CN115659128B - Signal noise reduction method based on ensemble empirical mode decomposition method and power spectrum - Google Patents
Signal noise reduction method based on ensemble empirical mode decomposition method and power spectrum Download PDFInfo
- Publication number
- CN115659128B CN115659128B CN202211587362.6A CN202211587362A CN115659128B CN 115659128 B CN115659128 B CN 115659128B CN 202211587362 A CN202211587362 A CN 202211587362A CN 115659128 B CN115659128 B CN 115659128B
- Authority
- CN
- China
- Prior art keywords
- power spectrum
- signal
- component
- eigenmode
- intrinsic mode
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
The invention discloses a signal noise reduction method based on a set empirical mode decomposition method and a power spectrum, which comprises the steps of screening an intrinsic mode function after the set empirical mode decomposition to reconstruct a signal, reasonably screening intrinsic modes according to the ratio of the peak part bandwidth of the power spectrum of the intrinsic mode function to the total bandwidth and the ratio of the number of extreme points of the power spectrum to the number of the extreme points of the first intrinsic mode function, and finally achieving the purpose of optimal noise reduction. The method is widely applicable to signal processing of a nonlinear system mixed with white noise and periodic noise, can well distinguish noise from effective signals and accurately eliminate noise interference to obtain pure signals, and has important engineering application value in the fields of bearing fault detection, milling chatter recognition, cutter wear monitoring and the like.
Description
Technical Field
The invention relates to the field of signal processing, in particular to a signal noise reduction method based on a set empirical mode decomposition method and a power spectrum.
Background
The method is characterized in that a group of white noises with normally distributed amplitudes are added in the original method, the noises are mutually counteracted through multiple average operations, signals with different scales are automatically decomposed to a proper scale by utilizing the uniform distribution characteristic of a white noise frequency spectrum, the modal aliasing phenomenon caused by an end point effect can be effectively inhibited without artificial interruption, and the dynamic characteristic of the original signal is preserved.
After the EEMD decomposition is completed, intrinsic Mode Functions (IMFs) of different scales are obtained, and the IMFs need to be screened when signals are reconstructed, however, IMF component signal screening does not have a screening criterion or a quantization standard, and in a plurality of similar components, only artificial subjective judgment and screening can be performed, which inevitably results in accuracy of subsequent reconstructed signals. That is, if the noise-dominant component is selected by mistake in the process, the reconstructed signal still contains noise, that is, the noise reduction is incomplete, so that errors exist in subsequent feature extraction, and the state change in the standard real system cannot be corrected. Therefore, the formulation of the IMF component signal screening criteria is necessary and important. Among them, the most common method is the correlation coefficient method, which calculates the pearson correlation coefficient of each component and the original signal, and uses the standard deviation of the correlation coefficient as a threshold, and if the correlation coefficient is greater than the threshold, the component is retained. Although the method can effectively remove noise in an analog signal, when the method is applied to an actual signal, it can be found that as the order of the IMF component signal increases, the correlation with the original signal decreases exponentially, i.e. the correlation is lower than the rest of the IMF component signals except the first few components, and then the threshold is also influenced by a plurality of IMF component signals with low correlation. Therefore, the method is difficult to apply to an actual signal. Meanwhile, many researchers combine means such as wavelet soft threshold and interval threshold, and although the threshold selection is more flexible and the signal-to-noise ratio is larger than that of the signal reconstructed by the correlation numerical method, the calculation is too complex, and wavelet decomposition needs to be performed on each component after EEMD decomposition is performed on the signal. Therefore, an accurate IMF component signal screening method which is suitable for real signals and is simple and convenient to calculate is needed, so that the signals after noise reduction are close to effective signals to the maximum extent, the subsequent signal feature identification is facilitated, and the real dynamic characteristics of the system are reflected.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a signal noise reduction method based on an ensemble empirical mode decomposition method and a power spectrum, which is used for a nonlinear system containing white noise and periodic noise. The method can effectively overcome the application defects in the prior art, quickly judge the noise content of the IMF component signals and reasonably select part of the components to reconstruct the signals, and prevent the excessive noise components of the reconstructed signals caused by improper component selection.
