127 results about "Time frequency distribution" patented technology

The invention relates to a bridge structure vibration monitoring and analysis method based on artificial intelligence, and the method specifically comprises the following steps: setting an acceleration sensor network at a key part of a bridge, collecting and marking a vibration signal sample, and forming a data set; a high-resolution time-frequency matrix of sample adaptive optimal kernel time-frequency distribution is calculated, a damage sensitive resonance frequency band is positioned according to spectrum kurtosis, and the time-frequency matrix is enhanced by the resonance frequency band subjected to adaptive gain enhancement; dividing a matrix frequency axis multi-scale binary tree, calculating and normalizing sub-band average energy, quantifying energy distribution uniformity through information entropy, and splicing multi-scale entropy values into feature vectors; then constructing and training a bridge detection data model; and finally, preprocessing new data, inputting the preprocessed new data into the trained model, and automatically analyzing to obtain a bridge health monitoring result. Through time-frequency optimization, multi-scale entropy and deep learning technologies, the problems of large noise, difficult feature extraction and low automation and accuracy of traditional monitoring can be solved.

The invention discloses a live detection method for insulation defects of high-voltage power equipment. The method comprises the following steps: firstly, synchronously arranging ultrahigh frequency sensors and acoustic emission sensors on a plurality of monitoring points on the surface of a gas insulated switchgear shell to form an array, and synchronously acquiring signals for filtering pretreatment; then, pulse events are extracted from the two types of signals respectively, and associated pulses in a time window are matched into matched pulse pairs representing the same discharge source; then, carrying out time-frequency transformation on the ultrahigh frequency and acoustic emission pulse waveform in each matched pulse pair, carrying out joint noise reduction by calculating a coherence coefficient between time-frequency distribution matrixes, and extracting a joint feature vector containing an energy ratio, a time parameter ratio and a frequency difference from the denoised time-frequency matrix; according to the invention, multi-source signals are fused, and high-sensitivity detection, high-precision positioning and high-accuracy identification of insulation defects are realized.

The invention provides a passive perception radar target classification method and system based on electromagnetic detection feature extraction, and the method comprises the steps: carrying out the multi-dimensional signal decomposition of an original scattered signal sequence through obtaining an electromagnetic radiation signal set of a target region; obtaining a time-frequency distribution feature set and a polarization response feature set corresponding to each signal acquisition period, extracting a target electromagnetic feature combination from the time-frequency distribution feature set and the polarization response feature set according to a preset electromagnetic feature association rule, generating a weight distribution parameter, and performing weighted fusion processing on the target electromagnetic feature combination to obtain a target electromagnetic feature combination; and the fusion feature vector set is input into an adaptive classification network, hierarchical classification operation is performed on the fusion feature vector set, the category identifier of the target scatterer and the corresponding classification confidence are output, and a radar detection optimization instruction is generated based on the category identifier and the corresponding classification confidence. According to the method, high-precision distinguishing of multiple types of scatterers is achieved under the passive sensing condition, and meanwhile the environmental adaptability and the calculation efficiency of the classification process are guaranteed.

The invention discloses a coal-fired power plant belt conveyor driving gear box bearing fault diagnosis method, which relates to the field of fault diagnosis and evaluation, and is used for extracting gear box bearing fault features based on an order tracking analysis method of tachometer-free instantaneous rotation frequency estimation. The method specifically comprises the following steps: S1, obtaining an instantaneous rotation frequency ridge line through the time frequency distribution of a vibration signal, carrying out the rotation frequency estimation of a rotating shaft, and improving the precision of the instantaneous rotation frequency ridge line through an improved multi-source ridge line fusion algorithm; and S2, extracting bearing fault feature order information by using a calculation order tracking COT algorithm, and adaptively processing a bearing fault diagnosis problem of the driving gearbox of the belt conveyor. According to the bearing fault diagnosis method for the rubber belt conveyor driving gearbox of the coal-fired power plant, the bearing inner ring fault is successfully diagnosed, and the feasibility of a bearing fault diagnosis strategy for the rubber belt conveyor driving gearbox is verified.

