450 results about "Variational mode decomposition" patented technology

The invention discloses a short-term load prediction method of a variational mode decomposition and deep belief network. The method comprises the following steps of 1) using a variational mode decomposition method to decompose original historical load data into a series of mode functions with different characteristics; 2) using an approximate entropy to calculate each modal function complexity, merging the modal functions whose approximate entropy values are similar into a new component, and carrying out characteristic analysis on each component; 3) in order to calculating correlation of an influence factor and an output variable, carrying out normalization processing on data; 4) combining a period characteristic of a load, and using a mutual information theory to select an input variable set from aspects of a historical load, a meteorology factor, a date type and the like; and 5) constructing the short-term load prediction method based on the deep belief network (DBN), and verifying method validity through a load prediction scene before 24h. By using the method, short-term load prediction precision is effectively increased and an electric power system load prediction problem can be well solved.

A hydroelectric generating set vibration signal noise reduction method based on mode autocorrelation analysis of SVD and VMD comprises the steps of constructing a Hankel matrix of a set vibration signal and performing singular value decomposition (SVD), selecting an effective singular value based on a mean value filtering strategy for reconstructing the signal, and realizing pre-filtering; performing decomposition through variational mode decomposition (VMD) for obtaining a series of mode functions, calculating an autocorrelation function of each mode component, selecting effective mode components according to an energy set of the autocorrelation function, and obtaining a signal after noise reduction through adding all effective mode components. According to the hydroelectric generating set vibration signal noise reduction method provided by the invention, a noise reduction experiment is carried out through simulation analysis and actual measurement of a vibration signal; and a result represents a fact that the method has relatively high noise reduction performance and can effectively improve hydroelectric generating set vibration signal analysis precision.

The invention discloses a GIS (Gas Insulated Switchgear) mechanical oscillation signal time frequency analysis method based on VMD (Variational Mode Decomposition) adaptive morphology. The GIS mechanical oscillation signal time frequency analysis method based on VMD adaptive morphology includes the steps: simulating different types of mechanical faults of GIS equipment; detecting oscillation signals of the GIS equipment in the normal condition and the simulation condition for many times; utilizing VMD to realize time frequency analysis of the oscillation signals, and finding out the change ofthe oscillation signal amplitude of the GIS device, following frequency distribution; and by integrating with Hilbert analysis, obtaining the characteristic criteria of the faults, by simulating different types of mechanical faults, establishing a GIS mechanical fault diagnosis database to realize time frequency analysis of the oscillation signals of the GIS equipment. The GIS mechanical oscillation signal time frequency analysis method based on VMD adaptive morphology performs time frequency analysis on the mechanical oscillation signals through the VMD algorithm, and can effectively processthe GIS oscillation signals so as to establish the GIS mechanical fault diagnosis database to provide theoretical basis for realizing field live detection of the GIS mechanical faults.

The invention discloses a rolling bearing fault diagnosis method based on variational mode decomposition, which is used to analyze a rolling bearing fault based on a vibration signal. First, fault information enhancement processing is performed on a signal collected by a sensor using a predictive filtering method; then, variational mode decomposition is performed on the filtered signal to get four modes; next, a mode most relevant to a rolling bearing fault is selected according to a fault information index; and finally, an envelope autocorrelation spectrum analysis of the filtered signal is made, and fault characteristic frequency matching is performed to get fault information.

The invention relates to a pipeline multi-point leakage positioning method based on improved VMD. The method comprises the following steps of collecting an original leakage signal of a pipeline; performing overall local area mean value decomposition on the original leakage signal to obtain a plurality of PF components; calculating a correlation coefficient of each PF component, screening out the required PF component according to the correlation coefficient, performing signal reconstruction according to the screened PF component, and determining K value of variational mode decomposition; performing variational mode decomposition on the reconstructed signal to obtain a plurality of IMF components, calculating a multi-scale entropy value of each IMF component, and screening the IMF components according to the multi-scale entropy value of each IMF component; and performing signal reconstruction on the screened IMF component, and performing cross-correlation positioning calculation on eachleakage signal subjected to blind source separation to complete pipeline leakage positioning. According to the method, the leakage signal of the pipeline can be effectively extracted, the influence of low relevant components and noise in the original leakage signal is eliminated, and the final positioning result is more accurate.