The purpose of the invention is realized by the following technical scheme:
a signal noise reduction method based on an ensemble empirical mode decomposition method and a power spectrum comprises the following steps:
step 1: decomposing the nonlinear signal by EEMD to obtain an IMF component signal and a residual error;
step 2: calculating and drawing a power spectrum of each IMF component signal;
and step 3: quantitatively analyzing the power spectrum of the IMF component signals, calculating the proportion of the peak bandwidth of the power spectrum of each IMF component signal to the total bandwidth, and screening the required IMF component signals;
wherein the content of the first and second substances,P i1 is the ratio of the bandwidth of the peak part of the power spectrum,W iB is the bandwidth of the peak portion of the power spectrum,iis an IMF component signal of 1 to n orders, and B is the total bandwidth;
and 4, step 4: counting the number of power spectrum extreme points of each IMF component signal, calculating the proportion of the power spectrum extreme points occupying IMF1, and screening the required IMF component signals;
wherein the content of the first and second substances,P i2 is the ratio of the power spectrum extreme points of each IMF component signal,N i is the firstiThe number of extreme points of each IMF component signal,N 1 the number of the power spectrum extreme points of the IMF1 component signal is shown;
and 5: superposing each IMF component signal meeting the conditions of the step 3 and the step 4 and the residual error obtained in the step 1 to complete signal reconstruction;
wherein the content of the first and second substances,Ris a reconstructed signal that is to be transmitted, nis the total number of IMF component signals,kit is indicated that the k-th one,ris a residual error that is a function of the error,
indicates that step 3 and step 4 are satisfiediAn intrinsic mode component.
Further, in step 1, the nonlinear signal refers to a nonlinear signal containing white noise and periodic noise.
Further, in step 2, calculating and plotting a power spectrum of each IMF component signal specifically includes:
2.1 The autocorrelation function is calculated first, and then fourier transform is performed to obtain the power spectrum of the IMF component signal.
2.2 Plotted with bandwidth as the abscissa and power as the ordinate, the power spectrum of the IMF component signal.
Further, in step 3, the screening of the required IMF component signal specifically includes:
the ratio P of the peak part bandwidth of the power spectrum of the IMF component signal to the total bandwidth is increased along with the increasing of the order of the IMF component signal i1 Less than 3%, retaining all IMF component signals following the IMF component signal, otherwise, discarding the IMF component signal.
Further, in step 4, the screening of the required IMF component signal specifically includes:
with the increasing of the order of the IMF component signal, the ratio P of the power spectrum extreme point of the IMF component signal to the power spectrum extreme point of the IMF1 component i2 Less than 3%, retaining all IMF component signals following the IMF component signal, otherwise, discarding the IMF component signal.
The invention has the following beneficial effects:
1) The method combines the power spectrum to select IMF component signals obtained by EEMD decomposition of original signals, performs quantitative analysis on the power spectrum of each IMF component signal, sets a proper threshold value, reasonably screens the IMF component signals to perform signal reconstruction, and enables the reconstructed signals after noise reduction by the method to be close to effective signals to the maximum extent.