The invention belongs to the technical field of rotating machinery signal processing, and discloses a multi-shafting rotating machinery rotating speed estimation method based on a displacement vibration signal, and the method comprises the steps: collecting the displacement vibration signal, and carrying out the short-time Fourier transform to obtain a time-frequency distribution matrix; screening out a safety harmonic set of each shafting based on the estimated rotating speed range; calculating a definition evaluation value of each safety harmonic wave and selecting an optimal harmonic wave as a target tracking harmonic wave; constructing a cost function fusing amplitude and frequency abrupt change penalty, and performing ridge line tracking in the local time-frequency matrix after noise reduction to obtain a preliminary estimation time-frequency curve; and constructing a demodulation phase function based on the curve, performing generalized demodulation and narrow-band filtering on an original signal, extracting a pure harmonic component, and finally reconstructing a final rotating speed curve of a corresponding shaft system through instantaneous phase differential. According to the method, high-precision and anti-interference decoupling and estimation of the rotating speed of each shafting under a strong interference background are realized.

The invention discloses a ship signal adaptive time-frequency analysis method based on instantaneous frequency direction self-adjustment. The ship signal adaptive time-frequency analysis method comprises the steps of collecting ship signals to be subjected to time-frequency analysis and performing short-time Fourier transform; estimating the instantaneous frequency of the ship signal based on short-time Fourier transform time-frequency distribution in combination with time-frequency narrow-band features; calculating the instantaneous frequency and the optimal window length after smoothing; segmenting the smoothed instantaneous frequency in time, calculating time-frequency domain direction pre-guidance of each time period, and constructing a Chirp atomic direction over-complete dictionary; segmenting the ship signal in time, and calculating the matching direction of the ship signal in each time period so as to calculate and improve Chirplet transformation time-frequency distribution; and calculating the optimal short-time Fourier transform in combination with the optimal window length, and correcting and improving Chirplet transform time-frequency distribution. According to the method, the robustness of short-time Fourier transform and the anisotropy of improved Chirplet transform are utilized at the same time, and time-frequency distribution with high robustness and high time-frequency resolution is obtained.

The invention provides a rotating speed extraction method, system and equipment based on time-frequency analysis and a medium. The method comprises the following steps: acquiring a signal x (t) to be analyzed; performing spectral analysis transformation on the to-be-analyzed signal x (t) to obtain a spectral analysis result; acquiring a three-dimensional time-frequency distribution diagram A (t) (f) of the to-be-analyzed signal x (t) according to the spectral analysis result; searching a peak frequency in the three-dimensional time-frequency distribution diagram A (t) (f) according to a preset extraction rotating speed frequency Nfn and a preset search range so as to obtain the rotating speed fn (t); according to the method, the rotating speed of the equipment can be accurately extracted according to the monitoring data under the condition of not depending on an external rotating speed sensor; the system has the same beneficial effects.

The invention discloses an ant colony optimization energy focusing cable insulation damage positioning method and device, and relates to the technical field of cable detection engineering, and the method comprises the steps: generating a Gaussian envelope linear frequency modulation chirp detection signal, injecting the detection signal into a to-be-detected cable, and collecting an incident signal and an echo signal containing a reflection component; constructing a parameterized time-frequency analysis kernel function family under a Cohen type time-frequency distribution framework, and searching and optimizing kernel parameters by using an ant colony optimization algorithm; and calculating the time frequency distribution and energy-time curve of the signal based on the energy focusing time frequency analysis kernel function obtained through optimization, extracting the arrival time corresponding to an energy peak value, and realizing cable insulation damage positioning according to the reflection time delay and the propagation wave velocity. According to the method, energy peak focusing and peak position stability are enhanced while cross terms are restrained, defect positioning precision and robustness are remarkably improved, and the method is suitable for engineering online detection and automatic implementation of cable insulation damage.

