CN112780503B - Fan blade protective paint damage monitoring method and system based on audio signals - Google Patents
Fan blade protective paint damage monitoring method and system based on audio signals Download PDFInfo
- Publication number
- CN112780503B CN112780503B CN202011478509.9A CN202011478509A CN112780503B CN 112780503 B CN112780503 B CN 112780503B CN 202011478509 A CN202011478509 A CN 202011478509A CN 112780503 B CN112780503 B CN 112780503B
- Authority
- CN
- China
- Prior art keywords
- damage
- rotating speed
- protective paint
- audio signals
- energy level
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
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)
- Wind Motors (AREA)
Abstract
The application relates to a method and a system for monitoring damage of fan blade protective paint based on audio signals. The sound signal containing the wind sweeping pneumatic noise of the fan within a period of time is collected through audio equipment, and real-time monitoring and fault diagnosis of damage of fan blade protective paint are achieved through analysis of the sound signal. The method and the device can monitor the health state of the blade in real time and achieve the purpose of nondestructive testing. Meanwhile, the method and the device can perform early warning on the blade undergoing deterioration and guide the time point of maintenance.
Description
Technical Field
The application relates to a fan blade protective paint damage monitoring method and system based on audio signals, and is applicable to the technical field of fan health monitoring.
Background
The blade of the wind driven generator is one of important parts of a wind turbine generator set for converting wind energy into mechanical energy, and is a basis for obtaining higher wind energy utilization coefficient and economic benefit, and the performance and the generating efficiency of the whole machine are directly influenced by the state of the blade. As the blade is running, particles in the air surrounding the blade will constantly erode the leading edge of the blade: including raindrops, hailstones, salt fog, sand and stones, field high-altitude organisms and the like, the direct influence is that the aerodynamic appearance of the blades is damaged, so that the aerodynamic performance of the whole unit is influenced, and the annual energy production loss is caused. Meanwhile, the erosion of the front edge can also cause the aerodynamic noise of the blade to show an increase in magnitude, so that the noise disturbs residents and other problems. If the erosion is too severe and cannot be properly treated for a long time, it may further affect the structural layer of the blade, eventually leading to longer maintenance times and difficulty in restoring the original aerodynamic profile. In order to retard or reduce the effects of corrosion at the leading edge of the blade, a protective coating is applied to the leading edge region of the blade tip. Therefore, the corrosion damage level of the front edge of the blade can be monitored by judging the damage degree of the protective paint, and timely and effective treatment can be ensured before the corrosion of the front edge develops to the structural layer.
At present, methods for monitoring damage of blade protective paint of a wind turbine generator mainly include methods such as visual detection of an artificial telescope and inspection of high-altitude spiders or operation platforms. In addition, by combining with the image intelligent processing technology which is suddenly advanced in recent years, the method of adopting the unmanned aerial vehicle to patrol is more intuitive and more detailed, and a good effect is achieved. In addition, some systems adopt a vibration sensor mounted on the blade or knock the blade to judge whether the blade is damaged. However, the above methods all have obvious disadvantages, such as the health state of the blade cannot be monitored in real time, the blade itself is damaged by the detection method, and the life safety of inspection personnel may be involved.
Chinese patent application 201810680085.0 proposes a method for monitoring and controlling rain erosion of a wind turbine generator system according to the current operating state of the wind turbine generator system and the current weather information of the environment, thereby preventing the blades of the wind turbine generator system from fatigue and accumulated damage. The scheme is a preventive control method and cannot identify and monitor the damage of the generated protective paint.
US9395337B2 discloses the use of a microphone device mounted at the bottom of a wind turbine tower to record the sound signals of the blades and to analyze the whistles in the sound signals by doppler shift analysis to locate the failure of the blades. The patent mainly refers to the principle and mechanism formula of locating the damage position of the blade by a Doppler analysis method, but the characteristic of the blade concerned is whistle sound which is one of the characteristics of various blade surface faults. Therefore, the method cannot diagnose the failure of the blade, and cannot judge the damage of the protective paint by the method.
The Chinese patent application 201811408152.X also provides a method for carrying out image splicing on a plurality of local areas of the blade acquired by the unmanned aerial vehicle so as to realize fault positioning. However, the unmanned aerial vehicle flies in the air, and is easily affected by factors such as wind speed and electromagnetic interference, shooting jitter is strong, the obtained image or video pixel is low, and small damage to the blade cannot be recognized in time. Meanwhile, the flight of the unmanned aerial vehicle is a regular inspection method, and continuous real-time monitoring can not be provided for the health state of the blade.
