CN113250911A - Fan blade fault diagnosis method based on VMD decomposition algorithm - Google Patents
Fan blade fault diagnosis method based on VMD decomposition algorithm Download PDFInfo
- Publication number
- CN113250911A CN113250911A CN202110525344.4A CN202110525344A CN113250911A CN 113250911 A CN113250911 A CN 113250911A CN 202110525344 A CN202110525344 A CN 202110525344A CN 113250911 A CN113250911 A CN 113250911A
- Authority
- CN
- China
- Prior art keywords
- sound signal
- signal
- short
- sound
- fault diagnosis
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 38
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000005236 sound signal Effects 0.000 claims abstract description 77
- 238000012706 support-vector machine Methods 0.000 claims abstract description 10
- 238000009432 framing Methods 0.000 claims abstract description 8
- 238000001914 filtration Methods 0.000 claims abstract description 5
- 238000010276 construction Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000037433 frameshift Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000011541 reaction mixture Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F05—INDEXING SCHEMES RELATING TO ENGINES OR PUMPS IN VARIOUS SUBCLASSES OF CLASSES F01-F04
- F05B—INDEXING SCHEME RELATING TO WIND, SPRING, WEIGHT, INERTIA OR LIKE MOTORS, TO MACHINES OR ENGINES FOR LIQUIDS COVERED BY SUBCLASSES F03B, F03D AND F03G
- F05B2260/00—Function
- F05B2260/80—Diagnostics
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a fan blade fault diagnosis method based on a VMD decomposition algorithm, which comprises the following steps of 1, collecting sound signals of a fan blade and filtering the sound signals to obtain filtered sound signals; 2, decomposing the filtered sound signal into a plurality of modal components by utilizing a VMD decomposition algorithm; step 3, performing framing processing on each modal component, and dividing each modal component into a plurality of frames of sound signals; windowing the sound signal subjected to the framing processing to obtain a sound signal subjected to windowing processing; step 4, calculating the short-time energy of each frame of sound signal, and then adding the short-time energy of all the frame of sound signals corresponding to each modal component to obtain the short-time energy of each modal component; and 5, forming a feature vector of the signal sample by the short-time energy of each modal component, inputting the feature vector into a support vector machine for training, and using the trained model for fault diagnosis. The method has high diagnosis accuracy and efficiency.
Description
Technical Field
The invention relates to the technical field of fault diagnosis, in particular to a fan blade fault diagnosis method based on a VMD decomposition algorithm.
Background
Since wind energy is a green and environmentally friendly renewable energy source, it has become a trend to vigorously develop and utilize wind energy for power generation against the background of energy shortage and increasingly important environmental requirements. In wind power equipment, blades are used as key parts of a wind turbine, and the wind turbine runs in a severe and variable environment for a long time, so that the fan blades are easy to crack, corrode and the like, and if the wind turbine is not timely found and maintained, the normal operation of the fan is affected, and huge economic loss is caused, so that fault diagnosis of the fan blades is very necessary.
At present, fan blade fault diagnosis methods are numerous, such as vibration diagnosis, fiber bragg grating, ultrasonic detection and unmanned detection, most of the methods need to install a sensor on the blade, stable work of the blade can be influenced, installation is difficult, cost is high, and practicability is poor.
Most of the existing vibration diagnosis methods are realized based on an EMD (empirical mode decomposition) algorithm, and feature information is extracted from vibration signals of the fan blades. The EMD algorithm decomposes nonlinear and non-stationary signals into a plurality of Intrinsic Mode Function (IMF) components with different frequency components, and has the characteristics of adaptivity, orthogonality, completeness and the like, but the recursive mode decomposition adopted by the EMD algorithm can cause an end point effect and a mode aliasing phenomenon, and the fault identification precision is influenced.