The invention provides a passive perception radar AI weak target detection method and system, and the method comprises the steps: obtaining a multi-dimensional electromagnetic signal data set of a target monitoring region, carrying out the signal feature extraction of the multi-dimensional electromagnetic signal data set, and obtaining the time-frequency distribution features and signal mode features corresponding to the multi-dimensional electromagnetic signal data set; based on a pre-trained weak target identification model, carrying out joint analysis processing on the time-frequency distribution characteristics and the signal mode characteristics, and generating weak target confidence parameters of potential weak target reflection signals; generating a weak target detection result of the target monitoring area according to a comparison result of the weak target confidence coefficient parameter and a preset confidence coefficient threshold value; and transmitting the weak target detection result to the target terminal device to trigger an alarm response operation. According to the method, the problem of characteristic loss caused by signal preprocessing and target detection module separation in a traditional method is solved, and the real-time performance of weak target detection and the system robustness are remarkably improved.

Example systems, methods, and apparatus are disclosed herein for a high-resolution time-frequency analysis for nonstationary signals. Time-frequency distributions (TFDs) are essential tools for analyzing nonstationary signals in various applications. The proposed technology, Zoom, a novel time-frequency distribution (TFD) designed to enhance the resolution and reduce cross-term interference. The performance of the proposed TFD is rigorously evaluated by studying its behavior across different smoothing values, allowing for a detailed analysis of its time-frequency localization capabilities. To quantify its effectiveness, the proposed technology computes several performance metrics, including time-frequency concentration and cross-term reduction.

The invention relates to a rotary machine composite fault detection method and system under a time-varying working condition, and belongs to the technical field of rotary machine fault diagnosis. The method comprises the steps of collecting a non-stationary vibration signal under a variable speed working condition, fusing a third-order Renyi entropy and kinematics prior to construct an adaptive time-frequency enhancement model, and outputting weighted time-frequency distribution; global optimization is carried out through a priori constraint Viterbi algorithm, and an instantaneous frequency ridge line is extracted under the condition of no rotating speed sensor; performing angular domain resampling by taking the ridge line as a phase reference, and converting a non-stationary signal into an angular domain stationary signal; constructing an extended difference mode decomposition convex optimization model, and iteratively solving an optimal difference spectrum; and determining a frequency spectrum segmentation threshold value based on self-adaptive change point analysis, performing decoupling separation and time domain reconstruction on the optimal difference spectrum, and completing composite fault diagnosis through feature comparison. The method does not need an external tachometer, can adaptively extract and separate the composite fault features under the variable speed working condition, is high in diagnosis accuracy, and is good in engineering application value.

The invention relates to the technical field of ultrasonic signal processing, in particular to an ultrasonic data processing method and system and electronic equipment, and the method comprises the steps: obtaining an original ultrasonic echo signal, and carrying out the time gain compensation, band-pass filtering and down-sampling preprocessing; extracting dynamic frequency dispersion characteristic parameters based on Coann class time-frequency distribution; constructing a depth self-adaptive phase compensation model, and correcting nonlinear phase distortion segment by segment; hilbert transform is adopted to extract an envelope, and after logarithmic compression and linear mapping are carried out, image reconstruction is completed by utilizing delay summation beam forming and an adaptive contrast enhancement algorithm. According to the method, the imaging definition is improved through cooperation of frequency dispersion modeling and phase correction, and the method has good depth adaptability and image visualization effect and is suitable for high-precision image acquisition requirements in scenes such as medical ultrasonic imaging and industrial nondestructive testing.

The application provides a method for reducing impurities in electrolyte based on a copper foil production device, comprising: identifying a pressure fluctuation abnormal area from a pressure time-frequency distribution map as an input of a pre-constructed high deposition area distribution model, outputting a position distribution of the high deposition area, and determining a potential high-risk node of impurity deposition in a pipeline; deploying a filtering device at the potential high-risk node of impurity deposition according to the determined potential high-risk node of impurity deposition in the pipeline, separating impurity particle sizes through a multi-stage filtering structure built in the filtering device, directionally collecting different particle size impurities separated, and obtaining impurity data.