2) The method can be effectively applied to nonlinear systems containing white noise and periodic noise, such as bearing fault detection, milling chatter recognition, cutter wear monitoring and the like, the noise content of IMF component signals is quickly and simply judged, part of components in the IMF component signals are reasonably selected to reconstruct the signals, and a series of problems that the reconstructed signal noise components are too much due to improper component selection, the subsequently extracted signal characteristics are wrong, the state change in a true system cannot be corrected and the like are prevented.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is a graph of simulated signals constructed in accordance with an embodiment of the present invention;
FIG. 3 is a graph of EEMD decomposition results for an analog signal according to an embodiment of the present invention;
FIG. 4 is a power spectrum of each IMF component signal of the analog signal according to an embodiment of the present invention;
FIG. 5 is a diagram of a calculation method of step 3 according to an embodiment of the present invention;
FIG. 6 is a diagram of a reconstructed signal according to an embodiment of the invention;
FIG. 7 is a diagram of an analog signal, not a reconstructed signal of the present method according to an embodiment of the present invention;
FIG. 8 is an IMF1-IMF5 component signal and its power spectrum according to an embodiment of the present invention;
FIG. 9 is an IMF6-IMF10 component signal and its power spectrum according to an embodiment of the present invention;
FIG. 10 is a graph of the original and reconstructed signals and their power spectra according to an embodiment of the invention;
fig. 11 is a diagram of a reconstructed signal of an actual signal without the method according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments, and the objects and effects of the present invention will become more apparent, it being understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.
As shown in FIG. 1, the signal noise reduction method based on Ensemble Empirical Mode Decomposition (EEMD) and power spectrum of the present invention comprises the following steps:
step 1: EEMD decomposition is carried out on the nonlinear signal to obtain an IMF component signal and a residual error.
As shown in fig. 2, the analog signal constructed for this example contains a high frequency signal, a low frequency noise signal and white gaussian noise, and the formula is as follows:
wherein the content of the first and second substances,tthe time is represented by the time of day,
represents a low-frequency signal->
Representing high-frequency noise, processing the high-frequency signal into a periodic noise, based on the value of the noise value, and evaluating the value of the noise value>
Representing white gaussian noise with a signal to noise ratio of 10dB added to the low frequency signal. EEMD decomposition is performed on a nonlinear signal containing white noise and periodic noise to obtain an IMF component signal and a residual error, as shown in FIG. 3, the EEMD decomposition result of the analog signal is obtained by dividing the signal in different scales to obtain 7 IMF component signals and a residual error.
Step 2: a power spectrum of each IMF component signal is calculated and plotted.
2.1 The autocorrelation function is calculated first, and then fourier transform is performed to obtain the power spectrum of the IMF component signal.
2.2 Plotted with bandwidth as the abscissa and power as the ordinate, the power spectrum of the IMF component signal.
As shown in fig. 4, is the power spectrum of each IMF component signal.
And step 3: and quantitatively analyzing the power spectrum of the IMF component signals, calculating the proportion of the peak bandwidth of the power spectrum of each IMF component signal to the total bandwidth, and screening the required IMF component signals.
Wherein the content of the first and second substances,P i1 is the ratio of the bandwidth of the peak part of the power spectrum,W Bi is the bandwidth of the peak portion of the power spectrum,iis the IMF component signal of 1 to n orders, and B is the total bandwidth.
TABLE 1 ratio of power spectrum peak portion bandwidth of each IMF component signal
Because the white noise signal can present tiny peak on the power spectrum, and the periodic noise presents more prominent peak on the power spectrum, no matter which kind of noise is presented on the power spectrum, only the IMF component signal with less peak part bandwidth, the signal can be closer to the real signal. Fig. 5 shows a calculation method according to the present embodiment. As shown in Table 1, for the ratio of the peak part bandwidth of the power spectrum of each IMF component signal to the total bandwidth, the ratio of the peak part bandwidth of the power spectrum of the 4 th IMF component signal to the total bandwidth increases with the increasing order of the IMF component signalP i1 And if the IMF component signal is less than 3%, all IMF component signals after the 4 th IMF component signal are reserved, namely IMF 4-IMF 7 are reserved.
And 4, step 4: and counting the number of the power spectrum extreme points of each IMF component signal, calculating the proportion of the power spectrum extreme points occupying IMF1, and screening the required IMF component signals.
Wherein the content of the first and second substances,P i2 is the ratio of the power spectrum extreme points of each IMF component signal,N i is the firstiThe number of extreme points of each IMF component signal,N 1 the number of the power spectrum extreme points of the IMF1 is shown, and the IMF1 is the 1 st IMF component signal.