In conclusion, the method for diagnosing the fan blade fault by using the sound signal provided by the invention does not need to install a vibration sensor on the blade, has high detection efficiency and better application prospect, and can improve the diagnosis accuracy.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to solve the technical problem of providing a fan blade fault diagnosis method based on a VMD decomposition algorithm.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a fan blade fault diagnosis method based on a VMD decomposition algorithm is characterized by comprising the following steps:
step 1, collecting sound signals of a fan blade and filtering the sound signals to obtain filtered sound signals;
2, decomposing the filtered sound signal into a plurality of modal components by utilizing a VMD decomposition algorithm;
step 3, performing framing processing on each modal component, and dividing each modal component into a plurality of frames of sound signals; then windowing the sound signal after the framing processing to obtain a sound signal after the windowing processing;
step 4, calculating the short-time energy of each frame of sound signal by using the formula (8), and then adding the short-time energy of all the frame of sound signals corresponding to each modal component to obtain the short-time energy of each modal component;
wherein wlen is the length of each frame of sound signal, E (i) represents the short-time energy of the ith frame of sound signal, yi(a) Representing the sound signal of the i-th frame after windowing, fnRepresenting the total frame number of the sound signal obtained after the frame division processing;
processing the fault signal sample and the normal signal sample according to the steps 1-4 to obtain short-time energy of each modal component of each signal sample, wherein the short-time energy of each modal component forms a feature vector of the signal sample; the feature vector of each signal sample is used as the input of a support vector machine, and the output is a fault signal or a normal signal; training the support vector machine to complete the construction of a fault diagnosis model;
and 5, diagnosing the sound signals of the fan blades in real time by using the fault diagnosis model obtained in the step 5, and outputting a diagnosis result.
And in the step 2, the filtered sound signals are decomposed into 5-8 modal components.
In the step 1, a Butterworth filter is adopted to filter the sound signal.
Compared with the prior art, the invention has the beneficial effects that:
1. the method collects the sound signals of the fan blades for fault diagnosis, is easier to implement compared with the prior method for collecting the vibration signals of the fan, and the sensors are required to be arranged on the blades for collecting the vibration signals, so that the method is difficult to install and has high cost; the sound signal acquisition device can be installed on a tower of the fan, is easy to install, and cannot influence the normal work of the fan.
2. The invention takes the short-time energy of the sound signal as the characteristic value of the fan fault, the continuous change of the sound signal in the time domain, no matter the change of the amplitude, the frequency and the like, can cause the change of the short energy, when the fan blade has a fault, the fault signal can be appeared in a certain frequency band or a plurality of frequency bands, the signal is amplified equivalently by calculating the short-time energy of the sound signal, the signal change of the frequency band is represented in the form of energy, therefore, the short-time energy can more sensitively reflect the fan fault, and the accuracy is higher.
3. The VMD decomposition algorithm is a self-adaptive and completely non-recursive modal variation and signal processing method, the self-adaptability of the VMD decomposition algorithm is represented by determining the modal decomposition number of a given sequence according to the actual situation, the optimal central frequency and the limited bandwidth of each mode can be matched in a self-adaptive manner in the subsequent searching and solving processes, the effective separation of inherent modal components (IMF) and the frequency domain division of signals can be realized, the effective decomposition components of the given signals can be obtained, the optimal solution of the variation problem can be obtained finally, the problems of end point effect and modal component aliasing of the EMD decomposition algorithm method are solved, and the diagnosis accuracy is improved.
4. The short-time energy is simple to calculate, and meanwhile, the efficiency of the SVM serving as a two-classifier is far higher than that of other multi-classification neural networks, so that the method has high effect and short time consumption, and can be suitable for real-time diagnosis.
Drawings
FIG. 1 is an overall flow chart of the present invention;
fig. 2 is a time domain and a frequency domain distribution of each modal component decomposed by the present embodiment;
FIG. 3 is a short-time energy plot of the first three modal components of the present embodiment;
fig. 4 is a short-time energy plot of the last three modal components of this embodiment.
Detailed Description
The technical solutions of the present invention are clearly and completely described below with reference to the specific embodiments and the accompanying drawings, and are not intended to limit the scope of the present application.