The Chinese patent application 201410017842.8 obtains the knocking sound and the environmental noise when the knocking hammer knocks the blade, adopts the methods of filtering, wavelet transformation, Fourier transformation, modal decomposition and the like to extract the characteristics, and adopts the neural network algorithm to directly predict the delamination condition of the blade. The method can not realize nondestructive and real-time detection, repeated knocking has potential threats to the structure and the performance of the blade, and the blade needs to be stopped to enter the tower cylinder during detection, so that the detection is not suitable for frequent operation. Meanwhile, the patent mentions that the signals acquired by knocking can only judge whether the knocking part is damaged or not, and can conclude that the whole blade is time-consuming and labor-consuming to inspect.
In addition to the above methods, the corrosion of the leading edge of the blade is protected from the perspective of corrosion-resistant materials, modification of the blade structure. In the prior art, a method for monitoring damage of protective paint at the front edge of a blade through a pneumatic noise sound signal generated by blade wind sweeping is not available.
Disclosure of Invention
The invention aims to provide a damage monitoring method and a damage monitoring system of fan blade protective paint based on audio signals, which can monitor the health state stage of a blade in real time and achieve the aim of nondestructive testing. Meanwhile, the method and the device can perform early warning on the blade undergoing deterioration and guide the time point of maintenance.
The application relates to a damage monitoring method of a fan blade protective paint based on an audio signal, which comprises a training stage and an online operation stage; the training phase comprises the following steps:
collecting a plurality of groups of audio signals within a period of time, and acquiring an average rotating speed value of each group of audio corresponding to the time;
converting the audio signal to obtain a spectrogram and converting the spectrogram into an energy level;
screening working conditions, and selecting data in a certain rotating speed range;
each group of audio signals is processed to obtain a sample, the rotating speed is used as an independent variable, the maximum energy level of the target frequency band is used as a dependent variable, linear fitting is carried out on the rotating speed and the energy level of a plurality of sample points, and residual distribution of the energy level is calculated;
respectively carrying out the operation of the steps on each frequency band to obtain a linear model;
the on-line operation phase comprises the following steps:
acquiring a group of audio signals and corresponding rotating speed values thereof in real time, and judging whether the rotating speed values are in a rotating speed range screened by training data;
predicting the predicted energy level of the target frequency band according to the rotating speed by adopting the model obtained by training, and calculating the residual error between the energy level of the group of audio signals and the predicted energy level;
selecting a plurality of groups of collected effective data, and calculating the average value of residual errors;
estimating the damage level of the protective paint by adopting a prediction factor according to the residual distribution of the training data and the average value of the online running residual;
when a plurality of frequency bands are selected for fitting, the final prediction factor takes the median of the prediction factors of each frequency band at the same time;
and judging the stage of the fault according to the current factor and the historical factor by using the rule.
The application also relates to a damage monitoring system of the fan blade protective paint based on the audio signal, which comprises a machine end hardware device and a station end server, wherein the machine end hardware device comprises a sound sensor and an edge hardware data acquisition system, and the edge hardware data acquisition system executes the damage monitoring method.
According to the damage monitoring method and system of the fan blade protective paint based on the audio signal, the following beneficial technical effects are achieved:
(1) the online real-time monitoring and the non-contact monitoring can be realized, the timeliness and the high efficiency of the monitoring are ensured, and the possible damage caused by the monitoring is avoided;
(2) based on the mechanism that the rotating speed of the blade and the energy level of wind sweeping sound are in a linear relation when the blade sweeps wind, the damage of the protective paint can be identified only by using an audio signal by utilizing the phenomenon that a relation curve translates along the direction with high energy level when the protective paint is damaged, so that the monitoring and the identification are more efficient and accurate;
(3) aiming at the mechanism that the damage failure of the protective paint has different rates at different stages, the damage degree of the protective paint can be judged, the precision of damage monitoring is improved, and a more accurate maintenance suggestion is provided for subsequent maintenance;
(4) the method for taking the median value based on the fitting multiple narrow frequency bands can effectively remove the interference caused by other types of faults and various environmental noises without carrying out denoising processing on data at an early stage, thereby saving a data processing program and improving the data processing efficiency;
(5) the method and the device map the wind sweeping energy level to the fault level finally according to the wind sweeping energy level variation trend of the blades, can early warn the deteriorating blades in time, and improve early warning efficiency and early warning accuracy.
Drawings
FIG. 1 shows a schematic of the trend of leading edge protective paint damage over time.
FIG. 2 shows a flow chart of a fan blade protective paint damage monitoring method of the present application.