The invention discloses an intelligent power grid false data injection attack detection method which combines a variational mode decomposition technology and a machine learning technology. The detection method is added behind a power system state estimation part, and comprises the following steps: firstly, decomposing a system state time sequence into a set of a plurality of sub-sequences with different center frequencies by using a variational mode decomposition technology; secondly, in order to compress redundant data to facilitate model training, expressing data segment characteristics by using characteristic quantities based on statistical indexes; and finally, constructing a false data injection attack detector under the smart power grid by taking the calculated feature set as a training set of an online sequential extreme learning machine. In the process, the credible authenticated data can be used as an extended training set to continuously optimize the performance of the detector with the online learning capability. The invention aims to establish an efficient and accurate false data injection attack detection method for the smart power grid to ensure safe control and stable operation of a power system.

The invention discloses a VMD and CNN-based cable early fault identification and classification method. The method comprises the following steps: step 1, obtaining a to-be-tested analog signal; step 2, selecting a bandwidth limiting factor alpha, a noise tolerance tau and a mode decomposition number K as parameters, and setting parameter values; step 3, performing variational mode decomposition onvarious analog signals, obtaining each mode and the center frequency thereof, and realizing frequency band division; step 4, extracting decomposition modal features and constructing feature vectors;step 5, inputting various signal feature vectors into the convolutional neural network, carrying out parameter modulation training and obtaining a classification result. By using the method, early faults and over-current disturbance of the cable can be accurately distinguished, cable maintenance is completed in time before the early faults become permanent faults, and stable operation of a power grid is maintained.

The invention discloses a VMD-GRU-based short-term wind speed prediction method. The method is characterized is characterized in that it comprises, acquiring data of wind speeds at the current momentand n moments closest to the current moment to form a time sequence; and preprocessing the time sequence to obtain a plurality of sub-sequences and residual components, respectively inputting each sub-sequence and each residual component into respective corresponding trained GRU models, outputting predicted values by the trained GRU models, and performing post-processing on all the predicted values to obtain future wind speed prediction data at the next moment. According to the method, non-stationary wind speed data is decomposed into sub-sequences and residual components with different frequencies by adopting a variational mode decomposition method, the stability of the sub-sequences and the residual components is good, and better prediction is facilitated; the method has good predictionprecision for the wind speed with strong volatility, randomness and uncertainty, and the operation state of the wind power generation device can be adjusted more reasonably.

The invention relates to a laser radar echo signal de-noising method based on variational mode decomposition. The method comprises the steps of according to a laser radar equation, simulating a laserradar echo signal; selecting an optimal decomposition layer number K in variational mode composition by means of detrended fluctuation analysis; performing variational mode decomposition on the laserradar echo signal; then calculating similarity between the laser radar echo signal and each mode probability density function after decomposition through Bhattacharyya distance, differentiating a correlation mode and an uncorrelated mode; then processing the uncorrelated mode by means of a moving average mode, and extracting a related signal; and finally reconstructing the correlation mode and theprocessed uncorrelated mode for realizing de-noising of the laser radar echo signal. The laser radar echo signal de-noising method can effectively filter the noise in the laser radar echo signal andimproves effective detecting distance and precision of the laser radar.

The invention discloses a short-term photovoltaic power prediction method based on VMD-IPSO-GRU, and belongs to the technical field of photovoltaic power generation and grid connection. Firstly, a historical photovoltaic power time sequence is decomposed into sub-sequences with different frequencies through variational mode decomposition, geographic information and component parameters contained in photovoltaic sequence data are fully mined, and signals and noise of original data are separated; secondly, main meteorological factors influencing photovoltaic output are determined through Spearman and Pearson correlation coefficients; and finally, gating cycle unit network models are established for the sub-sequences decomposed by the VMD respectively, and the GRU nerve is optimized through an improved particle swarm algorithm and an adaptive moment estimation algorithm, thereby improving the network convergence rate and the data fitting effect, accurately and efficiently finishing short-term photovoltaic power prediction, and avoiding errors caused by manual parameter adjustment.

