CN107290147B - The instantaneous Frequency Estimation method examined based on non-delayed cost function and t - Google Patents
The instantaneous Frequency Estimation method examined based on non-delayed cost function and t Download PDFInfo
- Publication number
- CN107290147B CN107290147B CN201710609997.4A CN201710609997A CN107290147B CN 107290147 B CN107290147 B CN 107290147B CN 201710609997 A CN201710609997 A CN 201710609997A CN 107290147 B CN107290147 B CN 107290147B
- Authority
- CN
- China
- Prior art keywords
- cost function
- value
- moment
- mean curve
- instantaneous frequency
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/028—Acoustic or vibration analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
-
- 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
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Abstract
The invention discloses the instantaneous Frequency Estimation methods based on non-delayed cost function and t inspection.Original signal is converted to time-frequency spectrum using Short Time Fourier Transform by this method, a plurality of ridge band is obtained using Canny detection algorithm, the exceptional value for excluding every vallate band is examined using t, pass through one synthesis ridge band with complete sharp edge of superposition building, the exceptional value for excluding synthesis ridge band is examined using t, calculate the Mean curve of synthesis ridge band, Mean curve is smoothed, calculate the confidence interval of the smooth Mean curve in 95% confidence level, smooth Mean curve and its confidence interval are mapped on target crestal line, obtain the reference line and the part region of search of target crestal line, target crestal line is extracted using non-delayed cost function.The present invention is suitable for estimating the instantaneous frequency of complicated multi -components frequency variation signal, overcomes conventional method in the defects of mechanical oscillation signal instantaneous Frequency Estimation, the accuracy and precision of estimated result is high, is convenient for engineer application.
Description
Technical field
The present invention relates to condition monitoring for rotating machinery and fault diagnosis field, and in particular to based on non-delayed cost function and
The instantaneous Frequency Estimation method that t is examined.
Background technique
Due to the complexity of working environment, rotating machinery often works under Variable Velocity Condition.Instantaneous Frequency Estimation is assessment
Running state of rotating machine and the important prerequisite for carrying out fault diagnosis.Currently used instantaneous Frequency Estimation method is a step cost
Function method (one-step cost function).One step cost function method can search for ridge point within the scope of local frequencies, but
It is that the central point of local frequencies range depends on the position of a upper ridge point, this leads to a step cost function, and there is delays.This
Outside, the width of local frequencies range is rule of thumb arbitrarily arranged, and width at any time is fixed, Bu Nengsui
The time change, this causes a step cost function to lack enough adaptivitys.Drawbacks described above leads to a step cost function
Method accuracy and precision when estimating instantaneous frequency is lower.
Summary of the invention
The problem to be solved in the present invention is against the above deficiency, to propose to examine based on non-delayed cost function and t instantaneous
Frequency estimating methods.Compared with the conventional method, the present invention is using the smooth Mean curve after mapping as the reference line of target crestal line,
Using the confidence interval after mapping as the local search section of target crestal line, therefore the central point of local frequencies search range is disobeyed
The position for relying a ridge point, without any delay, local frequencies search range can be set automatically, and search bandwidth can be with
The variation of time and change automatically, the accuracy and precision of instantaneous Frequency Estimation result is high.
In order to solve the above technical problems, the present invention provides the instantaneous Frequency Estimation examined based on non-delayed cost function and t
Method, which comprises the following steps:
Step 1: signal x(k) (k=1,2 ..., N) is converted to by time-frequency spectrum, N using Short Time Fourier Transform algorithm
The length of representation signal；
Step 2: one piece of regional area having compared with high s/n ratio is chosen from time-frequency spectrum, using Canny detection algorithm
The regional area is converted into bianry image, bianry image includes a plurality of ridge band；Regional area refers to including at least two vallate bands,
Signal-to-noise ratio is greater than the region of 80dB；
Step 3: exceptional value of every vallate with lower edges is excluded using t check algorithm；
Step 4: above-mentioned a plurality of ridge band is added to wherein according to mutual kinematics proportionate relationship, and a profile is most
On complete ridge band, one synthesis ridge band with complete sharp edge of building；Kinematics proportionate relationship refers to corresponding to ridge band
Machine part between transmission ratio；
Step 5: above-mentioned exceptional value of the synthesis ridge with lower edges is excluded using t check algorithm；
Step 6: calculate the Mean curve of above-mentioned synthesis ridge band, using 5 points three times smoothing algorithm Mean curve is carried out it is flat
Sliding processing, obtains smooth Mean curve, calculates the confidence interval of the smooth Mean curve in 95% confidence level；
Step 7: by above-mentioned smooth Mean curve and its confidence interval according to smooth Mean curve and target crestal line to be estimated
Between kinematics proportionate relationship be mapped on target crestal line；
Step 8: using the smooth Mean curve after mapping as the reference line of target crestal line, the confidence interval after mapping being made
For the local search section of target crestal line；
Step 9: ridge point being searched in the local search section corresponding to each moment using non-delayed cost function, is determined
Instantaneous frequency corresponding to each moment finally obtains the instantaneous frequency on entire time interval.