Fig. 3 shows an example of the relationship between the data rotation speed and the maximum sound pressure level in the embodiment of the present application.
FIG. 4 shows a graph of blade damage fan factor over time in an embodiment of the present application.
FIG. 5 is a graph showing fan factor over time for blade maintenance in an embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more apparent, embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be noted that the embodiments and features of the embodiments in the present application may be arbitrarily combined with each other without conflict.
According to the damage monitoring method of the fan blade protection paint based on the audio signals, the audio equipment is used for collecting the sound signals containing the fan wind sweeping pneumatic noise within a period of time, and the sound signals are analyzed, so that real-time monitoring and fault diagnosis of the damage of the fan blade protection paint are achieved. The damage monitoring system of the fan blade protective paint based on the audio signals comprises a machine end hardware device and a station end server, wherein the machine end hardware device comprises a sound sensor and an edge hardware data acquisition system, the sound sensor is used for acquiring sound signals of a fan, the environment and the like, and the station end server can run matched application software and display data. The edge hardware data acquisition system is used for executing the monitoring method described below.
In the present application, a blade with a leading edge protection paint has the following characteristics: 1) the erosion ranges and degrees of the three blades are similar, and the characteristic is reflected in the state that the wind sweeping energy levels of the three blades are always approximate along with time; 2) the damage of the protective paint is characterized in that the paint surface is gradually peeled off, the peeling process can cause the loss of the pneumatic appearance and the rough section, and the characteristic can be reflected in the change of the level of the wind sweeping energy; 3) protecting different stages of paint damage development, namely from a coating layer to a putty layer to a structural layer until cracking occurs, wherein the change rates of the energy levels are different, and the energy levels develop in a step shape along with time, as shown in figure 1; 4) the change in energy level is manifested over a large frequency range and not as specific failure band as a blade crack. Based on the fault characteristics, the judgment on the damage of the protective paint is realized through the energy level change and the change trend of the sound signal.
According to the damage monitoring method of the fan blade protective paint based on the audio signal, the specific step flow is shown in fig. 2 and comprises a training phase and an online operation phase.
The training phase comprises the following steps:
(1.1) collecting a plurality of groups of audio signals within a period of time, and acquiring an average rotating speed value of each group of audio corresponding to the time; the average rotating speed value can be obtained from an external system, and when the rotating speed cannot be obtained from the outside, the average rotating speed value can also be calculated by identifying the number of times of wind sweeping in the audio signal per unit time;
(1.2) carrying out short-time Fourier transform (STFT) conversion on the audio signal to obtain a spectrogram, and converting the spectrogram into an energy level, wherein the energy level refers to a parameter representing the energy of the wind sweeping sound;
(1.3) data screening: screening working conditions, selecting data in a certain rotating speed range, and filtering low-rotating-speed noise points and data points in full-time; the selected speed is generally a higher speed interval of the fan; the fan rotating speed range can be determined according to the fan parameters and whether the collection conditions are limited; full-load is the highest rotating speed value, namely after reaching the rotating speed under a certain wind speed, the wind speed is increased again, the rotating speed cannot be increased, and the power can also become approximate rated power;
and (1.4) obtaining one sample by each group of audio signals through the processing, so that a plurality of groups of audio signals can obtain a plurality of samples in a period of time. Taking the rotating speed as an independent variable and the maximum energy level of the target frequency band as a dependent variable, performing linear fitting on the rotating speed and the energy level of a plurality of sample points by adopting linear regression, and calculating residual distribution of the energy level, such as an n-sigma value;
and (1.5) when a plurality of frequency bands are selected, respectively carrying out the operation of the step (1.4) on each frequency band to obtain a linear model. Because the damage of the protective paint has the same performance in a large range of frequency bands, one or more target frequency bands (5000 Hz-16000 Hz) can be selected, and when a plurality of frequency bands are selected, the range of each frequency band is smaller and the linear relation is stronger.