The step 3 above has performed the first round of screening, and the second round of screening is screening the number of power spectrum extreme points of the IMF component signals. And counting the number of the power spectrum extreme points of each IMF component signal, wherein the extreme points comprise a maximum point and a minimum point, and the number of the power spectrum extreme points of each IMF component signal and the proportion of the power spectrum extreme points occupying the IMF1 component signal are shown in table 2.
TABLE 2 ratio of the number of power spectrum extremum points of each IMF component signal
Since the first filtering has removed part of the IMF component signals, there may be a lot of noise that although the peak portion bandwidth is less than 3%, the intensity of the extreme points in the portion is high, the number of the extreme points is too many, and the noise still exists. Therefore, the extreme points of each IMF component signal are counted, and as can be seen from Table 2, the number of the extreme points of the power spectrum of the 4 th IMF component signal accounts forP i2 And if the IMF component signal is less than 3%, retaining all IMF component signals after the 4 th IMF component signal, namely retaining IMF 4-IMF 7 component signals. Although the results of the second round of screening for the analog signal are still the same as those obtained in step 3, the round of screening is necessary for the actual signal.
And 5: and (5) superposing the IMF component signals meeting the conditions of the step (3) and the step (4) and the residual errors obtained in the step (1) to complete signal reconstruction.
Wherein the content of the first and second substances,Ris a reconstructed signal that is a function of the signal,nis the total number of IMF component signals,kis shown askThe number of the main components is one,ris a residual error that is a function of the error,
indicates fullStep 3 and step 4iAn intrinsic mode component.
As shown in fig. 6, the IMF component signals obtained after the analog signal is processed in steps 1 to 4 and the residual reconstructed signals are compared with fig. 2, and it can be seen that the noise in the original signal is completely removed. As shown in fig. 7, noise still exists in the time domain in the IMF3-7 component signal and the residual reconstructed signal, and the IMF5-7, IMF6-7 component signal and the residual reconstructed signal lose the waveform of the original signal, which indicates that the noise reduction is excessive. Pearson's correlation coefficient of the original signal and the reconstructed signal is calculated up to 0.9744. Therefore, the IMF component signals can be reasonably screened, noise reduction of the signals is guaranteed to the maximum extent, and effective signal waveforms cannot be excessively lost due to noise reduction.
The above is a specific embodiment of the present invention applied to analog signals, and the method is applied to vibration signals in the actual rolling process.
The original signal is decomposed into 17 th order IMF component signals and residual errors through step 1, as shown in fig. 8, the IMF1-IMF5 component signals obtained through step 1 and step 2 and their power spectra are shown, and it can be known from the figure that the power spectrum of the original signal has many fine and prominent peaks, which indicates that periodic signals and white noise exist in the signal. The IMF6-IMF10 component signals and their power spectra are shown in fig. 9. The bandwidth ratio of the peak part of the power spectrum of each IMF component signal in the table 3 is obtained through the step 3 (no calculation is carried out after the IMF10 component signal is obtained).
TABLE 3 Peak portion Bandwidth ratio of Power Spectrum of IMF component signals
As shown in Table 3, for the ratio of the peak part bandwidth of the power spectrum of each IMF component signal to the total bandwidth, the ratio of the peak part bandwidth of the power spectrum of the 7 th IMF component signal to the total bandwidth increases with the increasing order of the IMF component signalP i1 And if the IMF component signal is less than 3%, retaining all IMF component signals after the 7 th IMF component signal, namely retaining IMF 7-IMF 17 component signals.
And 4, counting the number of the extreme points of the power spectrum of each IMF component signal through the step 4, wherein the extreme points comprise maximum points and minimum points, and the number of the extreme points of the power spectrum of each IMF component signal and the proportion of the extreme points of the power spectrum of each IMF component signal in the IMF1 component signal are shown in the table 4.