Further, in the step 1 Short Time Fourier Transform algorithm the following steps are included:
1) Short Time Fourier Transform is carried out to signal x(k):
,
TF (t, f) representation signal x(k) Short Time Fourier Transform as a result, t represents time factor, f represents scale factor,
Function w (z) represents independent variable as the window function of z；
2) calculate signal x(k) time-frequency spectrum:
,
Spectrogram (t, f) represents x(k) time-frequency spectrum.
Further, in the step 2 Canny detection algorithm the following steps are included:
1) image f (x, y) is smoothed using Gaussian filter, eliminates noise in image and unrelated thin
Section:
,
,
G (x, y) represents smoothed out image, and G (x, y) represents 2-d gaussian filters device, the time point of x representative image, y
The Frequency point of representative image, symbol * represent convolutional calculation, and it is poor that σ represents Gauss standard；In the present invention, σ=1；
2) amplitude and direction of g (x, y) intensity gradient are calculated
,
,
,
M (x, y) represents the amplitude of intensity gradient, and θ (x, y) represents the direction of intensity gradient, gx(x, y) represent g (x,
Y) to the partial derivative of x, gy(x, y) represents g (x, y) to the partial derivative of y；
3) false edge is eliminated using non-maximum value coercion Acts:
On gradient direction θ (x, y), if non-zero gradient value is greater than two adjacent gradient values, the non-zero gradient value
It remains unchanged, otherwise, the non-zero gradient value zero setting；
4) image after above-mentioned elimination false edge is filtered using two threshold values of different sizes, two threshold values
It is denoted as T1And T2, T1<T2, by T1Obtained image is denoted as I1, by T2Obtained image is denoted as I2；In the present invention, T1=0.0063, T2
=0.0156；
5) from I2It is middle to reject the weak edge not connecting with strong edge, then connect I1And I2In edge form continuous side
Edge.
Further, in the step 3 t check algorithm the following steps are included:
1) estimate signal xnThe standard deviation of (n=1,2 ..., N),
,
Representative sample mean value, σ representative sample standard deviation, N representative sample length；
If 2), then x is rejectedn；tσ(N-1) standard deviation is represented as σ, and freedom degree is (N-1)
T distribution.
Further, in the step 9 non-delayed cost function the following steps are included:
1) local search section FB corresponding to k-th of momentkIt is defined as
,
fk(pmc) value of the smooth Mean curve k-th of moment after representing mapping,It is smooth equal after representing mapping
It is worth curve confidence interval in the half of k-th of moment width, m represents the length of target crestal line；
2) non-delayed cost function CF corresponding to k-th of momentkIs defined as:
,
,
fk(i) it represents in FBkThe frequency values taken in range, TF (tk, fk) TF (t, f) is represented k-th moment
Value, tkT is represented in the value at k-th of moment, fkF is represented in the value at k-th of moment, ekRepresent weight factor.
Further, relative error≤0.682%, average relative error≤0.066%.
The invention adopts the above technical scheme, compared with prior art, the invention has the following advantages that
1) present invention has real-time: the present invention is mapped as reference line to synthesize the smooth Mean curve of ridge band, can be instant
The central point for determining current time local frequencies search range avoids the dependence to previous ridge point, eliminates time delay,
With real-time.
