CN113158714A - Fault feature self-adaptive extraction method based on wavelet entropy and EEMD - Google Patents
Fault feature self-adaptive extraction method based on wavelet entropy and EEMD Download PDFInfo
- Publication number
- CN113158714A CN113158714A CN202011215620.9A CN202011215620A CN113158714A CN 113158714 A CN113158714 A CN 113158714A CN 202011215620 A CN202011215620 A CN 202011215620A CN 113158714 A CN113158714 A CN 113158714A
- Authority
- CN
- China
- Prior art keywords
- wavelet
- decomposition
- entropy
- imf
- signal
- 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.)
- Granted
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 43
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 29
- 230000009467 reduction Effects 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 17
- 230000003044 adaptive effect Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 10
- 238000009825 accumulation Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 9
- 238000005215 recombination Methods 0.000 claims description 8
- 230000006798 recombination Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 abstract description 3
- 238000012216 screening Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 12
- 238000004458 analytical method Methods 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
Abstract
The invention discloses a fault feature self-adaptive extraction method based on wavelet entropy and EEMD, which comprises the following steps: carrying out inherent modal decomposition on the collected mechanical product operation historical signal by using ensemble empirical modal decomposition; calculating the wavelet energy entropy of the intrinsic mode function component obtained by decomposition; extracting an inherent modal function component containing most complex information; performing wavelet threshold denoising on the extracted inherent modal function component; and recombining the intrinsic mode function components after noise reduction. The method utilizes wavelet energy entropy to self-adaptively determine the inherent modal function component with the most energy information, thereby completing self-adaptive screening of the inherent modal function without depending on experience and test and rapidly extracting the fault characteristic frequency band.
Description
Technical Field
The invention relates to a fault feature extraction technology, in particular to a fault feature self-adaptive extraction method based on wavelet entropy and EEMD.
Background
With the development and progress of science and technology, mechanical equipment plays an important role in modern industrial application, and with the increasing complexity of equipment structures, the feature extraction, further diagnosis and evaluation of the operation process information of the mechanical equipment are also increasingly important. Rotating machines are important components of mechanical devices, and their operating conditions are directly related to the performance of the devices. In order to ensure the long-term and efficient normal operation of the rotating machinery, maintenance is reasonably organized according to the health state of electromechanical equipment, and excessive maintenance and insufficient maintenance can be avoided. The implementation of the 'predicted maintenance' can effectively reduce the maintenance cost of the equipment, prolong the service life of the equipment and shorten unnecessary downtime. Therefore, the monitoring and diagnosis of the current state of the mechanical equipment are of great significance.
However, the rapid and accurate implementation of fault feature extraction becomes a significant difficulty in current mechanical fault diagnosis and research work. When a fault occurs, the complexity of the natural oscillation of the mechanical system changes, which brings about a great trouble to the analysis and processing of the fault signal. To detect mechanical equipment failures, a number of methods have been developed. At present, most of feature extraction technologies only rely on traditional root mean square and peak values to perform feature extraction, and a series of values are usually extracted through a fixed algorithm by the traditional feature extraction technology, without considering the actual operating environment and operating state of equipment. Furthermore, merely obtaining a change in value is not sufficient to understand what changes are made to the device.
The feature extraction technology is one of core technologies of mechanical equipment reliability evaluation and health management, and in a signal analysis technology, the most important is feature extraction and feature classification identification, and the quality of the feature extraction greatly influences the effect of subsequent feature classification identification, so the feature extraction is the most important, and the traditional feature extraction technology comprises the following steps: time domain analysis, frequency domain analysis and the like are not enough to represent specific change reasons of different signals, only a single numerical value represents, which cannot be known what kind of changes are generated by equipment, fault characteristic frequency is difficult to identify, and the number of key Intrinsic Mode Functions (IMF) cannot be identified. That is, one signal may require 4 Intrinsic Mode Functions (IMFs) to contain the desired fault information, while another signal may require 3 or 5. Choosing the appropriate eigenmode function (IMF) has a direct impact on the next more detailed feature extraction or feature classification.