The application discloses a wheeled traffic target classification and identification method, system, device and medium based on micro-motion information, and the method comprises the following steps: acquiring a micro-Doppler echo signal of a wheeled target; performing a short-time Fourier time-frequency transformation on the micro-Doppler echo signal to obtain a time-frequency distribution of the micro-Doppler echo signal; calculating an instantaneous frequency of the micro-Doppler echo signal, performing an EMD transformation on the instantaneous frequency, obtaining a plurality of modal components, and selecting a single-frequency component from the plurality of modal components to obtain a wheel rotation period; according to the time-frequency distribution of the micro-Doppler echo signal, the number of flicker points in the wheel rotation period is obtained, and the corresponding maximum speed in the time-frequency distribution is also obtained; and the data set composed of the number of flicker points in the wheel rotation period, the maximum speed and the corresponding vehicle type is input into a support vector machine to classify and identify the vehicle. The application can finely classify and identify different wheeled vehicles.

The invention discloses a target deformation detection method based on Doppler feature guidance, and the method comprises the following steps: 1, obtaining radar echo signal data, and calculating distance-slow time two-dimensional data; 2, acquiring signal data at a target position and signal data at a reference position, and calculating a phase at the target position and a phase at the reference position; 3, calculating time-frequency distribution of the signal data at the target, extracting Doppler features of the signal data, and recording an existence time set of the Doppler features; 4, compensating the phase at the target position by using the phase at the reference position to obtain target deformation; and 5, in the Doppler feature existence time set, performing deformation detection on the target deformation to obtain a target deformation displacement set and an occurrence time index set. According to the method, the radar is used for counteracting shaking of the platform perceived by the phase at the reference position, detection is guided by combining Doppler features related to target deformation, and the high deformation detection rate can be achieved with the low false alarm rate.

The application provides a locomotive gear fault detection method and device and a server, relates to the technical field of locomotive gear fault diagnosis, and comprises the following steps: performing median filtering and mean filtering on an original time-frequency distribution to obtain noise estimation values of each time in a time-frequency distribution matrix of the original time-frequency distribution, and determining a true signal component discrimination threshold of each time; determining a first time-frequency ridge line based on the noise estimation values and the true signal component discrimination threshold, and establishing a dynamic optimization penalty function according to the first time-frequency ridge line to determine a second time-frequency ridge line; connecting and processing discontinuous time-frequency ridge lines with endpoint acceleration values through linear interpolation fitting to determine a repaired second continuous time-frequency ridge line, so as to determine a target time-frequency ridge line; and determining a locomotive gear fault detection result by using the target time-frequency ridge line and a gear transmission ratio relationship. The application can significantly enhance the extraction capability of the time-frequency ridge line, and further improve the accuracy of gear fault detection.

The present invention discloses an intermittent sampling interference method and system based on pseudo-random sequence frequency shift modulation, which relates to the technical field of radar electronic countermeasures, including intermittent sampling of enemy radar signals to obtain intermittent sampling signals; setting a pseudo-random sequence and a frequency shift amount, constructing a frequency shift amount function and a frequency shift modulation function; using the frequency shift modulation function to modulate and forward the intermittent sampling signal to obtain a frequency shift modulation interference signal; then performing matched filtering processing, parsing the filtered interference signal to obtain an optimal frequency shift amount matrix; substituting the optimal frequency shift amount matrix into the filtered interference signal to obtain an optimal interference signal, and implementing interference on the enemy radar. The present invention causes false targets to randomly superimpose around the real target, forming strong false targets with random distributions of leading, nearby, and lagging; not only adjusting the amplitude and position distribution of the strong false targets, but also destroying the time-frequency distribution and spatial distribution characteristics of the intermittent sampling interference, having both suppression and deception interference effects.