2) present invention has adaptivity: the present invention is mentioned using the smooth Mean curve confidence interval mapping of synthesis ridge band
The subrange of confession, can adaptively determine local frequencies search range corresponding to each moment, and search bandwidth can be with
The variation of time and change automatically, do not need that search bandwidth is by virtue of experience arranged, to eliminate due to artificial origin and generate
Error.
3) the results showed that it is opposite accidentally by the maximum between obtained instantaneous Frequency Estimation value of the invention and measured value
Difference is 0.682%, average relative error 0.066%；Compared with the result of a step cost function method, maximum relative error is reduced
95.84%, average relative error reduces by 96.92%.
The present invention will be further described with reference to the accompanying drawings and examples.
Detailed description of the invention
Attached drawing 1 is the stream based on non-delayed cost function and t the instantaneous Frequency Estimation method examined in the embodiment of the present invention
Cheng Tu；
Attached drawing 2 is epicyclic gearbox vibration signal in the embodiment of the present invention；
Attached drawing 3 is the time-frequency spectrum of epicyclic gearbox vibration signal in the embodiment of the present invention；
Attached drawing 4 is the regional area with high s/n ratio chosen from time-frequency spectrum in the embodiment of the present invention；
Attached drawing 5 is the edge of the local image region detected in the embodiment of the present invention by Canny algorithm；
Attached drawing 6 is to eliminate every vallate with the result after abnormal point using t check algorithm in the embodiment of the present invention；
Attached drawing 7 is the synthesis ridge band being formed by stacking in the embodiment of the present invention using the kinematics proportionate relationship between ridge band
(undermost ridge band is to synthesize ridge band)；
Attached drawing 8 is to eliminate synthesis ridge with the result after abnormal point using t check algorithm in the embodiment of the present invention；
Attached drawing 9 is the mean value smoothing curve and its 95% confidence interval that ridge band is synthesized in the embodiment of the present invention；
Attached drawing 10 is the mean value smoothing curve mapped in the embodiment of the present invention and its confidence interval；
Attached drawing 11 is instantaneous Frequency Estimation value in the embodiment of the present invention.
Specific embodiment
Embodiment, as shown in Figure 1, based on the instantaneous Frequency Estimation method that non-delayed cost function and t are examined, including it is following
Step:
Step 1: signal x(k) (k=1,2 ..., N) is converted to by time-frequency spectrum, N using Short Time Fourier Transform algorithm
The length of representation signal；
Step 2: one piece of regional area having compared with high s/n ratio is chosen from time-frequency spectrum, using Canny detection algorithm
The regional area is converted into bianry image, bianry image includes a plurality of ridge band；Regional area refers to including at least two vallate bands,
Signal-to-noise ratio is greater than the region of 80dB；
Step 3: exceptional value of every vallate with lower edges is excluded using t check algorithm；
Step 4: above-mentioned a plurality of ridge band is added to wherein according to mutual kinematics proportionate relationship, and a profile is most
On complete ridge band, one synthesis ridge band with complete sharp edge of building；Kinematics proportionate relationship refers to corresponding to ridge band
Machine part between transmission ratio；
Step 5: above-mentioned exceptional value of the synthesis ridge with lower edges is excluded using t check algorithm；
Step 6: calculate the Mean curve of above-mentioned synthesis ridge band, using 5 points three times smoothing algorithm Mean curve is carried out it is flat
Sliding processing, obtains smooth Mean curve, calculates the confidence interval of the smooth Mean curve in 95% confidence level；
Step 7: by above-mentioned smooth Mean curve and its confidence interval according to smooth Mean curve and target crestal line to be estimated
Between kinematics proportionate relationship be mapped on target crestal line；
Step 8: using the smooth Mean curve after mapping as the reference line of target crestal line, the confidence interval after mapping being made
For the local search section of target crestal line；
Step 9: ridge point being searched in the local search section corresponding to each moment using non-delayed cost function, is determined
Instantaneous frequency corresponding to each moment finally obtains the instantaneous frequency on entire time interval.
Short Time Fourier Transform algorithm in step 1 the following steps are included:
1) Short Time Fourier Transform is carried out to signal x(k):
,
TF (t, f) representation signal x(k) Short Time Fourier Transform as a result, t represents time factor, f represents scale factor,
Function w (z) represents independent variable as the window function of z；
2) calculate signal x(k) time-frequency spectrum:
,
Spectrogram (t, f) represents x(k) time-frequency spectrum.