Disclosure of Invention
Therefore, the invention provides a fault feature self-adaptive extraction method based on wavelet entropy and Ensemble Empirical Mode Decomposition (EEMD) based on the device operation process state data according to the definition of the wavelet entropy and the EEMD.
The technical solution for realizing the purpose of the invention is as follows: a fault feature self-adaptive extraction method based on wavelet entropy and EEMD comprises the following steps:
step 1: carrying out inherent modal decomposition on the acquired original signal by using EEMD (ensemble empirical mode decomposition), adding white noise for noise reduction, and simultaneously sequentially expanding the original signal from high frequency to low frequency to obtain a plurality of IMF components by decomposition;
step 2: according to wavelet packet decomposition, calculating the wavelet energy entropy of each IMF component obtained in the step 1;
and step 3: extracting IMF components with most complex information according to the wavelet energy entropy calculated in the step 2;
and 4, step 4: performing wavelet threshold denoising on the IMF component extracted in the step 3;
and 5: and 4, performing signal accumulation recombination on the IMF component subjected to noise reduction in the step 4 to obtain a brand new fault characteristic frequency band section.
Further, the specific process of step 1 is as follows:
setting the total number of decomposition iterations as NE, and initializing the amplitude of the added white noise;
1) the m-th decomposition of the original signal with white noise added thereto, m being 1,2, …, NE, initially m being 1, is the original signal x (t) with white noise added thereto:
Xm(t)=X(t)+nm(t)
wherein, Xm(t) represents the mth denoised signal; n ism(t) represents white noise added by the mth decomposition;
2) decomposing the m-th noisy signal X by using an empirical mode methodm(t) obtaining the ith IMF component IMFiI ═ 1,2, …, NE; if m < NE, return to step 1), and m ═ m +1, repeat steps 1) and 2);
calculating NE IMF components and signal residual r obtained after NE decomposition:
further, the specific process of step 2 is as follows:
selecting the number of wavelet packet decomposition layers as N, and expressing the reconstruction coefficient of the original signal X (t) at the j-th node of the k-th layer as D ═ Dk(j),j=1,2,…,2kJ is a node corresponding to each IMF component of the kth layer, and k is 0,1,2, …, N;
computing wavelet energies E for node subbands of respective IMF componentsN,j:
Ek,j=|dk(j)|2
Calculating the ratio P of wavelet energy of node sub-band segment of each IMF componentj:
Calculating the wavelet energy entropy W of each IMF component obtained by decompositionE:
Further, step 3 comprises: comparing and analyzing the energy change of the wavelet energy entropy by a method of calculating and accumulating the wavelet energy entropy; and extracting IMF components with wavelet energy entropy having larger turning points.
Further, step 4 comprises: carrying out wavelet fixed threshold denoising on the extracted IMF component; selecting wavelet fixed threshold denoising according to denoising performance, selecting noise level estimation according to the first layer wavelet decomposition for adjustment, setting the decomposition layer number to be 3, and selecting db3 as a wavelet base.
Further, the signal accumulation and recombination in the step 5 refers to signal superposition of the IMF components denoised by the wavelet threshold in the step 4,
wherein X '(t) is a reconstructed fault frequency band signal, IMF'iAnd de-noising the ith IMF component of the wavelet.
Further, in step 1, the original signal includes any section of historical signal from the beginning of use to complete failure of the current device, or a device of the same type as the current device and in the same working environment.