The invention relates to a fan blade fault diagnosis method (a method for short, see fig. 1-4) based on a VMD decomposition algorithm, which comprises the following steps:
step 1, collecting sound signals of a fan blade, and filtering the sound signals by using a Butterworth filter in a formula (1) to filter low-frequency wind noise to obtain filtered sound signals; the Butterworth filter is characterized in that a frequency response curve in a pass frequency band is flat to the maximum extent, no ripple exists, and a frequency response curve in a stop frequency band gradually drops to zero; most of the noise of the fan sound signal comes from low-frequency wind noise, and most of the useful sound signal is in a high frequency band, so that the sound signal without wind noise is obtained through a Butterworth filter, wherein the expression of the Butterworth filter is as follows:
in the formula (1), H (w) is the signal amplitude, w is the signal frequency, wcTo cut-off frequency, ωpFor passband edge frequencies, n is the order of the Butterworth filter,
a value at the passband edge;
step 2, decomposing the filtered sound signals by utilizing a VMD decomposition algorithm to obtain modal components of a plurality of different frequency bands, namely decomposing the filtered sound signals into sound signals of a plurality of different frequency bands;
in the VMD decomposition algorithm, the Intrinsic Mode Function (IMF) is defined as a bandwidth-limited AM-FM function, assuming filteringThe sound signal f after wave reception is a multi-component signal, and is decomposed into K modal components u with limited bandwidth after VMD decomposition algorithmk(t), modal component ukThe expression of (t) is:
uk(t)＝Ak(t)cos[φk(t)]k＝1，2....K (2)
in the formula (2), Ak(t) is the modal component uk(t) instantaneous amplitude, [ phi ]k(t) represents the modal component uk(t) instantaneous phase, K being the number of modal components;
the mode component u is matched by the formula (3)k(t) performing Hilbert transform to obtain an analytic signal of the modal component;
[σ(t)+j/πt]uk(t) (3)
in the formula (3), σ (t) is a unit pulse function; j represents an imaginary unit;
order to
As modal component uk(t) estimated center frequency, where ωkAs modal component uk(t) using formula (4) to mix the analysis signal of the modal component with the estimated center frequency to obtain the modal component uk(t) an estimated bandwidth;
assuming that the sum of the estimated bandwidths of all the modal components is the minimum, and the constraint condition is that the sum of the estimated bandwidths of all the modal components is equal to the filtered sound signal, establishing a constraint variation model as follows:
in the formula (5), the reaction mixture is,
represents the partial derivative of time t, f representsFiltered sound signal, { u }kDenotes the k-th modal component after decomposition, { ωkDenotes the center frequency of the k-th modal component;
carrying out iterative solution on the formula (5) to obtain each modal component; the number of the modal components is 5-8, so that the accuracy and the real-time performance of diagnosis can be guaranteed; in this embodiment, the filtered sound signal is decomposed into 6 modal components, and fig. 2 shows the time domain and frequency domain distribution of the 6 modal components;
step 3, using the formula (6) to perform framing processing on each modal component, and dividing each modal component into fnA frame sound signal;
wherein N is the length of the modal component; overlap is the overlapping part between two frames of sound signals; wlen is the length of each frame of sound signal; inc is the displacement of the next frame of sound signal to the previous frame of sound signal, called frame shift;
windowing the sound signal after the framing processing to obtain a sound signal after the windowing processing, then:
yi(a)＝w(a)*x((i-1)*inc+a) (7)
where ω (a) is the window function, here the hanning window; y isi(a) Representing the i frame sound signal after windowing, wherein x is the expression of the sound signal in a time domain, x represents convolution, and a is the length of the i frame sound signal;
step 4, calculating the short-time energy of each frame of sound signal by using the formula (8), and then adding the short-time energy of all the frame of sound signals corresponding to each modal component to obtain the short-time energy of each modal component, namely the short-time energy of the sound signals in different frequency bands;
wherein E (i) represents the short-time energy of the sound signal of the ith frame;
FIGS. 3 and 4 show the short-term energies of the 6 modal components of the present embodiment;
processing the fault signal sample and the normal signal sample according to the steps 1-4 to obtain short-time energy of each modal component of each signal sample, wherein the short-time energy of each modal component forms a feature vector of the signal sample; taking the feature vector of each signal sample as the input of an SVM (support vector machine), wherein the output of the SVM is a fault signal or a normal signal, wherein 0 represents the fault signal, and 1 represents the normal signal; training the SVM to complete the construction of a fault diagnosis model;
in practical application, the sound signals of the fan blades are diagnosed in real time by using the fault diagnosis model, and a diagnosis result is output.