Canny detection algorithm in step 2 the following steps are included:
1) image f (x, y) is smoothed using Gaussian filter, eliminates noise in image and unrelated thin
Section:
,
,
G (x, y) represents smoothed out image, and G (x, y) represents 2-d gaussian filters device, the time point of x representative image, y
The Frequency point of representative image, symbol * represent convolutional calculation, and it is poor that σ represents Gauss standard；In the present invention, σ=1；
2) amplitude and direction of g (x, y) intensity gradient are calculated
,
,
,
M (x, y) represents the amplitude of intensity gradient, and θ (x, y) represents the direction of intensity gradient, gx(x, y) represent g (x,
Y) to the partial derivative of x, gy(x, y) represents g (x, y) to the partial derivative of y；
3) false edge is eliminated using non-maximum value coercion Acts:
On gradient direction θ (x, y), if non-zero gradient value is greater than two adjacent gradient values, the non-zero gradient value
It remains unchanged, otherwise, the non-zero gradient value zero setting；
4) image after above-mentioned elimination false edge is filtered using two threshold values of different sizes, two threshold values
It is denoted as T1And T2, T1<T2, by T1Obtained image is denoted as I1, by T2Obtained image is denoted as I2；In the present invention, T1=0.0063, T2
=0.0156；
5) from I2It is middle to reject the weak edge not connecting with strong edge, then connect I1And I2In edge form continuous side
Edge.
T check algorithm in step 3 the following steps are included:
1) estimate signal xnThe standard deviation of (n=1,2 ..., N),
,
Representative sample mean value, σ representative sample standard deviation, N representative sample length；
If 2), then x is rejectedn；tσ(N-1) standard deviation is represented as σ, and freedom degree is (N-
1) t distribution.
Non-delayed cost function in step 9 the following steps are included:
1) local search section FB corresponding to k-th of momentkIt is defined as
,
fk(pmc) value of the smooth Mean curve k-th of moment after representing mapping,It is smooth equal after representing mapping
It is worth curve confidence interval in the half of k-th of moment width, m represents the length of target crestal line；
2) non-delayed cost function CF corresponding to k-th of momentkIs defined as:
,
,
fk(i) it represents in FBkThe frequency values taken in range, TF (tk, fk) TF (t, f) is represented k-th moment
Value, tkT is represented in the value at k-th of moment, fkF is represented in the value at k-th of moment, ekRepresent weight factor.
The performance of algorithm of the present invention is verified using blower turbine epicyclic gearbox vibration data.
Vibration data is acquired from the gearbox-case of planetary gear train, data length N=2736825, sample frequency
fs= 5000 Hz。
Collected epicyclic gearbox vibration data is as shown in Figure 2.
Epicyclic gearbox vibration data shown in Fig. 2 is converted to by time-frequency spectrum using Short Time Fourier Transform algorithm, is obtained
The time-frequency spectrum arrived is as shown in Figure 3.
The regional area with high s/n ratio, obtained regional area such as Fig. 4 institute are chosen from time-frequency spectrum shown in Fig. 3
Show.
Edge detection is carried out to regional area as shown in Figure 4 using Canny detection algorithm, obtained image border is as schemed
Shown in 5.
The abnormal point of each vallate band in Fig. 5 is eliminated using t check algorithm, obtained result is as shown in Figure 6.
Each vallate band is added to the wherein most complete ridge band of a profile according to the kinematics proportionate relationship between ridge band
On, constructed synthesis ridge band is as shown in Figure 7 (undermost ridge band is to synthesize ridge band).
The abnormal point of synthesis ridge band is eliminated using t check algorithm, as a result as shown in Figure 8.
The smooth Mean curve of synthesis ridge band and its 95% confidence interval are calculated, as a result as shown in Figure 9.
According to the kinematics proportionate relationship between smooth Mean curve and target crestal line by smooth Mean curve and its confidence
On Interval Maps to target crestal line, the results are shown in Figure 10.
Using the ridge point of non-delayed cost function search target crestal line, obtained instantaneous frequency profile is as shown in figure 11.