The invention belongs to the field of power systems, and relates to a PSO-based on improved variational mode decomposition. The LSSVM short-term load prediction method comprises the following steps: S1, selecting a decomposition effect evaluation index; S2, setting an SMD decomposition upper limit; S3, optimizing the VMD parameters by using a particle swarm optimization algorithm, performing VMD decomposition, and finally obtaining a period corresponding to the center frequency of the modal component; S4, combining the modal components to obtain a combined component; S5, solving mutual information between the sequences of the influence factor data and the combination components and the predicted daily load sequence, and obtaining an influence factor input variable set according to a threshold requirement; S6, substituting the selected influence factor input variable into the PSO- LSSVM model. According to the method, the utilization efficiency of influence factor data is improved, andan optimized mode decomposition result is obtained; By quantifying the correlation between the internal structure components of the influence factors and the loads, effective influence factor variables are accurately selected, the number of the influence factors is increased, and the prediction precision is improved.

The invention discloses a bearing fault diagnosis method based on feature enhancement, which can effectively and rapidly extract bearing fault vibration signal shock characteristics while the signal data amount is reduced. Firstly, the bearing fault vibration signals are decomposed by variational mode decomposition (VMD), a kurtosis value and a component with the maximum cross-correlation functionwith the original signal are selected as the optimal components which have better block sparse characteristics. On the basis of the traditional online dictionary learning constraint model, l2, 1 normconstraints of a sparse coefficient are added. Under a new constraint model, sparse representation and dictionary learning are carried out alternatively, inter-block sparse characteristics of the newconstraint can be matched with block sparse characteristics of vibration signals in a sparse representation process, the redundant component in the signals is further removed, l2, 1 norm constraintsare added during the dictionary learning process at the same time, and an experimental result shows that dictionary atoms acquired from the dictionary learning process with new constraints added are more robust against noise interference. The dictionary obtained based on learning and the sparse coefficient are subjected to signal reconstruction, the signal shock characteristics with the redundantcomponents enhanced such as noise in the signals can be removed, and the fault information of the signals is further extracted to complete fault diagnosis.

The invention discloses a variable frequency scroll compressor fault diagnosis method based on improved VMD and SVM, and the method carries out the processing and analysis of a vibration signal of a variable frequency scroll compressor from three aspects: signal processing, feature extraction and classification recognition, and comprises the steps: firstly carrying out the vibration test of the variable frequency scroll compressor; acquiring vibration acceleration signals under different states of normality, scroll plate faults, bearing faults and crankshaft faults of the variable-frequency scroll compressor, and obtaining data samples of different fault types; taking the envelope entropy-correlation index as a fitness function, adopting a sparrow search algorithm (SSA) to optimize a variational mode decomposition (VMD) algorithm to process the vibration signal of the variable frequency scroll compressor, and obtaining intrinsic mode functions of different scales; and calculating multi-scale permutation entropies of different intrinsic mode functions to form a feature vector, inputting the feature vector into a classifier established based on a support vector machine (SVM) for training and predictive classification, and judging the fault type of the variable-frequency scroll compressor.

The invention discloses a wind speed section prediction method and system based on an artificial neural network, and the method and system are used for the short-time wind speed prediction of a wind field. The method comprises the steps: decomposing an initial complex time sequence into a plurality of time sequences in a simple structure through VMD (variational mode decomposition); carrying out the feature selection through GSO (Gram-Schmidt orthogonal); taking a processed wind speed sequence as the input of the ANN (artificial neural network), wherein the output of the ANN is the upper and lower bounds of wind speed at a future moment; finally training the weight and bias of the ANN through an MOGSA (Multi objective gravitational search algorithm), taking two contradictory indexes (coverage rate and section width) as an optimization target, and obtaining an optimal scheme set. The wind speed section predicted through the method has a higher coverage rate for the actual wind speed section, and is small in section width. The above combined model enables the prediction accuracy to be improved to a very high level.

The invention discloses a method for removing interference on radar imaging caused by object micro-motion. The method of the invention comprises steps of emitting M identical N-dimension discretized linear frequency modulation signals by a radar, obtaining an M-row and N-column two-dimensional distance-slow time domain echo data after an emitted signal and a received echo signal go through matched filtering, 2) performing compound variational mode decomposition on the distance-slow-time domain echo data of each column to obtain q modular functions, calculating energy of each modular function which is decomposed out, solving a ratio of energy of each modular function to energy of all modular functions, performing summation synthesis on the modular functions, the ratios of which are higher than a threshold to obtain new M-row N-column two-dimensional distance-slow-time-domain echo data and 3) performing windowing processing on the echo data of the M-row N-column two-dimensional distance-slow-time-domain which is obtained from the step 2 and then performing Fourier transformation along an azimuth to obtain an inverse synthesis aperture radar imaging.