A fault feature self-adaptive extraction system based on wavelet entropy and EEMD comprises:
the natural mode decomposition module is used for performing natural mode decomposition on the acquired original signal by using the EEMD, sequentially expanding the original signal from high frequency to low frequency while adding white noise for noise reduction, and decomposing to obtain a plurality of IMF components;
the wavelet energy entropy calculation module is used for calculating the wavelet energy entropy of each IMF component according to wavelet packet decomposition;
the IMF component extraction module is used for extracting IMF components with most complex information according to the wavelet energy entropy;
the denoising module is used for performing wavelet threshold denoising on the extracted IMF component with the most complex information;
and the fault characteristic frequency band section extraction module is used for performing signal accumulation recombination on the IMF components subjected to noise reduction to obtain a brand new fault characteristic frequency band section.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
Compared with the prior art, the invention has the beneficial effects that:
1) the invention fully utilizes the characteristics of ensemble empirical mode decomposition integrity and wavelet transformation locality to decompose signals to different frequency bands, and simultaneously carries out wavelet transformation denoising, combines the characteristics of wavelet packet transformation description signal locality and the characteristic that information entropy can reflect signal integrity, and utilizes the change of wavelet energy entropy to judge the change of specific frequency band segments, thereby all signals under different working conditions can adaptively select the optimal number of Intrinsic Mode Functions (IMF);
2) the invention can also carry out signal noise reduction on the extracted signal, which has very important significance for carrying out efficient and effective self-adaptive feature extraction in the technical fields of signal processing, feature extraction and the like of mechanical equipment reliability evaluation and health monitoring;
3) according to the invention, the acquired equipment operation process signal is decomposed through EEMD, the wavelet energy entropy is utilized to calculate the wavelet energy accumulation entropy of the decomposed IMF, and finally, the fault characteristic band segment is extracted in a self-adaptive manner according to the change of the wavelet entropy, so that a corresponding technical basis is provided for subsequent characteristic classification and the like.
Drawings
FIG. 1 is a diagram illustrating simulated signal components in accordance with an embodiment of the present invention.
FIG. 2 is a diagram illustrating the addition of Gaussian white noise to an artificial signal according to an embodiment of the present invention.
Fig. 3 is a flowchart of a fault feature adaptive extraction method based on wavelet entropy and EEMD according to an embodiment of the present invention.
FIG. 4 is a diagram illustrating decomposition results of an emulated signal through an EEMD in accordance with an embodiment of the present invention.
FIG. 5 is a diagram illustrating IMF wavelet entropy of an emulated signal passing through an EEMD in accordance with an embodiment of the present invention.
FIG. 6 is a diagram illustrating an example of an extracted signal according to an embodiment of the present inventionAnd (3) a denoising result schematic diagram for performing wavelet fixed threshold denoising.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, the present embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the present invention belongs.
In order to ensure the accuracy of the test result, the present embodiment uses the simulation signal of the device operation history signal (hereinafter referred to as the original signal) as the test basis for analysis, as shown in fig. 1, where the device operation process signal mainly refers to any section of history signal from the beginning of use to the complete failure, which is collected from the current device or a device of the same type as the current device and in the same working environment. The original signal to be decomposed in this embodiment includes a mixed signal with amplitude of 1 and vibration frequency of 1HZ and 100HZ, the 100HZ is artificially defined as the main frequency signal, the 1HZ signal is noise impact, and random white noise is added to the original signal, as shown in fig. 2.
Fig. 3 shows a flowchart of a fault feature adaptive extraction method based on wavelet entropy and EEMD provided by the present embodiment, which specifically includes the following steps:
step 1: performing inherent modal decomposition on the original signal by using EEMD;
selecting a simulation superimposed signal for research, wherein the added white noise signal-to-noise ratio Nstd is 0.2, the iteration total number NE is selected for 15 times, EEMD is carried out on the input original signal X (t), and the original signal is expanded from high frequency to low frequency. The specific process is as follows:
1) firstly, initializing an amplitude value of white noise to be added, decomposing a signal to be decomposed added with the white noise for the mth time, and setting m to be 1 at the initial time; adding white noise to the original signal X (t)
Xm(t)=X(t)+nm(t)
Wherein n ism(t) white noise added for the mth decomposition, xmAnd (t) is the signal after the mth time of noise addition.
2) Decomposing the noisy signal by using an empirical mode method to obtain 15 IMF components;
if m < 15, return to step 1), and m ═ m + 1; repeating steps 1) and 2), calculating each IMF component after m times of decomposition to decompose the original signal into 15 IMF components and a signal residual component:
wherein, IMFiThe ith IMF component obtained by the ith decomposition is obtained; r is the signal residual component. The first four IMF components IMF1-4 resulting from EEMD of the original signal X (t) are shown in FIG. 4.