Show through many experiments by the maximum relative error between obtained instantaneous Frequency Estimation value of the invention and measured value
It is 0.682%, average relative error 0.066%, and the instantaneous Frequency Estimation value and actual measurement for using a step cost function method to obtain
Maximum relative error between value is 16.39%, and average relative error 2.14%, maximum relative error of the present invention reduces
95.84%, average relative error reduces by 96.92%.
According to experimental result, think after analysis:
1) a traditional step cost function needs to rely on when determining the central point in current search section a upper ridge point
Position, there is time delay phenomenon, the present invention can be determined immediately using the smooth Mean curve after mapping as reference line
The center in current search section does not depend on a ridge point completely, therefore has real-time.
2) a traditional step cost function method lacks adaptivity, needs artificially to be arranged the region of search, and search width
It is fixed, thus inevitably brings error, the present invention is using the smooth Mean curve confidence interval after mapping come automatic
Determine local search section, search bandwidth can change automatically with the variation of time, not need manually to participate in, therefore have
Adaptivity.
3) compared with a traditional step cost function method, accuracy of the present invention and accuracy are high.
One skilled in the art would recognize that the above specific embodiments are only exemplary, it is to make ability
Field technique personnel can better understand the content of present invention, should not be construed as limiting the scope of protection of the present invention, as long as
Technical solution improvements introduced according to the present invention each falls within protection scope of the present invention.
Claims (4)
1. the instantaneous Frequency Estimation method examined based on non-delayed cost function and t, which comprises the following steps:
Step 1: signal x(k) k=1,2 ..., N being converted to by time-frequency spectrum using Short Time Fourier Transform algorithm, N represents letter
Number length；
Step 2: one piece of regional area having compared with high s/n ratio is chosen from time-frequency spectrum, it should using Canny detection algorithm
Regional area is converted into bianry image, and bianry image includes a plurality of ridge band；
Step 3: exceptional value of every vallate with lower edges is excluded using t check algorithm；
Step 4: above-mentioned a plurality of ridge band is most complete according to the mutual kinematics proportionate relationship wherein profile that is added to
Ridge band on, synthesis ridge band of the building one with complete sharp edge；
Step 5: above-mentioned exceptional value of the synthesis ridge with lower edges is excluded using t check algorithm；
Step 6: calculate the Mean curve of above-mentioned synthesis ridge band, using 5 points three times smoothing algorithm Mean curve is smoothly located
Reason, obtains smooth Mean curve, calculates the confidence interval of the smooth Mean curve in 95% confidence level；
Step 7: by above-mentioned smooth Mean curve and its confidence interval according between smooth Mean curve and target crestal line to be estimated
Kinematics proportionate relationship be mapped on target crestal line；
Step 8: using the smooth Mean curve after mapping as the reference line of target crestal line, using the confidence interval after mapping as mesh
Mark the local search section of crestal line；
Step 9: ridge point being searched for using non-delayed cost function in the local search section corresponding to each moment, is determined each
Instantaneous frequency corresponding to moment finally obtains the instantaneous frequency on entire time interval；
The calculating step of non-delayed cost function includes: in the step 9
1) local search section FB corresponding to k-th of momentkIt is defined as
,
fk(pmc) value of the smooth Mean curve k-th of moment after representing mapping,Smooth mean value after representing mapping is bent
For line confidence interval in the half of k-th of moment width, m represents the length of target crestal line；
2) non-delayed cost function CF corresponding to k-th of momentkIs defined as:
,
,
fk(i) it represents in FBkThe frequency values taken in range, TF (tk, fk) TF (t, f) is represented in the value at k-th of moment, tk
T is represented in the value at k-th of moment, fkF is represented in the value at k-th of moment, ekRepresent weight factor.
2. the instantaneous Frequency Estimation method according to claim 1 examined based on non-delayed cost function and t, feature are existed
In, Short Time Fourier Transform algorithm in the step 1 the following steps are included:
1) Short Time Fourier Transform is carried out to signal x(k):
,
TF (t, f) representation signal x(k) Short Time Fourier Transform as a result, t represents time factor, f represents scale factor, function
W (z) represents independent variable as the window function of z；
2) calculate signal x(k) time-frequency spectrum:
,
Spectrogram (t, f) represents x(k) time-frequency spectrum.