The invention discloses a method for predicting the remaining service life of a rolling bearing integrated with the KELM (Kernel Extreme Learning Machine), and belongs to the technical field of the bearing service life prediction. The method is used to solve the problem that the prediction of the remaining service life of the rolling bearing has difficulty in prediction and low prediction accuracy. The method firstly extracts features of a vibration signal based on the variational mode decomposition, introduces a new similarity dimension reduction method for features dimension reduction, and further extracts the features-CEF (Cyclic Enhancement Features) with strong monotonicity, similarity, and stability. Multiple KELM models are constructed through that the CEF extracted by the multiplebearings is used as the input of the KELM, the ratio of the current service life to the whole life, p, that is, the life percentage is used as the output. A prediction model integrated with KELM is constructed by combining the random forest to obtain a current prediction result p value. The CEF of the test bearing is input into the prediction model, the current p value is predicted, and the secondorder exponential smoothing method is used for fitting to predict the RUL of the bearing. The experimental verification shows that the proposed prediction method has higher prediction accuracy than other literatures.

The invention discloses an antifriction bearing VMD, spectral kurtosis and smooth iteration envelope analysis method. According to the method, firstly an original signal is decomposed by using a variational mode decomposition method; then the noise component and the trend term in the decomposition result are eliminated by using rearrangement and substitution operation of the data; then the signal after the first time of filtering is analyzed by using a spectral kurtosis method so that the center frequency and bandwidth of an optimal filter are obtained; then the second time of filtering is performed on the signal after the first time of filtering by using the filter; then envelope analysis is performed on the signal after the second time of filtering by using a smooth iteration envelope analysis method; and finally the fault type of an antifriction bearing is determined according to the envelope spectrum. The method is suitable for processing complex antifriction bearing fault signals and can accurately judge the fault type of the antifriction bearing so that the method has great anti-noise performance and robustness and is convenient for engineering application.

The invention discloses a seismic fluid prediction method based on VMD and TK energy operators. The method comprises the following steps: S1, performing variational mode decomposition on a seismic signal to obtain a set of modal components that reflect a natural mode of vibration inside the seismic signal and have an instantaneous frequency with real physical meaning; S2, getting TK energy spectrums of each modal component by using the TK energy operators to obtain more accurate energy distribution on a signal time frequency; and S3, performing effective and accurate seismic fluid prediction by searching fluid characteristics of ''enhanced low frequency energy and attenuated high frequency energy'' of oil and gas contained in the reservoir. The seismic fluid prediction method based on VMDand TK energy operators combines the variational mode decomposition with the TK energy operators, which has stronger identification for the fluid-containing characteristics of the reservoir, and can effectively improve the accuracy of seismic fluid prediction.

The invention discloses a shock echo signal analysis method based on variational mode decomposition, which comprises the following steps: 1) selecting measurement parameters and working parameters ofequipment according to requirements of field measurement environment, and collecting shock echo signals; 2) setting parameters in variational modal decomposition according to the shock echo signals; 3) decomposing the collected shock echo signals into several eigen modulus functions by a variational modal decomposition method to solve the variational problem; 4) using Hilbert transform to obtain the Hilbert time-frequency spectrum of the eigen modulus functions; 5) integrating the Hilbert time-frequency spectrum of different frequencies in the time domain to obtain the final marginal spectrum.The method identifies the defects of the concrete component by using the marginal spectrum of the shock echo signals, and can suppress the noise interference under strong noise, and achieves higher resolution compared with the traditional Fourier transform.

The invention provides a bearing fault classification method and system based on a deep learning network, and the method comprises the steps: setting a sampling frequency, and collecting the vibrationsignal data of a bearing under different working conditions; segmenting the obtained vibration signal data to construct a plurality of samples; decomposing the vibration signal data of each sample toobtain a plurality of modal components so as to realize effective component separation; constructing a deep network with a residual error unit, and determining an appropriate network depth by using arandom search method; inputting the training set into a deep residual network for iterative training and obtaining a classification model; and inputting the test set into the classification model toobtain a fault classification result. According to the classification method provided by the invention, variational mode decomposition and a deep residual network are combined; the problems that noiseinterference exists in input data, cross aliasing exists in effective components, network deepening causes identification gradient disappearance, and performance degradation causes poor classification effect are solved, fault feature extraction not affected by rotating speed changes is achieved, and the fault classification accuracy is improved.