The invention relates to the technical field of voice signal processing, and particularly discloses an AI accompanying robot voice synthesis method and system based on voiceprint recognition, and the method comprises the steps: collecting a user voice signal, carrying out the preprocessing, and obtaining time-frequency distribution data and a system state parameter set; based on a time-frequency analysis result, extracting robust identity features robust to emotion changes; based on a nonlinear dynamic analysis result, extracting dynamic emotional characteristics such as emotional confusion degree, stability, intensity baseline and fluctuation intensity; standardizing the two types of features and carrying out dynamic weighted fusion according to confidence to form a decoupling feature vector for executing identity confirmation and emotion recognition in parallel; when low-confidence identity judgment and high-intensity emotion conflict occur, historical interaction records are introduced for arbitration correction; and finally generating a synthesis instruction of emotional rhythm control, and driving an end-to-end speech synthesis model to output natural speech with both user identity timbre and corresponding emotional color.

This application relates to a terahertz video synthetic aperture radar moving target imaging method based on time-frequency analysis. The method includes: constructing a radar echo model of the moving target based on instantaneous range and radar echo; processing the radar echo model by performing a short-time Fourier transform on the compressed echo signal; obtaining the Doppler center frequency of the scene echo from the obtained time-frequency distribution map; performing Doppler frequency zero-padding, range migration correction, and azimuth processing on the compressed echo signal according to a pre-set function; observing the range cell position of the moving target from the obtained imaging results; and extracting and performing time-frequency analysis on the moving target signal. After determining the Doppler center frequency of the moving target through time-frequency analysis and the time-frequency map, velocity estimation is performed; and the image of the moving target is refocused based on the characteristics of the time-frequency distribution line of the moving target echo. This method can quickly achieve velocity estimation and image refocusing of moving targets.

The application discloses a ship active signal time-frequency analysis method based on time-frequency parameter fine estimation, and comprises the following steps: (1) obtaining a ship active signal to be processed; (2) calculating a single-frequency signal energy fluctuation characteristic of the ship active signal to be processed; (3) calculating a received single-frequency signal time existing range based on the single-frequency signal energy fluctuation characteristic; (4) calculating a received single-frequency signal frequency based on the received single-frequency signal time existing range; and (5) calculating a received frequency modulation signal time-frequency distribution based on the received single-frequency signal frequency. The method is suitable for a low signal-to-noise ratio condition, can effectively calculate a time-frequency distribution of a frequency modulation signal in a two-component ship active signal composed of the frequency modulation signal and a single-frequency signal, and provides a basis for target feature extraction based on the ship active signal.

This application belongs to the field of rotating machinery signal processing technology, and discloses a method for estimating the rotational speed of multi-axis rotating machinery based on displacement vibration signals. The method involves acquiring displacement vibration signals and performing short-time Fourier transform to obtain a time-frequency distribution matrix; selecting a safe harmonic set for each axis system based on the estimated rotational speed range; calculating the sharpness evaluation value of each safe harmonic and selecting the optimal harmonic as the target tracking harmonic; constructing a cost function that fuses amplitude and frequency mutation penalties; performing ridge tracking in the denoised local time-frequency matrix to obtain a preliminary estimated time-frequency curve; constructing a demodulation phase function based on this curve; performing generalized demodulation and narrowband filtering on the original signal to extract clean harmonic components; and finally reconstructing the final rotational speed curve of the corresponding axis system through instantaneous phase differentiation. This method achieves high-precision, anti-interference decoupling and estimation of the rotational speed of each axis system under strong interference conditions.

The application relates to the technical field of audio processing, and discloses a voice activity detection method, a system, a terminal and a storage medium. The method comprises the following steps: continuously monitoring an input audio signal through a first detection layer, performing preliminary voice judgment on each frame of audio signal based on basic acoustic characteristics, and generating a wake-up signal to start a second detection layer when the judgment result meets a first preset trigger condition; the second detection layer extracts at least two acoustic characteristics reflecting voice harmonic structure and time-frequency distribution characteristics of the audio signal in response to the wake-up signal; all acoustic characteristics reflecting voice harmonic structure and time-frequency distribution characteristics are subjected to fusion analysis, and when the fusion analysis result meets a second preset trigger condition, it is determined that the current audio signal is a candidate audio signal, and an activation signal is generated to start a third detection layer; the third detection layer inputs the candidate audio signal into a preset deep learning model after processing the candidate audio signal in response to the activation signal, so as to output a voice activity detection result.

