resources. According to the method, the short-time power disturbance and the permanent power disturbance can be effectively distinguished, and a basis is provided for on-line and off-line safe and stable control.

The invention provides a radar signal depth feature extraction method and system based on adversarial sample defense, electronic equipment and a storage medium, and relates to the technical field of radar signal processing.The radar signal depth feature extraction method comprises the steps that a spacecraft synthetic aperture radar original echo signal is collected and converted into a time-frequency feature map, and intra-pulse and inter-pulse features are extracted in parallel to generate a combined matrix; receiving a deception jamming signal by using a polarization radar group, generating a signal data body without polarization influence through complex coherent superposition, inputting the deception jamming signal and the signal data body into a feature decoupling adversarial network, and separating bullet micro-motion target feature data by means of mutual information maximization constraint; and finally, inputting into a multi-scale local attention module to extract multi-band characteristic components, performing adaptive weight coefficient fusion according to power entropy, performing polarization channel energy correction, and outputting anti-interference fingerprint characteristics with micro-Doppler characteristic enhancement, so that anti-interference processing of the spacecraft synthetic aperture radar signals can be realized. And warhead target anti-interference fingerprint features with micro-Doppler feature enhancement characteristics are effectively extracted.

The invention discloses a distribution network power line anti-interference communication method based on adaptive filtering, and belongs to the technical field of distribution network power line communication. According to the method, a communication signal adopting a frequency shift keying modulation mode is acquired from a power line coupler of a distribution network through a signal acquisition module, the communication signal is preprocessed by using a digital signal processing unit, spectral analysis is performed on the preprocessed signal, and characteristic parameters of interference noise are extracted; a coefficient vector of a finite impulse response adaptive filter is initialized based on the characteristic parameters, the order of the filter is in direct proportion to the reciprocal of the dominant interference frequency, fine adjustment is carried out based on noise energy distribution through a table look-up method or a linear mapping relation, and the weight coefficient of the filter is adjusted in real time in the communication process. The adjustment step length is dynamically adjusted according to the real-time estimated signal-to-noise ratio, and finally, the processed signal is output through the adaptive filter and data demodulation is carried out. The method is mainly used for improving the anti-interference capability and the communication reliability of the power private line communication of the distribution network.

The invention discloses an SAR image despeckle method and device, and relates to the technical field of image processing. The method comprises the following steps: constructing an SAR image despeckle model based on an encoder-decoder structure; acquiring an SAR observation image polluted by noise; extracting shallow layer features of the SAR observation image; separating the shallow features into high-frequency features and low-frequency features; aggregating the spatial context information of the high-frequency features and encoding the spatial context information into a query matrix, a key matrix and a value matrix; calculating high-frequency output characteristics according to cross covariance attention among the matrixes; extracting noise distribution features in the low-frequency features and splicing the noise distribution features with the low-frequency features to obtain low-frequency output features; learning complementarity weight coefficients of the high-frequency output features and the low-frequency output features, and fusing the high-frequency output features and the low-frequency output features to obtain deep features; reconstructing the deep features output by the last layer of the decoder, and outputting a residual image; and connecting the residual image with the SAR observation image residual to obtain an SAR freckle-removed image.