In order to verify the effectiveness of the method, an EMD decomposition algorithm and a VMD decomposition algorithm are respectively adopted to carry out fault diagnosis on the wind generating set with the fan model of W2000C-93-80, and the test results are shown in Table 1;
TABLE 1 comparison of test results of different methods
It can be known from the table that the accuracy rate obtained by utilizing the method to diagnose the fan blade is higher than that of the EMD decomposition algorithm, because the VMD decomposition algorithm can be utilized to effectively decompose the sound signal, the endpoint overlapping phenomenon is avoided, meanwhile, the short-time energy of the modal component is utilized as the characteristic of the fault diagnosis model, which is equivalent to that the sound signal is amplified and then the fault diagnosis model is utilized to identify, and the short-time energy can more sensitively reflect the fan fault, so the method has more advantages in the fan blade fault diagnosis aspect.
Nothing in this specification is said to apply to the prior art.
Claims (3)
1. A fan blade fault diagnosis method based on a VMD decomposition algorithm is characterized by comprising the following steps:
step 1, collecting sound signals of a fan blade and filtering the sound signals to obtain filtered sound signals;
2, decomposing the filtered sound signal into a plurality of modal components by utilizing a VMD decomposition algorithm;
step 3, performing framing processing on each modal component, and dividing each modal component into a plurality of frames of sound signals; then windowing the sound signal after the framing processing to obtain a sound signal after the windowing processing;
step 4, calculating the short-time energy of each frame of sound signal by using the formula (8), and then adding the short-time energy of all the frame of sound signals corresponding to each modal component to obtain the short-time energy of each modal component;
wherein wlen is the length of each frame of sound signal, E (i) represents the short-time energy of the ith frame of sound signal, yi(a) Representing the sound signal of the i-th frame after windowing, fnRepresenting the total frame number of the sound signal obtained after the frame division processing;
step 5, constructing a fault diagnosis model;
processing the fault signal sample and the normal signal sample according to the steps 1-4 to obtain short-time energy of each modal component of each signal sample, wherein the short-time energy of each modal component forms a feature vector of the signal sample; the feature vector of each signal sample is used as the input of a support vector machine, and the output is a fault signal or a normal signal; training the support vector machine to complete the construction of a fault diagnosis model;
and 5, diagnosing the sound signals of the fan blades in real time by using the fault diagnosis model obtained in the step 5, and outputting a diagnosis result.
2. The wind turbine blade fault diagnosis method based on the VMD decomposition algorithm according to claim 1, wherein in step 2, the filtered sound signal is decomposed into 5-8 modal components.
3. The VMD decomposition algorithm-based fan blade fault diagnosis method of claim 1, wherein the step 1 is performed by filtering the sound signal with a butterworth filter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110525344.4A CN113250911B (en) | 2021-05-12 | 2021-05-12 | Fan blade fault diagnosis method based on VMD decomposition algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110525344.4A CN113250911B (en) | 2021-05-12 | 2021-05-12 | Fan blade fault diagnosis method based on VMD decomposition algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113250911A true CN113250911A (en) | 2021-08-13 |
| CN113250911B CN113250911B (en) | 2022-02-18 |
Family
ID=77181844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110525344.4A Expired - Fee Related CN113250911B (en) | 2021-05-12 | 2021-05-12 | Fan blade fault diagnosis method based on VMD decomposition algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113250911B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115434872A (en) * | 2022-08-11 | 2022-12-06 | 兰州理工大学 | Wind turbine generator gearbox composite fault diagnosis method based on AVMD and improved RSSD |
| WO2023024242A1 (en) * | 2021-08-23 | 2023-03-02 | 洛阳轴承研究所有限公司 | Vibration test method and system for bearing unit |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108594161A (en) * | 2018-05-03 | 2018-09-28 | 国网重庆市电力公司电力科学研究院 | Foreign matter voice signal noise-reduction method, system in a kind of electric energy meter |
| CN109139390A (en) * | 2018-09-27 | 2019-01-04 | 河北工业大学 | A kind of fan blade fault recognition method based on acoustical signal feature database |