TABLE 4 power spectrum extremum number ratio of each IMF component signal
Since the first filtering has removed part of the IMF component signals, there may be a lot of noise that although the peak portion bandwidth is less than 3%, the intensity of the extreme points in the portion is high, the number of the extreme points is too many, and the noise still exists. Therefore, the extreme points of each IMF component signal are counted, and as can be seen from Table 4, the power spectrum extreme point number of the 8 th IMF component signal is in proportion to the number of the extreme pointsP i2 And if the IMF component signal is less than 3%, retaining all IMF component signals after the 8 th IMF component signal, namely retaining IMF 8-IMF 17 component signals.
The reconstructed signal obtained in step 5 is compared with the original signal as shown in fig. 10, and it can be seen that the noise in the original signal is completely removed, and the power spectrum is smooth and has no obvious peak. As shown in FIG. 11, the signal power spectrum reconstructed by the IMF 7-17 component signals still has few peaks, i.e. noise still exists, and although the power spectrum reconstructed by the IMF 8-17 and IMF 9-17 component signals is very smooth, the signal time domain diagram shows that important parts in effective signals are also removed. Therefore, the method can reasonably screen the IMF component signals for signal reconstruction, and the reconstructed signals subjected to noise reduction by the method are close to effective signals to the maximum extent.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and although the invention has been described in detail with reference to the foregoing examples, it will be apparent to those skilled in the art that various changes in the form and details of the embodiments may be made and equivalents may be substituted for elements thereof. All modifications, equivalents and the like which come within the spirit and principle of the invention are intended to be included within the scope of the invention.
Claims (2)
1. A signal noise reduction method based on an ensemble empirical mode decomposition method and a power spectrum is characterized by comprising the following steps:
step 1: decomposing the nonlinear signal by using a set empirical mode decomposition method to obtain an intrinsic mode component and a residual error;
step 2: calculating and drawing a power spectrum of each intrinsic mode component;
in step 2, calculating and drawing a power spectrum of each eigenmode component, specifically including:
2.1 Calculating an autocorrelation function, and then carrying out Fourier transformation to obtain a power spectrum of the intrinsic mode component;
2.2 With the bandwidth as the abscissa and the power as the ordinate, drawing a power spectrum of the eigenmode component;
and step 3: quantitatively analyzing the power spectrum of the intrinsic mode components, calculating the proportion of the peak bandwidth of the power spectrum of each intrinsic mode component to the total bandwidth, and screening the required intrinsic mode components;
wherein, P 1i Is the ratio of the bandwidth of the peak part of the power spectrum, W Bi Is the bandwidth of the peak part of the power spectrum, i is the 1-n order eigenmode component signal, B is the total bandwidth;
in step 3, screening the required eigenmode components, specifically including:
the ratio P of the peak part bandwidth of the power spectrum of the intrinsic mode component to the total bandwidth is increased along with the increasing of the order of the intrinsic mode component 1i If the ratio is less than 3%, retaining all the intrinsic mode components behind the intrinsic mode component, otherwise, discarding the intrinsic mode component;
and 4, step 4: counting the number of power spectrum extreme points of each eigenmode component, calculating the proportion of the power spectrum extreme points occupying the first eigenmode, and screening the required eigenmode components;
wherein, P 2i Is the ratio of the extreme points of the power spectrum of each eigenmode component, N i Is the number of the ith eigenmode component extreme points, N 1 The number of the power spectrum extreme points of the first eigenmode component;
in step 4, screening the required eigenmode components, specifically including:
with the increasing of the order of the eigenmode component, the ratio P of the power spectrum extreme point of the eigenmode component to the power spectrum extreme point of the first eigenmode component 2i If the ratio is less than 3%, retaining all the eigenmode components behind the eigenmode component, otherwise, discarding the eigenmode component;
and 5: superposing each intrinsic mode component meeting the conditions of the step 3 and the step 4 and the residual error obtained in the step 1 to complete signal reconstruction;
wherein R is a reconstructed signal, n is the total number of eigenmode components, k represents the kth, R is a residual, IMF i Indicating the ith eigenmode component satisfying step 3 and step 4.