The invention provides a noise reduction method based on variational mode decomposition and a minimum mean square error adaptive filter. This method combines the advantages of variational modal decomposition and least mean square adaptive filter, the noisy and noisy signals are decomposed into K eigenmode components by using variational mode decomposition algorithm, and then each eigenmode component is filtered by least mean square adaptive filter. Finally, the denoised signal is reconstructed from the K filtering results. This method not only has good function of signal-noise separation, butalso can effectively suppress the phenomenon of mode aliasing. The adaptive filter can automatically adjust the filter parameters iteratively to meet the requirements of certain criteria, so as to achieve optimal filtering, has and have the function of self-tuning and tracking.

The invention discloses a method for predicting the remaining service life of a lithium ion battery based on VMD-HGWO-SVR. Prediction of the remaining service life of the lithium ion battery is an important part of battery health management. The method comprises the following specific steps: firstly, carrying out multi-scale decomposition on lithium battery capacity degradation data by using a variational mode decomposition method, setting a proper threshold value according to correlation coefficient analysis, and reconstructing a mode function meeting conditions to obtain battery capacity data after capacity regeneration and noise fluctuation are eliminated; then, training an SVR model based on the preprocessed battery capacity data, and optimizing hyper-parameters of the SVR by adoptingan improved grey wolf optimization algorithm HGWO; and finally, predicting the remaining service life of the lithium battery by using the trained VMD-HGWO-SVR model. According to the method, the influence of capacity regeneration and noise fluctuation in the lithium battery capacity data on the prediction precision of the residual life of the lithium battery is solved, the grey wolf optimization algorithm is improved in three aspects to prevent the prediction model from falling into a local optimal solution during training, and the proposed method is stable in prediction performance and more accurate in prediction result.

The invention discloses an optimal power flow calculation method for a multi-period electricity-gas interconnection system based on wind speed prediction and is applicable to the field of power system optimization control. According to the method, firstly, a wind speed prediction method based on variational mode decomposition and gaussian process regression is proposed, and accordingly a probability distribution curve for currently predicating the wind speed is obtained; an electricity-gas interconnection system multi-period optimal power flow model is established, the minimum total operation cost serves as the target, and the model relates to operation constraint of an power system and a natural gas system; punishment cost and standby cost are adopted to describe influences caused by wind power overestimation and wind power underestimation respectively. It is indicated through the embodiment that the power system and the natural gas system restrict each other, comprehensive optimization is beneficial for obtaining of a globally optimal solution, and safety and reliability of the systems are further guaranteed. Besides, the wind power punishment cost and wind power standby cost have great influences on a regulation scheme, a reference is provided for optimized operation of the systems under the background that new energy is introduced, and decision support is provided for scheduling personnel.

The invention discloses an intelligent identification method for prestressed concrete beam grouting compactness. According to the method, variational mode decomposition, finite element simulation, support vector machine and other technologies are used, wherein impact echo vibration signals under different working conditions are simulated through finite elements, a support vector machine training sample is constructed by using variational mode decomposition and hilbert transform, and finally the prestressed concrete beam grouting compactness is quantitatively predicted with the trained support vector machine according to an impact echo detection signal. According to the present invention, with the intelligent identification method, the disadvantages of low efficiency, poor precision and the like during the existing impact echo detection process can be effectively overcome, the influence of the human factors in the detection process can be eliminated, and the detection precision and the working efficiency can be improved to the greatest extent.

The invention relates to an on-load tap-changer fault identification method based on vibration signal feature extraction. The method comprises steps that firstly, a vibration signal of an on-load tap-changer OLTC in operation is acquired by an acceleration sensor, and the vibration signal is preprocessed; the acquired tap-changer vibration signal is subjected to variation mode decomposition, thatis VMD decomposition, false components are further eliminated, and the kurtosis of an effective modal component is calculated as the fault feature quantity of the OLTC; and lastly, tap-changer fault diagnosis is performed according to the tap-changer fault feature quantity. The method is advantaged in that OLTC mechanical fault features can be effectively extracted, and the effect is better than EMD.