| CN109887510A (en) * | 2019-03-25 | 2019-06-14 | 南京工业大学 | Voiceprint recognition method and device based on empirical mode decomposition and MFCC |
| US20200284687A1 (en) * | 2019-02-19 | 2020-09-10 | Dalian University Of Technology | A method for automatically detecting free vibration response of high-speed railway bridge for modal identification |
| CN112113784A (en) * | 2020-09-22 | 2020-12-22 | 天津大学 | Equipment state monitoring method based on equipment acoustic signals and EMD |
-
2021
- 2021-05-12 CN CN202110525344.4A patent/CN113250911B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108594161A (en) * | 2018-05-03 | 2018-09-28 | 国网重庆市电力公司电力科学研究院 | Foreign matter voice signal noise-reduction method, system in a kind of electric energy meter |
| CN109139390A (en) * | 2018-09-27 | 2019-01-04 | 河北工业大学 | A kind of fan blade fault recognition method based on acoustical signal feature database |
| US20200284687A1 (en) * | 2019-02-19 | 2020-09-10 | Dalian University Of Technology | A method for automatically detecting free vibration response of high-speed railway bridge for modal identification |
| CN109887510A (en) * | 2019-03-25 | 2019-06-14 | 南京工业大学 | Voiceprint recognition method and device based on empirical mode decomposition and MFCC |
| CN112113784A (en) * | 2020-09-22 | 2020-12-22 | 天津大学 | Equipment state monitoring method based on equipment acoustic signals and EMD |
Non-Patent Citations (1)
| Title |
|---|
| 李春琦: "基于声学特征的旋转机械故障诊断方法及其DSP实现", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023024242A1 (en) * | 2021-08-23 | 2023-03-02 | 洛阳轴承研究所有限公司 | Vibration test method and system for bearing unit |
| CN115434872A (en) * | 2022-08-11 | 2022-12-06 | 兰州理工大学 | Wind turbine generator gearbox composite fault diagnosis method based on AVMD and improved RSSD |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113250911B (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Fault diagnosis for a wind turbine generator bearing via sparse representation and shift-invariant K-SVD | |
| CN109977920A (en) | Fault Diagnosis of Hydro-generator Set method based on time-frequency spectrum and convolutional neural networks | |
| CN113250911B (en) | Fan blade fault diagnosis method based on VMD decomposition algorithm | |
| Li et al. | Period-assisted adaptive parameterized wavelet dictionary and its sparse representation for periodic transient features of rolling bearing faults | |
| CN108489719B (en) | A kind of combined failure of rotating machinery diagnostic method based on the unusual spectral factorization of G-P | |
| CN104677623A (en) | On-site acoustic diagnosis method and monitoring system for wind turbine blade failure | |
| CN105928702B (en) | Variable working condition box bearing method for diagnosing faults based on form PCA | |
| CN105678343A (en) | Adaptive-weighted-group-sparse-representation-based diagnosis method for noise abnormity of hydroelectric generating set | |
| CN111308260B (en) | Electric energy quality monitoring and electric appliance fault analysis system based on wavelet neural network and working method thereof | |
| CN102629243A (en) | End effect suppression method based on neural network ensemble and B-spline empirical mode decomposition (BS-EMD) | |
| Cheng et al. | Adaptive sparsest narrow-band decomposition method and its applications to rolling element bearing fault diagnosis | |
| US20240088657A1 (en) | Fractional domain noise reduction method for power signal | |
| CN111289796A (en) | Method for detecting subsynchronous oscillation of high-proportion renewable energy power system | |
| CN110909827A (en) | Noise reduction method suitable for fan blade sound signals | |
| CN113919525A (en) | Power station fan state early warning method, system and application thereof | |
| CN105352726A (en) | Fault diagnosis method for gear | |
| CN108197073A (en) | A kind of improved electric vehicle rechargeable electrical energy signature analysis method | |
| CN111400959A (en) | Blade fault diagnosis method and device of wind generating set | |
| CN113988125A (en) | Torsional vibration signal instantaneous frequency extraction method based on improved synchronous compression transformation | |
| CN103149047A (en) | Cooling tower acoustic diagnosis method based on nonlinear mixed model | |
| Guo et al. | Gear fault diagnosis of wind turbine based on discrete wavelet transform | |
| CN104459186A (en) | Tachometer-free order-ratio analyzing method based on sparse segmentation fitting and integral approximation | |
| CN110222390B (en) | Gear crack identification method based on wavelet neural network | |
| CN110147637A (en) | Based on the small impact-rub malfunction diagnostic method for involving the greedy sparse identification of harmonic components | |
| CN115438693A (en) | Method for reducing noise of fault vibration signal of wind turbine generator transmission chain |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220218 |