The on-line operation phase comprises the following steps:
and (2.1) acquiring a group of audio signals and corresponding rotating speed values in real time, and judging whether the rotating speed values are in the rotating speed range screened by the training data. When the rotating speed is in the screening range, carrying out STFT conversion on the audio signal and calculating the energy level of the audio signal; when the rotating speed is not in the screening range, subsequent operation is not carried out;
(2.2) predicting the predicted energy level of the target frequency band according to the rotating speed by adopting the model obtained by training, and calculating the residual error between the energy level of the group of audio signals and the predicted energy level;
(2.3) selecting a plurality of groups of collected effective data, and calculating the average value of residual errors;
(2.4) evaluating the damage level of the protective paint according to the residual distribution of the training data and the average value of the online running residual, wherein the formula can be as follows:
wherein x is an online operation residual error average value, y is a prediction factor which is the damage level of the protective paint, base is a current fan damage level reference value, when y is 0, the protective paint is not damaged, and when y is 1, the protective paint is seriously damaged; the nsigma is the sum of the average value of the residual errors of the training data and n times of standard deviation, and n can be determined according to the requirement of monitoring precision;
(2.5) when a plurality of frequency bands are selected for fitting, finally, the prediction factors take the median of the factors of each frequency band at the same time;
(2.6) judging the stage of the fault according to the current factor and the historical factor by using a rule, and outputting the fault stage and whether mutation occurs at the current stage; mutation here means that switching occurs at different stages of injury as shown in figure 1.
Examples
In the embodiment, a sound signal acquired in real time in a certain wind field is used as a case for introduction.
Fig. 3 is a graph comparing the original data with the relationship between the rotation speed and the maximum sound pressure level after data screening and frequency band selection, where the left side is the original data, the right side is the data after data screening and target frequency band selection, the abscissa is the rotation speed, and the ordinate is the maximum sound pressure level (Max SPL). It can be seen that the energy level of the whole frequency range of the original data has a certain linear relation, but the variance is large, and the variance between the rated rotating speed and some specific rotating speeds is obviously larger. After the energy level is recalculated by data screening such as rotating speed and the like and selecting a frequency band with obvious wind sweeping sound characteristics, the linear relation between the maximum sound pressure level and the rotating speed is obvious, and the variance is small.
Fig. 4 is a graph showing the damage of the blade protective paint at the later stage of a certain fan, wherein the factor changes with time, and it can be seen that the factor value gradually increases from a certain time period and increases once again after a certain period of plateau. The fan finds that the damage is close to the structural layer when being manually checked, and the damage corresponds to the factor value well. In the example, 5000-12000Hz frequency dividing bands are used for fitting, and the graphs in FIG. 4 are respectively a factor graph of each frequency band along with time, a final fault factor along with time graph output after the factors are integrated, and an output fault stage along with time graph from top to bottom. It can be seen that the development trends and factor ranges of the factors of the frequency bands are consistent, and the fault stage and the manual inspection result correspond well. In the following figures, S1 indicates the beginning of damage of the outer shape coating, S2 indicates the damage of the putty, and S3 indicates the damage of the approaching structure layer.
FIG. 5 is a graph of factor versus time before and after a damaged blade blower is serviced. It can be seen that the trend of failure and degradation occurs before maintenance, the factors of each frequency band are simultaneously increased, and the factors can be automatically recovered and maintained at the level of no alarm after maintenance. In the figure, S4 represents a blade leading edge degradation stage, and S5 represents a blade maintenance stage.
The damage monitoring method and system of the present application can be used to diagnose damage to other finishes besides protective paint, such as a putty layer and/or paint layer disposed under a blade protective paint or film.
Although the embodiments disclosed in the present application are described above, the descriptions are only for the convenience of understanding the present application, and are not intended to limit the present application. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims.
Claims (4)
1. A fan blade protective paint damage monitoring method based on audio signals is characterized by comprising a training phase and an online operation phase;
the training phase comprises the following steps:
(1.1) collecting a plurality of groups of audio signals within a period of time, and acquiring an average rotating speed value of each group of audio corresponding to the time;
(1.2) converting the audio signal to obtain a spectrogram, and converting the spectrogram into an energy level;
(1.3) screening working conditions, and selecting data in a certain rotating speed range;
(1.4) each group of audio signals are processed in the steps (1.1) to (1.3) to obtain a sample, the rotating speed is used as an independent variable, the maximum energy level of the target frequency band is used as a dependent variable, linear fitting is carried out on the rotating speed and the energy level of a plurality of sample points, and residual distribution of the energy level is calculated;
(1.5) respectively carrying out the operation of the step (1.4) on each frequency band to obtain a linear model;
the on-line operation phase comprises the following steps:
(2.1) acquiring a group of audio signals and corresponding rotating speed values thereof in real time, and judging whether the rotating speed values are in a rotating speed range screened by training data;
(2.2) predicting the predicted energy level of the target frequency band according to the rotating speed by adopting the model obtained by training, and calculating the residual error between the energy level of the group of audio signals and the predicted energy level;
(2.3) selecting a plurality of groups of collected effective data, and calculating the average value of residual errors;
(2.4) estimating the damage level of the protective paint by adopting a prediction factor according to the residual distribution of the training data and the average value of the online running residual;
(2.5) when a plurality of frequency bands are selected for fitting, the final prediction factor takes the median of the prediction factors of each frequency band at the same time;
and (2.6) judging the stage of the fault according to the current factor and the historical factor by using the rule.