The invention provides a power load prediction method based on variational mode decomposition and a gating cycle unit, and belongs to the technical field of power load prediction. The power load prediction method comprises the following steps: S1, acquiring original load data; S2, decomposing the original load data into a modal function by adopting a variational modal decomposition method; S3, performing normalization processing on the modal function, and dividing the modal function into a training set and a test set; S4, performing gating cycle unit network training on the training set, and establishing a prediction model according to a training result; S5, detecting the prediction model by using a loss function until a loss function value is reduced to be less than a threshold value; andS6, substituting the test set into the prediction model and superposing to obtain a power load prediction result, and evaluating the prediction result. The method is accurate in prediction of data with volatility and randomness, can effectively improve the speed of the model, and can well solve the problem of load prediction of a power system.

The invention relates to a magnetic resonance sounding (MRS) signal noise filtering field, in particular to a magnetic resonance sounding signal noise filtering method based on variational mode decomposition, which is mainly used for processing power frequency resonance noise and random white noise in the magnetic resonance sounding signals. A 'three-VMD' decomposition approach is provided to better achieve the efficient removal of noise in the noisy MRS signals. MRS signals collected by a magnetic resonance sounding water detector are subjected to band-pass filtering and Fourier transform todetermine the frequencies and the number of the power frequency resonance interference and the single frequency interference included in the MRS signals, the first, second and third VMD decompositionare employed to respectively remove the Gaussian white noise, most of the power frequency and the power frequency being the closest to the signals in the noisy MRS signals, and finally, target MRS signals are extracted and obtained. The magnetic resonance sounding signal noise filtering method can solve the customary modal aliasing problem after a traditional modal decomposition method is employed, and is high in signal-to-noise ratio and high in adaptability compared with the traditional MRS signal denoising method.

The invention discloses an electric energy quality analysis method based on variational mode decomposition multi-scale permutation entropy. The method comprises the steps of collecting original training data of electric energy quality monitoring points when the electric energy quality monitoring points are disturbed by different types; adopting variational mode decomposition to decompose the components to obtain K IMF components; calculating the multi-scale permutation entropy of each IMF component, and constructing a feature vector of the original training data; selecting R features from thefeature vectors to form an optimized feature vector of the original training data; taking the optimized feature vector of the original training data and the corresponding disturbance type as input data and output data respectively, and training an ELM neural network model to obtain an electric energy quality disturbance classifier; and acquiring optimized feature vectors of voltage signals of to-be-detected power quality monitoring points according to same method; inputting the optimized feature vectors into the power quality disturbance classifier to obtain the disturbance type of the power quality monitoring point to be detected. According to the method, the real-time diagnosis efficiency of the disturbance type of the power quality is greatly improved.

The invention discloses a mine micro-seismic signal identification method based on the features of energy distribution, which belongs to the field of signal analysis and identification. The method includes the following steps: reading a micro-seismic signal x(t) to be identified; carrying out VMD (Variational Mode Decomposition) on x(t) to get K variational modal components arranged in order according to the frequency from high to low; calculating the band energy of each modal component, and extracting the energy percentage of each modal component in the original signal to constitute an energy distribution vector P; calculating the energy distribution X-axis center-of-gravity coefficient cx on the basis of the energy distribution vector P; identifying the mine micro-seismic signal according to an identification threshold T: determining that the signal is a mine coal rock fracture micro-seismic signal if cx>T, and determining that the signal is a blasting vibration signal if cx<=T; and finally, adaptively updating the value of the identification threshold T. Through the method, a coal rock fracture micro-seismic signal and a blasting vibration signal can be distinguished. The method has the characteristics of strong adaptability, high accuracy, and the like.

The invention relates to a desert seismic signal denoising method based on the VMD approximate entropy and a multi-layer perceptron and belongs to the field of geophysical technology. The two-dimensional desert seismic record is subjected to variational mode decomposition to obtain a series of eigenmode components, the approximate entropy of each eigenmode component is calculated, all the eigenmode components are respectively divided into an effective signal dominant component and a noise dominant component, a characteristic quantity is constructed through effective signal correlation, the characteristic quantity is inputted into the multi-layer perceptron for classification, the valid signal portion determined by the multi-layer perceptron classifier is reserved, the noise portion determined by the multi-layer perceptron classifier is removed, the noise dominant component is denoised through combining an autocorrelation coefficient with tzhe multi-layer perceptron, and lastly, desertseismic signal denoising is realized through reconstruction. The desert seismic signal denoising method is advantaged in that noise removal under strong noise and low signal to noise ratio conditionsis achieved, the desert seismic signal denoising method has fast speed, high accuracy and strong anti-interference ability, and denoising of desert seismic signals under low frequency and low SNR canbe realized.