Step 2: calculating the wavelet packet energy of each IMF component by using a wavelet decomposition method and calculating the entropy value thereof, wherein the specific process is as follows:
in this embodiment, the number of wavelet packet decomposition layers is selected to be 3'db 3' is a wavelet base, and represents the reconstruction coefficient of the original signal x (t) at the jth node of the kth layer as D ═ Dk(j) J is 1,2, …,8, j is a node corresponding to each IMF component, k is 0,1,2, 3;
computing wavelet energies E for node subbands of respective IMF componentsk,j:
Ek,j=|dk(j)|2
Calculating the ratio P of wavelet energy of node sub-band segment of each IMF componentj,PjThe method has very sensitive performance to energy change, and can embody the distribution of wavelet energy and the energy relative relation of each subband sequence signal:
calculating the wavelet energy entropy WE of each IMF component obtained by decomposition:
the energy entropy of the wavelet obtained by decomposing the original signal is shown in tables 1(a) and (b):
TABLE 1 wavelet energy entropy of original signal
(a)
(b)
In the table, the number of the first and second,wavelet energy entropy representing the decomposed 1 st IMF component;a wavelet energy entropy representing the signal after the merging of the 1 st and 2 nd IMF components of the decomposition; by analogy in the following way,represents the wavelet energy entropy of the signal after the merging of the 1 st to 9 th IMF components of the decomposition.
And step 3: extracting an inherent mode function with most complex information as a fault characteristic frequency band signal;
the entropy values of the wavelet energy entropies obtained by the above calculation are compared, and whether the corresponding frequency band has a sudden change or not is judged according to the definition of the entropy values, the result obtained by the embodiment is shown in fig. 5, and the result shows that the energy change of the input signal occurs when the input signal is in the IMF6, so that the IMFs 1-5 and the IMFs 6-9 are extracted to be two segments of characteristic frequency bands respectively. Due to the characteristics of the ensemble empirical mode decomposition algorithm (the signal is spread from high frequency to low frequency) and the basic concept of entropy, the low-frequency impact can be judged to occur in the IMF6, and therefore the IMF6-9 is extracted as the characteristic fault frequency band signal.
And 4, step 4: performing wavelet threshold denoising on the extracted inherent modal function;
the extracted IMF6-9 and the residual IMF1-5 are denoised by a wavelet fixed threshold denoising method to obtain an IMF1-5 wavelet fixed threshold denoising signal and an IMF6-9 wavelet fixed threshold denoising signal, wherein the fixed threshold denoising method is selected, and the noise level estimation according to the first layer wavelet decomposition is selected for adjustment, and the result is shown in FIG. 6 and FIG. 7.
And 5: recombining the noise-reduced inherent mode functions;
and performing signal accumulation recombination on the wavelet fixed threshold denoising signal to obtain a brand new fault characteristic frequency band segment. Namely, the natural mode functions subjected to the wavelet threshold denoising in the step 4 are subjected to signal superposition:
where X '(t) is the reconstructed fault band signal, IMF'iAnd de-noising the ith IMF component of the wavelet.
Since the simulation data in this embodiment set 1HZ as the interference noise, the extracted eigenmode function, i.e. the reconstructed fault band signal X' (t), restores the signal, and thus the method of the present invention is verified.
In conclusion, through the steps, the invention can adaptively obtain the fault frequency band sections of the signals under different conditions, can adaptively extract the optimal number of the intrinsic mode functions from the process signals, can also perform signal noise reduction on the extracted signals, and has very important significance for efficiently and effectively performing adaptive feature extraction.
The invention can adaptively extract the optimal number of the natural modal functions according to signals under different conditions, and provides certain reference guidance for the selection of the natural modal functions under different conditions. The method is not only suitable for the simulation data set in the case, but also suitable for other mechanical equipment and key parts thereof, and provides reasonable reference for feature extraction. In addition, the invention has good expandability and provides a certain reference function for other technical personnel in the technical field.