2. A method for monitoring damage to a fan blade protective paint according to claim 1, wherein in step (1.1) the average rotational speed value is obtained from an external system or calculated by identifying the number of wind sweeps per unit time in the audio signal.
3. The damage monitoring method of the fan blade protective paint according to claim 1 or 2, characterized in that after the step (2.6), the failure stage is continuously output and whether the sudden change of the damage stage occurs currently is judged.
4. A damage monitoring system of fan blade protective paint based on audio signals comprises a machine end hardware device and a station end server, wherein the machine end hardware device comprises a sound sensor and an edge hardware data acquisition system, and the edge hardware data acquisition system executes the damage monitoring method according to any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011478509.9A CN112780503B (en) | 2020-12-15 | 2020-12-15 | Fan blade protective paint damage monitoring method and system based on audio signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011478509.9A CN112780503B (en) | 2020-12-15 | 2020-12-15 | Fan blade protective paint damage monitoring method and system based on audio signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112780503A CN112780503A (en) | 2021-05-11 |
| CN112780503B true CN112780503B (en) | 2021-12-28 |
Family
ID=75750948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011478509.9A Active CN112780503B (en) | 2020-12-15 | 2020-12-15 | Fan blade protective paint damage monitoring method and system based on audio signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112780503B (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104101652B (en) * | 2014-07-10 | 2017-02-15 | 南京航空航天大学 | Audio signal based wind power blade damage monitoring method and system |
| CN207195098U (en) * | 2017-07-31 | 2018-04-06 | 上海绿孚科技有限公司 | A kind of blade state monitoring system of the wind power generating set based on acoustic processing |
| CN110792563B (en) * | 2019-11-04 | 2020-09-15 | 北京天泽智云科技有限公司 | Wind turbine generator blade fault audio monitoring method based on convolution generation countermeasure network |
-
2020
- 2020-12-15 CN CN202011478509.9A patent/CN112780503B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112780503A (en) | 2021-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2868643C (en) | Wind turbine and method for evaluating the health state of the blades thereof | |
| CN107796611B (en) | Alarm system for detecting abnormal work of wind driven generator | |
| CN112727704B (en) | Method and system for monitoring corrosion of leading edge of blade | |
| JP2022552449A (en) | Methods and apparatus for inspecting wind turbine blades, and equipment and storage media therefor | |
| CN111400961B (en) | Wind generating set blade fault judging method and device | |
| CN104677623A (en) | On-site acoustic diagnosis method and monitoring system for wind turbine blade failure | |
| CN112067701B (en) | Fan blade remote auscultation method based on acoustic diagnosis | |
| CN111161756A (en) | Method for extracting and identifying abnormal whistle contour in wind sweeping sound signal of fan blade | |
| CN109139390B (en) | Fan blade fault identification method based on acoustic signal feature library | |
| CN111120388B (en) | Fan state combined monitoring method and system | |
| CN110554090A (en) | Wind turbine generator and crack monitoring system and method of variable-pitch bearing of wind turbine generator | |
| CN109236587A (en) | It is a kind of for detecting the alarm system of wind-driven generator abnormal work | |
| WO2021037751A1 (en) | Detection of abnormal conditions on a wind turbine generator | |
| CN112727703B (en) | Fan blade protective film damage monitoring method and system based on audio signal | |
| CN111927717B (en) | System and method for online monitoring noise of transmission chain of fan engine room | |
| CN114183312A (en) | System and method for monitoring state of blades of wind turbine generator | |
| CN112780503B (en) | Fan blade protective paint damage monitoring method and system based on audio signals | |
| CN117685177A (en) | Method for reducing rain erosion of blade and monitoring equipment | |
| AU2021415086B2 (en) | Blade fault diagnosis method, apparatus and system, and storage medium | |
| Chen et al. | Acoustic-based whistle detection of drain hole for wind turbine blade | |
| CN117929546B (en) | Early damage monitoring method for internal material of wind driven generator in blade transportation | |
| CN111188742A (en) | Wind generating set blade icing detection method based on optical fiber acceleration sensor | |
| CN112926626B (en) | Fan blade fault detection method based on sparse Bayesian learning and power spectrum separation | |
| CN117723303B (en) | Acoustic monitoring method for wind generating set bearing | |
| CN118309616A (en) | Operation and maintenance method and system of offshore wind power equipment |
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 |