The application provides a clastic rock reservoir seismic sequence high-resolution division method and system, S1 obtains the sensitive well logging curve of the well closest to the seismic trace, adopts a cumulative prediction error method to process the sensitive well logging curve to obtain a cumulative prediction error curve, and performs generalized S transformation on the sensitive well logging curve to obtain time-frequency distribution characteristics; S2 performs long-term and short-term cyclic sequence interface identification by using the cumulative prediction error curve and the time-frequency distribution characteristics to obtain long-term and short-term cyclic sequence interfaces; S3 calibrates the long-term and short-term cyclic sequence interfaces to the well seismic trace, performs well seismic trace seismic sequence division, adopts an atomic matching tracking technology to perform well-seismic time-frequency matching analysis in the time-frequency domain, identifies seismic implicit sequence interfaces, and obtains a clastic rock reservoir high-resolution seismic sequence framework after all seismic traces are calculated. The application obtains a high-resolution seismic sequence framework, and provides strong support for improving the precision and accuracy of clastic rock reservoir prediction.

This invention provides a method and apparatus for fault diagnosis of rolling bearings. The method involves acquiring the vibration acceleration signal of the rolling bearing and then obtaining its time-frequency distribution, time-domain waveform recursion graph, and spectral recursion graph. A combined tensor is then formed based on these graphs to construct a fault diagnosis model for the rolling bearing. Finally, test sample signals are input into the trained fault diagnosis model to obtain the fault diagnosis result. This invention combines the recursion graph analysis method with time-domain and frequency-domain signals, improving the diagnostic accuracy and generalization performance of the fault diagnosis model. Furthermore, through training and testing the fault diagnosis model, the diagnostic accuracy of the fault model under complex environments can be further improved.

The invention belongs to the technical field of time-frequency spectrum decomposition, and discloses a time-frequency spectrum decomposition method and device based on empirical matching pursuit and a storage medium, the time-frequency spectrum decomposition method comprises the following steps: constructing an empirical mode decomposition (EMD) dictionary suitable for time-frequency analysis according to input seismic data; performing matching pursuit on a seismic signal to be decomposed by using the EMD dictionary to obtain a time-frequency atom combination of the seismic signal; and performing multiple synchronous compression transformation on the time-frequency atoms of the seismic signal to obtain the time-frequency distribution of the seismic signal. According to the method, a new time-frequency characterization method is deduced, a time-frequency spectrum decomposition strategy based on empirical matching pursuit is designed, and the time-frequency resolution of seismic signals is improved by introducing multiple synchronous compression transformation.

The application discloses a Doppler through-wall radar positioning method based on neural network time-frequency analysis, comprising the following steps: constructing a multi-target motion echo signal after demodulation of a Doppler through-wall radar; performing time-frequency analysis processing on the signal through Winger-Ville distribution to obtain corresponding time-frequency distribution and construct a fusion data set; constructing a multi-input neural network model, training the model by using the fusion data set, updating the model parameters, and constructing an optimal time-frequency distribution prediction model; acquiring real-time collected echo signals, performing time-frequency analysis processing through the distribution to obtain corresponding time-frequency distribution, and constructing a real-time fusion data set; obtaining a predicted time-frequency distribution; extracting the multi-target instantaneous frequency of the predicted time-frequency distribution, estimating the distance and angle information of the target through a Doppler through-wall radar positioning algorithm, and completing the positioning of the multi-target; and the method can eliminate the interference of time-frequency distribution cross terms, improve positioning accuracy, and improve time-frequency analysis effect.