3. the instantaneous Frequency Estimation method according to claim 1 examined based on non-delayed cost function and t, feature are existed
In, Canny detection algorithm in the step 2 the following steps are included:
1) image f (x, y) is smoothed using Gaussian filter, eliminates the noise and Extraneous details in image:
,
,
G (x, y) represents smoothed out image, and G (x, y) represents 2-d gaussian filters device, the time point of x representative image, and y is represented
The Frequency point of image, symbol * represent convolutional calculation, and it is poor that σ represents Gauss standard；
2) amplitude and direction of g (x, y) intensity gradient are calculated
,
,
,
M (x, y) represents the amplitude of intensity gradient, and θ (x, y) represents the direction of intensity gradient, gx(x, y) represents g (x, y) to x
Partial derivative, gy(x, y) represents g (x, y) to the partial derivative of y；
3) false edge is eliminated using non-maximum value coercion Acts:
On gradient direction θ (x, y), if non-zero gradient value is greater than two adjacent gradient values, which is kept
It is constant, otherwise, the non-zero gradient value zero setting；
4) image after above-mentioned elimination false edge is filtered using two threshold values of different sizes, two threshold values are denoted as T1
And T2, T1<T2, by T1Obtained image is denoted as I1, by T2Obtained image is denoted as I2；
5) from I2It is middle to reject the weak edge not connecting with strong edge, then connect I1And I2In edge formed continuous boundary.
4. the instantaneous Frequency Estimation method according to claim 1 examined based on non-delayed cost function and t, feature are existed
In: t check algorithm in the step 3 the following steps are included:
1) estimate signal xnThe standard deviation of n=1,2 ..., N,
,
Representative sample mean value, σ representative sample standard deviation, N representative sample length；
If 2), then x is rejectedn；tσ(N-1) standard deviation is represented as σ, and freedom degree is (N-1)
T distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609997.4A CN107290147B (en) | 2017-07-25 | 2017-07-25 | The instantaneous Frequency Estimation method examined based on non-delayed cost function and t |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609997.4A CN107290147B (en) | 2017-07-25 | 2017-07-25 | The instantaneous Frequency Estimation method examined based on non-delayed cost function and t |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107290147A CN107290147A (en) | 2017-10-24 |
| CN107290147B true CN107290147B (en) | 2019-11-01 |
Family
ID=60102631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710609997.4A Active CN107290147B (en) | 2017-07-25 | 2017-07-25 | The instantaneous Frequency Estimation method examined based on non-delayed cost function and t |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107290147B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5955674A (en) * | 1997-10-31 | 1999-09-21 | Eaton Corporation | Driveline vibration system diagnostics |
| CN101587017A (en) * | 2009-06-19 | 2009-11-25 | 湖南大学 | Gear fault diagnosis method based on part mean decomposition cycle frequency spectrum |
| CN103154932A (en) * | 2010-08-11 | 2013-06-12 | 伊夫·列扎 | Method for analyzing signals providing instantaneous frequencies and sliding fourier transforms, and device for analyzing signals |
| CN104215459A (en) * | 2014-08-26 | 2014-12-17 | 中国直升机设计研究所 | Bearing fault diagnosis method |
| CN105067262A (en) * | 2015-08-04 | 2015-11-18 | 潍坊学院 | Rolling bearing state monitoring method |
| CN105092239A (en) * | 2014-05-09 | 2015-11-25 | 潍坊学院 | Method for detecting early stage fault of gear |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6829537B2 (en) * | 2003-03-10 | 2004-12-07 | Conocophillips Company | Contour-based attribute seismic mapping |
| CN101672646B (en) * | 2009-10-09 | 2011-11-23 | 国家卫星气象中心 | Satellite-image automatic, geometric and precise correction method |
| CN104749432B (en) * | 2015-03-12 | 2017-06-16 | 西安电子科技大学 | Based on the multi -components non-stationary signal instantaneous Frequency Estimation method for focusing on S-transformation |
-
2017