2. The method of signal noise reduction based on ensemble empirical mode decomposition and power spectrum according to claim 1, wherein: in step 1, the nonlinear signal refers to a nonlinear signal containing white noise and periodic noise.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211587362.6A CN115659128B (en) | 2022-12-12 | 2022-12-12 | Signal noise reduction method based on ensemble empirical mode decomposition method and power spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211587362.6A CN115659128B (en) | 2022-12-12 | 2022-12-12 | Signal noise reduction method based on ensemble empirical mode decomposition method and power spectrum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115659128A CN115659128A (en) | 2023-01-31 |
| CN115659128B true CN115659128B (en) | 2023-04-07 |
Family
ID=85017114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211587362.6A Active CN115659128B (en) | 2022-12-12 | 2022-12-12 | Signal noise reduction method based on ensemble empirical mode decomposition method and power spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115659128B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116155623B (en) * | 2023-04-17 | 2023-08-15 | 湖南大学 | Digital audio encryption method and system based on power grid frequency characteristic embedding |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106645943A (en) * | 2016-09-14 | 2017-05-10 | 哈尔滨工业大学 | Weak signal denoising method based on wavelet theory and EEMD |
| CN107783200A (en) * | 2017-11-21 | 2018-03-09 | 吉林大学 | Joint EMD and TFPF algorithms a kind of all-wave magnetic resonance signal random noise method for reducing |
| CN108229382A (en) * | 2017-12-29 | 2018-06-29 | 广州供电局有限公司 | Vibration signal characteristics extracting method, device, storage medium and computer equipment |
| KR20190092753A (en) * | 2018-01-31 | 2019-08-08 | 한국항공대학교산학협력단 | Body of revolution fault analysis apparatus and method using transmission error |
| WO2019179340A1 (en) * | 2018-03-19 | 2019-09-26 | 河北工业大学 | Eemd- and msb-based failure feature extraction method for rolling-element bearing |
| CN114325598A (en) * | 2021-12-10 | 2022-04-12 | 网络通信与安全紫金山实验室 | Signal denoising method and device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247931A (en) * | 2017-05-27 | 2017-10-13 | 哈尔滨理工大学 | A kind of De-noising Technique For Vibration Signals of Rotating Machinery |
| CN109785854B (en) * | 2019-01-21 | 2021-07-13 | 福州大学 | Speech enhancement method combining empirical mode decomposition and wavelet threshold denoising |
| CN110132403A (en) * | 2019-04-02 | 2019-08-16 | 天津大学 | A kind of vacuum pump vibration signal noise-reduction method based on EEMD and wavelet threshold |
| CN110514294A (en) * | 2019-08-30 | 2019-11-29 | 鞍钢矿业爆破有限公司 | A kind of blasting vibration signal noise-reduction method based on EMD and VMD |
| CN111839492B (en) * | 2020-04-20 | 2022-10-18 | 合肥工业大学 | Heart rate non-contact type measuring method based on face video sequence |
| CN113633288A (en) * | 2021-06-25 | 2021-11-12 | 杭州电子科技大学 | EEMD-based fatigue electroencephalogram feature extraction method |
| CN114098656B (en) * | 2022-01-26 | 2022-05-06 | 广东工业大学 | Signal noise reduction method and system based on empirical mode decomposition and bit plane conversion |
| CN114676736B (en) * | 2022-05-13 | 2022-08-26 | 浙江工业大学 | Method for predicting surface quality of rolled workpiece |
-
2022
- 2022-12-12 CN CN202211587362.6A patent/CN115659128B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106645943A (en) * | 2016-09-14 | 2017-05-10 | 哈尔滨工业大学 | Weak signal denoising method based on wavelet theory and EEMD |