Particularly, the invention also provides a fault feature self-adaptive extraction system based on wavelet entropy and EEMD, which comprises:
the natural mode decomposition module is used for performing natural mode decomposition on the acquired original signal by using the EEMD, sequentially expanding the original signal from high frequency to low frequency while adding white noise for noise reduction, and decomposing to obtain a plurality of IMF components;
the wavelet energy entropy calculation module is used for calculating the wavelet energy entropy of each IMF component according to wavelet packet decomposition;
the IMF component extraction module is used for extracting IMF components with most complex information according to the wavelet energy entropy;
the denoising module is used for performing wavelet threshold denoising on the extracted IMF component with the most complex information;
and the fault characteristic frequency band section extraction module is used for performing signal accumulation recombination on the IMF components subjected to noise reduction to obtain a brand new fault characteristic frequency band section.
Each module has the same processing method as the extraction method.
In particular, the present invention also provides a terminal device, comprising: the adaptive signal characteristic frequency band extraction method based on the wavelet entropy and the EEMD is implemented by the aid of a memory, a processor and a computer program which is stored in the memory and can run on the processor.
The present invention further provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the above-mentioned adaptive signal characteristic frequency band extraction method based on wavelet entropy and EEMD.
It is to be noted that the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention, and all of the technical solutions are covered in the protective scope of the present invention.
Claims (10)
1. A fault feature self-adaptive extraction method based on wavelet entropy and EEMD is characterized by comprising the following steps:
step 1, performing inherent modal decomposition on an acquired original signal by using EEMD (ensemble empirical mode decomposition), adding white noise and reducing noise, and simultaneously sequentially expanding the original signal from high frequency to low frequency to obtain a plurality of IMF (intrinsic mode function) components by decomposition;
step 2, according to wavelet packet decomposition, calculating the wavelet energy entropy of each IMF component obtained in the decomposition in the step 1;
step 3, extracting IMF components with most complex information according to the wavelet energy entropy calculated in the step 2;
step 4, performing wavelet threshold denoising on the IMF component extracted in the step 3;
and 5, performing signal accumulation recombination on the IMF component subjected to noise reduction in the step 4 to obtain a brand-new fault characteristic frequency band section.
2. The wavelet entropy and EEMD-based fault feature adaptive extraction method according to claim 1, wherein the specific process of step 1 is as follows:
setting the total number of decomposition iterations as NE, and initializing the amplitude of the added white noise;
step 1-1, decomposing the original signal added with white noise m at mth time, where m is 1,2, …, NE, and initially, where m is 1, and adding white noise to the original signal x (t):
Xm(t)=X(t)+nm(t)
wherein, Xm(t) is the signal after the mth time of noise addition; n ism(t) white noise added for the mth decomposition;
step 1-2, decomposing the m-th denoised signal X by using an empirical mode methodm(t) obtaining the ith IMF component IMFiI ═ 1,2, …, NE; if m < NE, returning to step 1), and m ═ m +1, repeating step 1-1 and step 1-2;
calculating NE IMF components and signal residual r obtained after NE decomposition;
3. the wavelet entropy and EEMD-based fault feature adaptive extraction method according to claim 2, wherein the specific process of step 2 is as follows:
selecting the number of wavelet packet decomposition layers as N, and expressing the reconstruction coefficient of the original signal X (t) at the j-th node of the k-th layer as D ═ Dk(j),j=1,2,…,2k0,1,2, …, where N, j is a node corresponding to each IMF component of the kth layer;
computing wavelet energies E for node subbands of respective IMF componentsk,j:
Ek,j=|dk(j)|2
Calculating the ratio P of wavelet energy of node sub-band segment of each IMF componentj:
Calculating the wavelet energy entropy W of each IMF component obtained by decompositionE:
4. The wavelet entropy and EEMD-based fault feature adaptive extraction method according to claim 1, wherein step 3 comprises: comparing and analyzing the energy change of the wavelet energy entropy by a method of calculating and accumulating the wavelet energy entropy; and extracting IMF components with wavelet energy entropy having larger turning points.
5. The wavelet entropy and EEMD-based fault feature adaptive extraction method according to claim 1, wherein step 4 comprises: carrying out wavelet fixed threshold denoising on the extracted IMF component; selecting wavelet fixed threshold denoising according to denoising performance, selecting noise level estimation according to the first layer wavelet decomposition for adjustment, setting the decomposition layer number to be 3, and selecting db3 as a wavelet base.
6. The wavelet entropy and EEMD-based fault feature adaptive extraction method according to claim 1, wherein the signal accumulation and recombination in step 5 is to perform signal superposition on IMF components subjected to wavelet threshold denoising in step 4:
wherein X '(t) is a reconstructed fault frequency band signal, IMF'iAnd de-noising the ith IMF component of the wavelet.
7. The adaptive wavelet entropy and EEMD-based fault feature extraction method according to claim 1, wherein in step 1, the original signal comprises any section of historical signal from the beginning of use to complete fault of current equipment or equipment of the same type and working environment as the current equipment.
8. A fault feature self-adaptive extraction system based on wavelet entropy and EEMD is characterized by comprising the following steps:
the natural mode decomposition module is used for performing natural mode decomposition on the acquired original signal by using the EEMD, sequentially expanding the original signal from high frequency to low frequency while adding white noise for noise reduction, and decomposing to obtain a plurality of IMF components;
the wavelet energy entropy calculation module is used for calculating the wavelet energy entropy of each IMF component according to wavelet packet decomposition;
the IMF component extraction module is used for extracting IMF components with most complex information according to the wavelet energy entropy;
the denoising module is used for performing wavelet threshold denoising on the extracted IMF component with the most complex information;
and the fault characteristic frequency band section extraction module is used for performing signal accumulation recombination on the IMF components subjected to noise reduction to obtain a brand new fault characteristic frequency band section.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the computer program is executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011215620.9A CN113158714B (en) | 2020-11-04 | 2020-11-04 | Fault feature self-adaptive extraction method based on wavelet entropy and EEMD |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011215620.9A CN113158714B (en) | 2020-11-04 | 2020-11-04 | Fault feature self-adaptive extraction method based on wavelet entropy and EEMD |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113158714A true CN113158714A (en) | 2021-07-23 |
CN113158714B CN113158714B (en) | 2024-04-16 |
Family
ID=76882326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011215620.9A Active CN113158714B (en) | 2020-11-04 | 2020-11-04 | Fault feature self-adaptive extraction method based on wavelet entropy and EEMD |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113158714B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114034381A (en) * | 2021-11-12 | 2022-02-11 | 广东电网有限责任公司江门供电局 | Distribution transformer vibration extraction method and system based on wavelet information entropy |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105093066A (en) * | 2015-08-12 | 2015-11-25 | 华北电力大学 | Line fault judgment method based on wavelet analysis and support vector machine |
CN105115594A (en) * | 2015-10-09 | 2015-12-02 | 北京航空航天大学 | Gearbox vibration signal fault feature extraction method based on wavelet entropy and information fusion |
CN106405339A (en) * | 2016-11-11 | 2017-02-15 | 中国南方电网有限责任公司 | Power transmission line fault reason identification method based on high and low frequency wavelet feature association |
CN108338784A (en) * | 2017-01-25 | 2018-07-31 | 中国科学院半导体研究所 | The Denoising of ECG Signal of wavelet entropy threshold based on EEMD |
CN108562828A (en) * | 2018-03-14 | 2018-09-21 | 哈尔滨理工大学 | The method for improving electrical network low voltage ride-through capability based on Wavelet Detection |
CN111161171A (en) * | 2019-12-18 | 2020-05-15 | 三明学院 | Blasting vibration signal baseline zero drift correction and noise elimination method, device, equipment and system |
CN111721527A (en) * | 2020-05-18 | 2020-09-29 | 浙江工业大学 | Wind generating set gearbox fault positioning method based on CMS system big data combined standard deviation and wavelet entropy |
-
2020
- 2020-11-04 CN CN202011215620.9A patent/CN113158714B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105093066A (en) * | 2015-08-12 | 2015-11-25 | 华北电力大学 | Line fault judgment method based on wavelet analysis and support vector machine |
CN105115594A (en) * | 2015-10-09 | 2015-12-02 | 北京航空航天大学 | Gearbox vibration signal fault feature extraction method based on wavelet entropy and information fusion |
CN106405339A (en) * | 2016-11-11 | 2017-02-15 | 中国南方电网有限责任公司 | Power transmission line fault reason identification method based on high and low frequency wavelet feature association |
CN108338784A (en) * | 2017-01-25 | 2018-07-31 | 中国科学院半导体研究所 | The Denoising of ECG Signal of wavelet entropy threshold based on EEMD |
CN108562828A (en) * | 2018-03-14 | 2018-09-21 | 哈尔滨理工大学 | The method for improving electrical network low voltage ride-through capability based on Wavelet Detection |
CN111161171A (en) * | 2019-12-18 | 2020-05-15 | 三明学院 | Blasting vibration signal baseline zero drift correction and noise elimination method, device, equipment and system |
CN111721527A (en) * | 2020-05-18 | 2020-09-29 | 浙江工业大学 | Wind generating set gearbox fault positioning method based on CMS system big data combined standard deviation and wavelet entropy |
Non-Patent Citations (1)
Title |
---|
史历程等: "基于小波能谱熵和集成经验模态分解的 传感器故障诊断耦合算法研究", 动力工程学报, vol. 8, no. 8, pages 624 - 632 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114034381A (en) * | 2021-11-12 | 2022-02-11 | 广东电网有限责任公司江门供电局 | Distribution transformer vibration extraction method and system based on wavelet information entropy |
CN114034381B (en) * | 2021-11-12 | 2023-10-10 | 广东电网有限责任公司江门供电局 | Distribution transformer vibration extraction method and system based on wavelet information entropy |
Also Published As
Publication number | Publication date |
---|---|
CN113158714B (en) | 2024-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jin et al. | A novel adaptive EEMD method for switchgear partial discharge signal denoising | |
CN106353649B (en) | A kind of partial discharge signal denoising method based on lifting wavelet transform | |
CN102930149B (en) | Sensor network sensing information denoising processing method based on principal component analysis (PCA) and empirical mode decomposition (EMD) | |
Dash et al. | Mining for similarities in time series data using wavelet-based feature vectors and neural networks | |
CN110717472A (en) | Fault diagnosis method and system based on improved wavelet threshold denoising | |
CN112486137A (en) | Method and system for constructing fault feature library of active power distribution network and fault diagnosis method | |
CN113158714A (en) | Fault feature self-adaptive extraction method based on wavelet entropy and EEMD | |
Bardet | Testing for the presence of self‐similarity of Gaussian time series having stationary increments | |
Dong et al. | Multi‐task learning method for classification of multiple power quality disturbances | |
Lee | Improvement of decomposing results of empirical mode decomposition and its variations for sea-level records analysis | |
Elshorbagy et al. | Analysis of cross-correlated chaotic streamflows | |
CN113188651B (en) | Vibration signal feature extraction and tool wear value prediction method based on IMF-PSD spectral line | |
CN116667920A (en) | Transient signal detection method combining principal component analysis and Hilbert-Huang transform | |
CN116225752A (en) | Fault root cause analysis method and system for micro-service system based on fault mode library | |
Gholizade-Narm et al. | Using repetitive fuzzy method for chaotic time series prediction | |
CN114441111A (en) | Pipeline leakage infrasonic wave signal analysis method and system | |
CN113218433A (en) | Sensor fault detection and data restoration method based on time-varying graph signal processing | |
CN113217102A (en) | Acoustic emission signal feature extraction method, acoustic emission signal feature identification device and storage medium | |
Li et al. | Chaotic frequency hopping prediction based on temporal convolutional network | |
Tahani et al. | Multiple fault detection based on wavelet denoising: Application on wind turbine system | |
Lei et al. | A genetic training algorithm of wavelet neural networks for fault prognostics in condition based maintenance | |
CN117370843A (en) | Method, device and equipment for identifying life body electric shock current | |
CN116719085B (en) | High-resolution processing method, device and equipment for seismic records and storage medium | |
Ma et al. | Novel noise reduction method based on improved empirical wavelet transform and kurtosis for partial discharge signal of high‐voltage cables | |
Kanirajan et al. | A Comparison of Back propagation and PSO for training RBF Neural Network for Wavelet based Detection and Classification of Power Quality Disturbances |
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 |