- 2017-07-25 CN CN201710609997.4A patent/CN107290147B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5955674A (en) * | 1997-10-31 | 1999-09-21 | Eaton Corporation | Driveline vibration system diagnostics |
| CN101587017A (en) * | 2009-06-19 | 2009-11-25 | 湖南大学 | Gear fault diagnosis method based on part mean decomposition cycle frequency spectrum |
| CN103154932A (en) * | 2010-08-11 | 2013-06-12 | 伊夫·列扎 | Method for analyzing signals providing instantaneous frequencies and sliding fourier transforms, and device for analyzing signals |
| CN105092239A (en) * | 2014-05-09 | 2015-11-25 | 潍坊学院 | Method for detecting early stage fault of gear |
| CN104215459A (en) * | 2014-08-26 | 2014-12-17 | 中国直升机设计研究所 | Bearing fault diagnosis method |
| CN105067262A (en) * | 2015-08-04 | 2015-11-18 | 潍坊学院 | Rolling bearing state monitoring method |
Non-Patent Citations (1)
| Title |
|---|
| 基于EEMD和Hilbert变换的齿轮箱故障诊断;林近山;《机械传动》;20101231;第34卷(第5期);第62-64页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107290147A (en) | 2017-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107368457B (en) | The instantaneous Frequency Estimation method examined based on LoG operator and Grubbs | |
| TW201300742A (en) | Abnormal sound diagnostic equipment | |
| CN108463630B (en) | Method for assessing the incident flow on a rotor blade of a wind turbine, and method for controlling a wind turbine and wind turbine | |
| CN111664929B (en) | System for separating periodic amplitude peaks and non-periodic amplitude peaks in machine vibration data | |
| CN107391935B (en) | The instantaneous Frequency Estimation method examined based on non-delayed cost function and Grubbs | |
| US9431024B1 (en) | Method and apparatus for detecting noise of audio signals | |
| KR20110104953A (en) | Device for diagnosing clogging of lead pipe and method for diagnosing blockage | |
| CN107389342B (en) | The instantaneous Frequency Estimation method examined based on Roberts operator and Grubbs | |
| US20190346343A1 (en) | System for Separating Periodic Amplitude Peaks from Non-Periodic Amplitude Peaks in Machine Vibration Data | |
| Zhang et al. | Improved local cepstrum and its applications for gearbox and rolling bearing fault detection | |
| CN109883705A (en) | Motor rolling bearing part spot corrosion method for diagnosing faults and its diagnostic system | |
| CN107389329B (en) | The instantaneous Frequency Estimation method examined based on non-delayed cost function and PauTa | |
| CN104568132B (en) | Reference signal constraint-based mechanical characteristic acoustic signal frequency-domain semi-blind extraction method | |
| CN107368456B (en) | The instantaneous Frequency Estimation method examined based on Sobel operator and t | |
| CN107368814B (en) | The instantaneous Frequency Estimation method examined based on Sobel operator and PauTa | |
| CN107389341B (en) | The instantaneous Frequency Estimation method examined based on Prewitt operator and t | |
| CN107290147B (en) | The instantaneous Frequency Estimation method examined based on non-delayed cost function and t | |
| CN107357760B (en) | The instantaneous Frequency Estimation method examined based on Roberts operator and PauTa | |
| CN107391934B (en) | The instantaneous Frequency Estimation method examined based on Prewitt operator and PauTa | |
| CN107368458B (en) | The instantaneous Frequency Estimation method examined based on Sobel operator and Grubbs | |
| CN107340129B (en) | The instantaneous Frequency Estimation method examined based on LoG operator and PauTa | |
| CN107356429B (en) | The instantaneous Frequency Estimation method examined based on LoG operator and t | |
| WO2019007970A1 (en) | Method for automatic detection of physical modes in a modal analysis model | |
| CN107389343B (en) | The instantaneous Frequency Estimation method examined based on Roberts operator and t | |
| CN107368686B (en) | The instantaneous Frequency Estimation method examined based on Prewitt operator and Grubbs |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210817 Address after: 262100 No. 66 Yinfeng Road, high tech Zone, Weifang City, Shandong Province Patentee after: SHANDONG IPRE DETECTION TECHNOLOGY Co.,Ltd. Address before: 261061 No. 5147 Dongfeng East Street, Shandong, Weifang Patentee before: WEIFANG University |
|
| TR01 | Transfer of patent right |