| CN107783200A (en) * | 2017-11-21 | 2018-03-09 | 吉林大学 | Joint EMD and TFPF algorithms a kind of all-wave magnetic resonance signal random noise method for reducing |
| CN108229382A (en) * | 2017-12-29 | 2018-06-29 | 广州供电局有限公司 | Vibration signal characteristics extracting method, device, storage medium and computer equipment |
| KR20190092753A (en) * | 2018-01-31 | 2019-08-08 | 한국항공대학교산학협력단 | Body of revolution fault analysis apparatus and method using transmission error |
| WO2019179340A1 (en) * | 2018-03-19 | 2019-09-26 | 河北工业大学 | Eemd- and msb-based failure feature extraction method for rolling-element bearing |
| CN114325598A (en) * | 2021-12-10 | 2022-04-12 | 网络通信与安全紫金山实验室 | Signal denoising method and device |
Non-Patent Citations (2)
| Title |
|---|
| Jiqiu Tan等.Diesel Engine Fault Diagnosis and Denoising Disposal of the Vibration Signals Based on EMD Method.《 2013 Fourth International Conference on Digital Manufacturing & Automation》.2013,第488-491页. * |
| 剡昌锋;周礼龙;韦尧兵;由理;吴旭东.基于EEMD降噪与二次相位耦合的滚动轴承故障诊断.兰州理工大学学报.2016,第42卷(第03期),第35-40页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115659128A (en) | 2023-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE102005033256B4 (en) | Device and method for generating a spectrogram mask trigger and spectrogram mask | |
| CN115659128B (en) | Signal noise reduction method based on ensemble empirical mode decomposition method and power spectrum | |
| EP2277039A1 (en) | Method and device for the classification of sound-generating processes | |
| DE102014226073A1 (en) | Method and device for operating a radar system of a motor vehicle | |
| CN110503060B (en) | Spectral signal denoising method and system | |
| CN112115856B (en) | Electroencephalogram quality assessment method, storage medium and system | |
| DE102013200941A1 (en) | Method and apparatus for determining a trigger condition for a rare signal event | |
| CN109781245B (en) | Objective evaluation method for pulse noise of diesel engine | |
| CN112859012A (en) | Radar deception jamming identification method based on cascade convolution neural network | |
| DE3408404A1 (en) | RADAR SIMULATION PROCEDURE | |
| CN113657737A (en) | Water quality evaluation method, system, equipment and storage medium based on automatic monitoring data | |
| CN117390434A (en) | Transient electromagnetic signal time delay estimation method based on multi-scale decomposition and reconstruction | |
| US7459962B2 (en) | Transient signal detection algorithm using order statistic filters applied to the power spectral estimate | |
| CN110780162B (en) | Method for extracting partial discharge signal of primary and secondary fusion power distribution switch and detection device | |
| US20090248336A1 (en) | Analyzer for signal anomalies | |
| CN112539887A (en) | WT-LCD-WD-based pipeline leakage signal denoising method | |
| DE102020118540A1 (en) | SYSTEMS, DEVICES AND METHODS FOR REAL-TIME FAULT DETECTION | |
| DE112008001249T5 (en) | Radio receiving apparatus and noise elimination method in this apparatus | |
| CN116136518A (en) | Chromatograph | |
| CN116667920A (en) | Transient signal detection method combining principal component analysis and Hilbert-Huang transform | |
| DE60100646T2 (en) | Method and device for analog-digital conversion using previous signal samples | |
| CN113064122A (en) | Performance evaluation method, system and medium of P-band SAR (synthetic Aperture Radar) interference suppression algorithm | |
| DE102005037628B3 (en) | Method for detecting an object by means of radar | |
| CN117874433B (en) | Monitoring system of deep denitrification biological filtration equipment for water treatment | |
| DE102018220600A1 (en) | Particle detection method and apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |