42 results about "Hilbert spectrum" patented technology

The invention provides a freezing circle watershed hydrological prediction method and system based on multi-source data fusion, and relates to the technical field of hydrological monitoring and sediment monitoring, and the method comprises the steps: obtaining a vibration signal of the interaction of a water flow and a riverbed and meteorological data in a region, and carrying out the preprocessing; the preprocessed meteorological data and vibration signals are input into a multi-source data fusion model, sequential feature extraction is conducted on the meteorological data and the vibration signals through a multi-layer perceptron in the multi-source data fusion model, the extracted sequential features are fused through a gating circulation unit, and multi-source data fusion features are obtained; and carrying out empirical mode decomposition and Hilbert spectrum analysis on the preprocessed vibration signal to obtain a plurality of signal characteristics such as instantaneous frequency and instantaneous amplitude of the vibration signal, inputting the signal characteristics and the multi-source data fusion characteristics into a hydrological parameter prediction model, and outputting corresponding predicted values of the water flow velocity, the flow and the sediment transport capacity.

The invention discloses a dynamic fluidized bed system and a dynamic fluidized separation method, belongs to the technical field of fluidized particle separation system measurement, and aims to solve the problem of poor detection precision of a mixed separation state of pulverized coal in a separation process in the prior art. In the dynamic fluidized bed system, a pressure acquisition unit acquires pressure signals at different positions of a bed layer in the vertical direction and transmits the pressure signals to a controller, the controller receives the pressure signals, calculates pressure difference signals, converts the pressure difference signals into density signals, and processes the density signals to generate a Hilbert map of particles in the bed layer; and judging whether the bed layer is in a mixed state or a separated state according to the Hilbert spectrum. The method can be used for dynamic fluidized separation of raw coal.

The invention belongs to the technical field of machine learning, and provides a harmonic reducer fault diagnosis method combining a deep migration network and a fusion sample, which comprises the following steps: collecting dynamic characteristic signals of a harmonic reducer by using a multi-channel sensor, and carrying out full-period division; performing signal decomposition on the dynamic characteristic signals after the whole period division by using a dragonfly optimization algorithm, and extracting an intrinsic mode function set; carrying out Hilbert transform to obtain a Hilbert spectrum, obtaining time-frequency images in three axial directions, carrying out multichannel image fusion on the time-frequency images through an image integration method in a wavelet domain, constructing a fused image sample, dividing the fused image sample into a training set and a test set, and carrying out label calibration; training a CBAM-based fault diagnosis model by using the training set of the calibration label, and performing reverse parameter adjustment by considering domain migration loss and cross entropy loss; and inputting the test set into the trained CBAM-based fault diagnosis model, and outputting a fault diagnosis result. According to the invention, fault diagnosis under different working conditions can be accurately obtained.

The invention discloses an excitation load grounding fault detection method, system and device based on EEMD and Hilbert spectrum analysis and a medium, and belongs to the technical field of power system energy storage equipment, and the method comprises the steps: collecting and preprocessing an electric signal, and constructing a discretization time domain data sequence; performing iterative decomposition on the sequence to obtain multiple groups of components; time domain structure analysis is carried out, and target components containing fault mutation features are screened; frequency domain transformation is carried out on the target component to construct spectrum distribution, and normalization processing is carried out; and then time-frequency joint transformation is carried out, energy distribution and energy entropy indexes are calculated, and diagnosis is completed. According to the method, modal aliasing is suppressed by using EEMD adaptive noise injection and a set average strategy, accurate separation of high-frequency transient disturbance and low-frequency harmonic waves is realized, multi-dimensional feature combined diagnosis is formed by combining frequency domain screening and time-frequency analysis, and the fault detection precision and reliability are improved.

The invention relates to the technical field of fan blade fault diagnosis, and discloses a fan blade fault diagnosis system and method based on HHT and DBSCAN. Synchronously acquiring vibration signals acquired by a three-axis acceleration sensor and wind speed and rotating speed working condition data acquired by an SCADA (Supervisory Control and Data Acquisition) system; eMD empirical mode decomposition is carried out on the collected vibration signals, the vibration signals are decomposed into a limited number of IMF signals, and effective IMF signals are screened; generating a Hilbert spectrum through HHT, extracting multi-scale frequency band energy features, and fusing SCADA data to carry out working condition adaptive normalization on the features; and then, processing the normalized features by adopting a self-adaptive DBSCAN clustering algorithm based on K-distance map parameters, and realizing fault diagnosis by identifying an abnormal cluster with a sample proportion of less than 5% and combining with a frequency band energy deviation threshold. According to the method, the problems that a traditional method is not thorough in decomposition of non-stationary signals, depends on labeled samples and is insufficient in clustering robustness are solved, and unsupervised and high-precision blade state monitoring is achieved.

The invention belongs to the technical field of battery replacement, and provides a battery replacement cabinet system applied to an electric bicycle, which comprises the following steps: analyzing the fluctuation condition of temperature data and establishing a dynamic threshold compensation mechanism by using the temperature data collected by a temperature sensing monitoring unit and a composite monitoring unit; the method comprises the following steps: performing time sequence construction on temperature data collected by a temperature sensing monitoring unit and a composite monitoring unit, analyzing the time sequence to obtain an intrinsic mode function component, converting the intrinsic mode function component to obtain a Hilbert spectrum, distinguishing a low-frequency intrinsic mode function component and a high-frequency intrinsic mode function component according to the Hilbert spectrum, and calculating the intrinsic mode function component according to the low-frequency intrinsic mode function component and the high-frequency intrinsic mode function component. Integrating the low-frequency intrinsic mode function and the high-frequency intrinsic mode function to obtain a temperature change time domain envelope line; analyzing the slope of the smooth rising stage of the temperature change time domain envelope, judging whether the charging temperature change of the battery is normal or not in real time, and if not, generating a pre-alarm signal; and based on the generated pre-alarm signal, analyzing a Pearson's correlation coefficient between the monitoring units, and positioning and adjusting a temperature anomaly region.

The invention discloses a grinding wheel online state monitoring system and method based on multi-sensor fusion. The monitoring system is composed of an acquisition module, a data preprocessing module and a man-machine interaction module. Firstly, multi-source information collection is carried out through a sensor, then time-frequency domain analysis and wavelet transformation are carried out on the collected information in a targeted mode, and data feature values are extracted; according to the method, image characteristic values such as a Hilbert spectrum are obtained through Hilbert-Huang transform and fast Fourier transform, high-dimensional data are subjected to importance evaluation through a reelief algorithm to obtain multi-modal data with high correlation, a ResNet-DenseNet-LSTM hybrid network model is built, accurate monitoring of the online state of the grinding wheel is achieved through data driving, and the accuracy of the online state of the grinding wheel is improved. And finally, integrating a monitoring algorithm in an upper computer interface. According to the invention, sufficient monitoring precision and accuracy can be ensured, online monitoring is realized, and a complicated process of offline data processing is avoided.

The invention provides a small sample radiation source individual identification model training method and device and electronic equipment, and relates to the technical field of radiation source individual identification, and the method comprises the steps: obtaining an original signal sample, and carrying out the form conversion of the original signal sample, and obtaining a multi-domain signal sample; the multi-domain signal sample comprises same-direction-orthogonal component data, bispectrum data and Hilbert spectrum data; performing phase rotation on the multi-domain signal sample to obtain an enhanced signal sample; performing feature extraction on the enhanced signal sample, and performing feature fusion on the extracted features to obtain multi-domain embedded features; performing classification based on multi-domain embedded features to obtain a prediction result of an enhanced signal sample; and constructing multi-task loss based on the multi-domain embedded features, the prediction result and the sample label of the original signal sample, and performing model optimization based on the multi-task loss to obtain a trained radiation source individual identification model. According to the invention, radiation source individual identification in a small sample scene can be carried out, and the identification accuracy is improved.

The invention discloses a complex nonlinear MIMU (Micro Inertial Measurement Unit) signal noise reduction method based on improved modal decomposition. Dynamic adaptive optimization of noise parameters is realized by introducing Hilbert spectrum analysis into a CEEMDAN (Complex Empirical Empirical Mode Decomposition Number) algorithm; an IMF component discrimination model is constructed based on multi-dimensional feature fusion, and accurate classification of IMF components is realized; and aiming at a discrimination result, adopting a hierarchical processing strategy of combining variational mode decomposition and empirical wavelet transform for different types of mode components to realize high-quality signal reconstruction. Compared with the prior art, the method has the advantages that the signal decomposition quality is remarkably improved, the multi-feature fusion discrimination model is excellent in performance when the boundary fuzzy region is processed, the problem of discontinuity of a traditional hard threshold method at the feature boundary is solved, the signal-to-noise ratio is remarkably improved, the root-mean-square error is greatly reduced, and the noise reduction effect is remarkably improved.

The invention belongs to the technical field of RSW pumps of CANDU6 nuclear power stations, aims to solve the problem that modal aliasing and false components exist in EMD decomposition in traditional HHT, and discloses a time-frequency analysis method and system for non-stationary vibration signals of an RSW pump of a CANDU6 nuclear power station, and the method comprises the steps: obtaining a rotor vibration signal, carrying out VMD decomposition on the rotor vibration signal, obtaining K modal components, and carrying out VMD decomposition on the K modal components; performing HT transformation on the K modal components, obtaining a Hilbert spectrum of an original signal, finally obtaining the original signal, and performing diagnosis on a rotor fault according to the change of a characteristic frequency in the Hilbert spectrum; the system comprises a vibration signal acquisition module, a signal decomposition and center frequency evaluation module, a modal component and center frequency solving module, a constraint problem conversion module, a modal component calculation module, a Hilbert spectrum acquisition module, an analysis signal construction module and an original signal restoration module. According to the method, the VMD method with excellent decomposition performance is combined with the HT, and the fault feature recognition capability of a traditional HHT method in processing rotor non-stationary vibration signals is optimized.

The invention relates to the technical field of inertial sensor state data analysis, and discloses a frequency domain processing-based inertial sensor data analysis method, which realizes frequency domain analysis on actual error composition of an inertial sensor, thereby improving the accuracy of effective system performance analysis of the inertial sensor. The method specifically comprises the following steps: S1, acquiring an original output signal by adopting an MEMS inertial measurement unit of a sensor; s2, decomposing a time domain signal in the original output signal into a solid mode equation and a residual term through a mode decomposition algorithm; extracting a solid-state model equation by using a screening algorithm through empirical mode decomposition, and updating a residual error; selecting a center frequency by using signal characteristics and priori knowledge through variational mode decomposition, extracting a solid-state mode equation at the center frequency, and modeling as a function about amplitude phase frequency; s3, performing Hilbert-Huang transform on the components of the solid mode equation to generate a Hilbert spectrum in time-frequency domain joint distribution; and S4, constructing a sensor performance evaluation model based on the Hilbert spectrum.

The present invention provides a multi-channel annular array imaging method for pile integrity detection, comprising: a plurality of acceleration sensors are evenly distributed in a ring shape along the circumference and pasted on the side of the pile, and the acceleration sensors are connected to a multi-channel array receiving device based on FPGA; a reflected time domain signal is obtained by repeatedly striking the top of the pile foundation with a nylon hammer, and a time domain signal with high repeatability is collected; a multi-channel signal coupling algorithm is used to overlap the time domain information in the same phase to obtain a time domain enhancement signal, so as to realize one-dimensional damage location detection; a HHT sub-modal processing algorithm is used to perform HHT on the time domain signal with high repeatability to obtain a three-dimensional Hilbert spectrum of time-frequency-energy, and a ring array imaging algorithm is applied to obtain a three-dimensional imaging diagram of the pile, and the type and position of the pile body defects are quantitatively displayed through the three-dimensional imaging diagram. The present invention solves the technical problem that only one-dimensional time domain curves and frequency response curves can be obtained at present, resulting in low defect location accuracy.

The invention relates to the technical field of ultrasonic road surface recognition, in particular to a road surface pothole area recognition method for road repairing and flattening. The method comprises the following steps: acquiring an ultrasonic echo signal sequence; after the ultrasonic echo sequence is equally divided, the local scattering degree is determined based on the equally divided waveform entropy difference; the waveform entropy of the ultrasonic echo signal sequence forms an entropy sequence, after coarse graining is conducted on the entropy sequence, the optimal time delay and the optimal embedding dimension are obtained, and then a state vector is determined; constructing a recursive matrix based on the similarity of the state vectors, and determining a first surface feature based on the feature of the recursive matrix and the local scattering degree; a second surface feature based on an entropy value of frequency energy in a Hilbert marginal spectrum converted based on an ultrasonic echo signal sequence and a Hilbert spectrum time point amplitude; complexity is determined based on the surface features, the complexity serves as a weight to adjust an amplitude threshold value, and then the signal is enhanced; and identifying the pothole area by comparing the enhanced signal with the safety signal. According to the invention, the detection precision of the pothole area is improved.

The invention provides a fault detection method and device for an AC / DC hybrid micro-grid. The implementation scheme of the fault detection method for the AC / DC hybrid micro-grid is as follows: acquiring a current signal of the AC / DC hybrid micro-grid; performing Hilbert-Huang transform on the current signal to obtain a Hilbert spectrogram; and obtaining a fault detection result of the AC / DC hybrid microgrid based on the Hilbert spectrogram.

The application discloses a kind of PCCP pipeline abnormal vibration high frequency automatic identification method, system and equipment, belong to concrete pipeline vibration automatic monitoring field. Including: the modal decomposition is carried out to the broken wire vibration data monitored, and Hilbert spectrum calculation is carried out, obtain time-frequency spectrum;Calculate the vibration energy curve after each filtering with time variation;Curve is normalized, and the one-dimensional vibration energy data after normalization is dimension, form filter threshold-time-filtered energy matrix;Based on energy-time curve before filtering selects key analysis time period, and obtains the corresponding filter threshold-time-filtered energy matrix sub-matrix;To the row (or column) of each filter frequency threshold value corresponding to sub-matrix selects upper quartile and handles, form filter frequency threshold-normalized filtered energy mean curve;According to the maximum on the curve, obtain the highest frequency of PCCP pipeline abnormal vibration.The application can realize automatic identification.

The application discloses a GIS interference source positioning method based on VFTO analysis, electronic equipment and a storage medium, and is used for a WAPI device of a GIS system. The method comprises the following steps: extracting VFTO waveform data in a wireless local area network security protocol AP of the GIS system; performing noise reduction processing on the VFTO waveform data, and decomposing the VFTO waveform data after noise reduction into a plurality of intrinsic mode functions of different frequency components; performing Hilbert transformation on each intrinsic mode function, and constructing a Hilbert spectrum; and positioning an interference source based on the propagation characteristics of the interference signal and the physical layout of the system through the Hilbert spectrum. The VFTO generated by the WAPI device is analyzed and processed, and the interference source is determined, so that the complex industrial environment of strong electromagnetic interference can be coped with, and the stability and reliability of the WAPI wireless environment are ensured.

The application discloses an oscillation wave detection-based extra-high voltage power transformer winding fault simulation device and diagnosis method. Firstly, the test transformer winding is tested, the transformer winding connection point is connected with a high-frequency high-voltage switch and a high-frequency high-voltage direct current power supply, the high-frequency high-voltage switch is periodically actuated, thereby exciting the extra-high voltage transformer body structure vibration; according to the test data, an ITD transformation is adopted to establish a signal model X t ; then, an empirical mode decomposition (EMD) method is adopted to decompose the signal into a sum of a plurality of mutually orthogonal intrinsic mode function (IMF) components, a Hilbert transformation is performed on each IMF component to obtain an instantaneous frequency and an instantaneous amplitude, thereby obtaining a Hilbert spectrum of the signal to perform decomposition and reconstruction, improve the signal-to-noise ratio, calculate I(X,Y) to judge and remove most of the redundant noise and interference signals; further reconstruct the required signal, after the EMD decomposition of the non-steady-state signal, the mutual information is utilized to screen the obtained IMF components of each order, and a new x(t) is obtained. Finally, the winding vibration signal after the simulation deformation fault is solved through a signal sequence sample entropy calculation formula, and the winding deformation fault severity is judged, and the transformer fault degree is identified.

The application provides a flexible direct current power transmission system fault detection method, comprising the following steps: obtaining an original current signal; setting a search range for a parameter of a variational mode decomposition, thereby obtaining an original setting parameter; decomposing the original current signal into mode functions with different center frequencies; selecting a value corresponding to a minimum relative entropy as an optimal parameter; performing a final variational mode decomposition on the original current signal by using the optimal parameter, thereby obtaining final mode functions in an ordered arrangement; selecting a second final mode function from the final mode functions; performing a Hilbert-Huang transform on the second final mode function, thereby obtaining a Hilbert spectrum of the second final mode function; performing a synchronous compression transform on the Hilbert spectrum, thereby obtaining an instantaneous energy density of the Hilbert spectrum; and comparing an amplitude of the instantaneous energy density with a steady-state threshold, thereby realizing fault detection. The application can realize accurate and rapid fault detection on a flexible direct current power transmission system.

The invention discloses an extra-high voltage power transformer winding fault simulation device and diagnosis method based on oscillatory wave detection, and the method comprises the steps: firstly testing a test transformer winding, connecting a connection point of the transformer winding with a high-frequency high-voltage switch and a high-frequency high-voltage DC power supply, and enabling the high-frequency high-voltage switch to carry out the periodic operation; therefore, the extra-high-voltage transformer body structure is excited to vibrate; establishing a signal model Xt by adopting ITD transformation according to the test data; the method comprises the following steps of: firstly, acquiring a signal, then decomposing the signal into the sum of a plurality of mutually orthogonal intrinsic mode function (IMF) components by adopting an empirical mode decomposition (EMD) method, and then performing Hilbert transformation on each IMF component to obtain an instantaneous frequency and an instantaneous amplitude, thereby obtaining a Hilbert spectrum of the signal to perform decomposition and reconstruction, improving a signal-to-noise ratio and solving I (X, Y) to judge and remove most of redundant noise and interference signals; and further reconstructing the required signal, and screening the obtained IMF components of each order by using mutual information after the unsteady-state signal is decomposed by EMD to obtain a new x (t). And finally, solving a winding vibration signal after the deformation fault is simulated through a solved signal sequence sample entropy calculation formula, judging whether the deformation fault of the winding is serious according to the winding vibration signal, and identifying the fault degree of the transformer.

This application relates to the technical field of vehicle control, and in particular to a method, apparatus, device, and medium for testing time synchronization networks. The method includes: acquiring the state equation and system parameters of a preset nonlinear dynamic disturbance; injecting a phase perturbation sequence into the device under test (DUT) to capture the synchronization error response sequence output by the DUT under perturbation excitation; performing variational mode decomposition on the synchronization error response sequence to obtain K eigenmode functions with finite bandwidth; performing a Hilbert transform on each eigenmode function to obtain the Hilbert spectrum of the tested spatiotemporal synchronization loop; constructing a parameterized reduced-order model of the system based on the eigenmode functions, predicting the stability domain boundary and instability bifurcation point of the system in the parameter space, and generating a synchronization stability boundary map to verify the bidirectional synchronization capability of the DUT based on the synchronization stability boundary map. This method can improve testing efficiency by adding active excitation to deeply evaluate the dynamic performance of the device and thus performing automatic time testing.

The application discloses a power transmission line robot automatic icing identification and deicing method, the method comprises the following steps: obtaining ice layer dielectric properties and meteorological data, calculating ice energy level index; after successful connection, based on quantum heuristic algorithm, point cloud matching is performed on the ice layer and an ideal power transmission line spiral stranded model, and a topological singular point density value is calculated to dynamically generate an adaptive fractal rolling path; in the moving rolling process, the Hausdorff dimension change of ice cracks and the nonlinear distortion component of the roller motor are fused, and a brittle-plastic phase transition critical factor is calculated; for the ice crushing removal effect, the sound force coupling dissipation rate is calculated through the Hilbert spectrum of the acoustic emission signal, and a model predictive control framework is constructed to realize efficient centrifugal throwing of ice chips; after the operation section is finished, a global residual risk potential field is generated based on the sound force coupling dissipation rate, and secondary fine polishing or intelligent unhooking return is performed according to the cleanliness convergence value. The application improves the deicing efficiency while ensuring the safety of the power transmission line body.

The application discloses a kind of underwater vehicle propeller shaft frequency and blade frequency variational holographic spectrum identification method and device, variational holographic spectrum method utilizes the thought of holographic representation, realizes the robust high-dimensional holographic representation of nonlinear non-stationary signal, obtains variational holographic spectrum diagram, by feature extraction to variational holographic spectrum diagram, realize the identification and estimation of underwater vehicle propeller shaft frequency, blade frequency and blade number;The method introduces variational mode decomposition, effectively solves the mode aliasing problem of EMD in holographic hilbert spectrum analysis method;According to the characteristics that propeller modulation signal in underwater vehicle radiation noise is low frequency slow change, segmented fourier transform is introduced, the problem that there is large error when hilbert spectrum analysis method is processed low frequency slow change signal is solved.The method solves the problem of insufficient representation when the commonly used radiation noise analysis method LOFAR analysis, DEMON analysis and various improved algorithms are analyzed in nonlinear non-stationary signal.

The application relates to the technical field of vehicle-mounted control, in particular to a time synchronization network test method and device, equipment and a medium, the method comprising the following steps: acquiring a state equation and system parameters of a preset nonlinear dynamic disturbance; injecting a phase perturbation sequence into a measured device; capturing a synchronization error response sequence output by the measured device under the perturbation excitation; performing variational modal decomposition on the synchronization error response sequence to obtain K intrinsic modal functions with limited bandwidth; performing Hilbert transformation on each intrinsic modal function to obtain a Hilbert spectrum of the measured time-space synchronization loop; constructing a parameterized reduced-order model of the system based on the intrinsic modal functions, predicting a stable domain boundary and an unstable bifurcation point of the system in a parameter space, and generating a synchronization stability boundary atlas to verify bidirectional synchronization capability of the measured device according to the synchronization stability boundary atlas. The dynamic performance of the device can be deeply evaluated by increasing active excitation, so that automatic time testing is performed, and the testing efficiency is improved.

This application relates to the field of ultrasonic pavement recognition technology, specifically to a method for identifying pothole areas in road repair and leveling. The method includes: acquiring an ultrasonic echo signal sequence; dividing the ultrasonic echo sequence equally and determining the degree of local dispersion based on the waveform entropy difference of the division; constructing an entropy sequence from the waveform entropy of the ultrasonic echo signal sequence, coarsening it, obtaining the optimal time delay and optimal embedding dimension, and then determining the state vector; constructing a recursive matrix based on the similarity of the state vectors, and determining the first surface feature based on its features and the degree of local dispersion; determining the second surface feature based on the entropy value of the frequency energy in the Hilbert marginal spectrum transformed from the ultrasonic echo signal sequence and the amplitude at time points in the Hilbert spectrum; determining the complexity based on the surface features, using it as a weight to adjust the amplitude threshold, thereby enhancing the signal; and identifying pothole areas by comparing the enhanced signal with a safe signal. This application improves the detection accuracy of pothole areas.

The present invention provides a method for identifying structural fatigue cracks based on the collaboration of VMD-HHT, comprising the following steps: applying a harmonic excitation at the end position of the structure to be identified, collecting the acceleration response data at different positions of the structure, and using wavelet packet transform to perform decomposition, noise suppression and reconstruction processing in sequence to obtain a reconstructed acceleration signal with noise reduction; performing variational mode decomposition on the reconstructed acceleration signal to obtain intrinsic mode functions; performing Hilbert-Huang transform on each intrinsic mode function respectively to obtain a Hilbert spectrum set; when multi-order secondary high-frequency signals appear in the Hilbert spectrum set, a fatigue crack occurs at the response data acquisition position of the structure, and the greater the instantaneous frequency amplitude of the multi-order secondary high-frequency signals, the greater the fatigue crack. This method combines the advantages of VMD and HHT, overcomes the problem of false intrinsic mode function information, improves the ability to decouple different frequency components in narrow-band signals, and can effectively distinguish frequency differences.