The application discloses a noise attention mechanism-based electroencephalogram time-frequency domain denoising method, which performs time-frequency transformation on a noisy electroencephalogram signal to generate a complex time-frequency feature; extracts a time domain and a frequency domain noise mask through a double-path neural network and fuses the time domain and the frequency domain noise mask; a double-stage complex convolutional network is constructed, a noise attention module of the double-stage complex convolutional network utilizes feature splitting and a spatial weight mechanism, a time-frequency distribution of noise is located in a first stage, and clean signal reservation and noise suppression are cooperatively optimized in a second stage; finally, a denoised signal is reconstructed through inverse transformation; the application simultaneously combines time domain and frequency domain information of the complex convolutional neural network electroencephalogram signal, and based on a noise attention mechanism of time-frequency domain segmentation, more accurately separates artifacts in the electroencephalogram signal and fully reserves useful electroencephalogram signals, while improving the interpretability of the model, and solves the problem that existing electroencephalogram signal denoising methods fail to consider time-frequency domain combination denoising; in addition, the application also includes a system, a device and a medium capable of storing and running the above method.

The invention discloses a denoising radar intra-pulse signal modulation high-precision identification method, which comprises the following steps of: 1, acquiring a radar signal to be identified, and carrying out Choi-Williams time-frequency distribution preprocessing on the radar signal to be identified to obtain a noise-containing two-dimensional time-frequency image to be identified with corresponding Choi-Williams time-frequency distribution; 2, constructing and training a real image denoising network, and enabling the real image denoising network to convert the input noise-containing two-dimensional time-frequency image input into a corresponding noise-free two-dimensional time-frequency image; and step 3, constructing and training a Swin-Transform network, and obtaining a feature extraction intelligent identification network model. And step 4, inputting the noise-containing two-dimensional time-frequency image to be identified in the step 1 into the real image denoising network which completes training to obtain a noise-free two-dimensional time-frequency image. And step 5, inputting the noise-free two-dimensional time-frequency image in the step 4 into the Swinin-Transform network obtained in the step 3, and obtaining a corresponding recognition result of the radar signal.

The invention discloses an underwater rotating machinery vibration feature extraction method based on time-frequency energy analysis. The method comprises the following steps: preprocessing an original mechanical vibration signal of the underwater rotating machine to obtain a filtered vibration signal; performing Fourier synchronous compression transformation and then constructing a time-frequency distribution matrix; establishing an optimal path model containing a smooth constraint term, and obtaining an instantaneous frequency trajectory so as to estimate an optimal time-varying rotating speed; performing time domain-angle domain mapping on the filtered vibration signal, constructing an angle domain vibration signal, and further constructing a time-order distribution matrix; and determining a dominant order component and an optimal operation condition of vibration from two dimensions of an order domain and a time domain. According to the method, the vibration signal characteristics of the underwater rotating machinery under the variable working condition can be extracted with high precision, the rotating speed time mapping relation is estimated under the condition of no rotating speed sensor, the excitation characteristics and the vibration intensity are extracted through the order and rotating speed energy distribution characteristics, and the method has important engineering application value in vibration monitoring and optimization design.

The invention provides an aero-engine strong non-stationary cross frequency signal time-frequency analysis method and system, and relates to the technical field of aero-engine signal processing and fault diagnosis, and the method comprises the steps: obtaining a vibration signal, and obtaining initial time-frequency distribution; performing adaptive ridge line extraction on the initial time-frequency distribution, and determining a time-frequency block range of an intersection region; carrying out local sub-blocking on the time frequency distribution in the intersection area to obtain optimized time frequency distribution; performing component reconstruction on the vibration signal by adopting Vold-Kalman order filtering to obtain final time frequency distribution; according to the method, the ridge line fracture time-frequency block is determined through bidirectional retrieval repair, the intersection area and the hierarchical cascade reconstruction component are optimized through comprehensive evaluation indexes, the ridge line continuity and instantaneous frequency extraction precision of the intersection area are remarkably improved, and high-resolution representation and accurate separation of strong non-stationary intersection frequency signals are achieved.