This invention relates to the field of reservoir prediction technology, and particularly to a method and apparatus for establishing a low-frequency model based on Hilbert-Huang transform for seismic inversion. The method includes: partially stacking CRP gathers that reflect AVO characteristics from CRP gather data; decomposing the partially stacked data volume into several intrinsic mode function (IMF) data volumes; performing a Hilbert-Huang transform on the IMF data volumes to obtain a Hilbert spectrum; performing low-frequency energy enhancement on the partially stacked data volumes based on the Hilbert spectrum to obtain a low-frequency enhanced data volume; and establishing a low-frequency model using a support vector machine method combined with well-seismic analysis, under the constraints of a subdivided layer interpretation structural framework. This invention expands the effective signal at the low-frequency end without changing the relative relationships between the partially stacked data volumes, improving the vertical resolution of seismic data while maintaining its lateral resolution, thus providing effective technical support for unconventional reservoir oil and gas exploration and development.

The invention relates to the technical field of fan blade fault detection, and particularly discloses a fan blade fault prediction method and system based on EEMD-HHT and CBCNN, and the method comprises the steps: obtaining an original vibration signal, obtaining an effective IMF based on EEMD technology processing, and carrying out the HHT transformation of the effective IMF to obtain a Hilbert spectrum; the method comprises the following steps: pre-constructing a CNN network and adding a CBAM attention module to form a CBCNN fault prediction model; dividing all Hilbert spectrums to obtain a verification set, a test set and a training set for training a CBCNN fault prediction model, and performing model training optimization processing until a final CBCNN fault prediction model is obtained; and obtaining a vibration signal used for prediction, inputting the vibration signal to the final CBCNN fault prediction model, and outputting a prediction result. According to the method, integrated empirical mode decomposition (EEMD) is adopted to improve HHT, a CBMA attention module is introduced into a CNN to enhance a CNN model, training is performed based on a training set containing a detection result, and fault type prediction is realized.

The invention provides a PVDF (polyvinylidene fluoride)-based blasting monitoring system and a PVDF-based blasting monitoring method, which can realize accurate capture and time sequence analysis of each blasting event so as to accurately evaluate the blasting quality and provide a reliable basis for safe danger elimination, and comprises a vibration sensing module, a signal acquisition terminal and a control module, the signal acquisition terminal processes the electric signal, converts the electric signal into a digital vibration signal and transmits the digital vibration signal; the data monitoring terminal processes the received digital vibration signal by using an HHT algorithm to generate a Hilbert spectrum representing the time-frequency characteristic of the vibration signal; based on the Hilbert spectrum, identifying an effective blasting signal corresponding to the single blasting event; according to the time sequence of the effective blasting signals, an actual blasting time sequence is reconstructed; and the actual blasting time sequence is compared with a preset blasting scheme time sequence, and the blasting operation effect is judged.

The present application relates to the technical field of inertial sensor state data analysis, and discloses a data analysis method for an inertial sensor based on frequency domain processing, which realizes frequency domain analysis on the actual error composition of the inertial sensor, thereby improving the accuracy of effective coefficient performance analysis of the inertial sensor. Specifically, S1, a MEMS inertial measurement unit of the sensor is used to collect original output signals; S2, a time domain signal in the original output signal is decomposed into a solid mode equation and a residual term through a modal decomposition algorithm; the solid mode equation is extracted through empirical mode decomposition using a screening algorithm, and the residual is updated; the center frequency is selected using signal characteristics and prior knowledge through variational modal decomposition, the solid mode equation at the center frequency is extracted, and modeling is performed as a function of amplitude, phase and frequency; S3, Hilbert spectrum of time-frequency domain joint distribution is generated by performing Hilbert-Huang transformation on the solid mode equation component; and S4, a sensor performance evaluation model is constructed based on the Hilbert spectrum.