CN112882017B - Wind power blade damage monitoring method and system based on Doppler radar - Google Patents
Wind power blade damage monitoring method and system based on Doppler radar Download PDFInfo
- Publication number
- CN112882017B CN112882017B CN201911199627.3A CN201911199627A CN112882017B CN 112882017 B CN112882017 B CN 112882017B CN 201911199627 A CN201911199627 A CN 201911199627A CN 112882017 B CN112882017 B CN 112882017B
- Authority
- CN
- China
- Prior art keywords
- blade
- time
- frequency
- damage
- echo signals
- 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
- 230000006378 damage Effects 0.000 title claims abstract description 93
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012544 monitoring process Methods 0.000 title claims abstract description 31
- 238000012806 monitoring device Methods 0.000 claims abstract description 23
- 230000007797 corrosion Effects 0.000 claims abstract description 17
- 238000005260 corrosion Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000001228 spectrum Methods 0.000 claims description 59
- 238000012360 testing method Methods 0.000 claims description 25
- 238000005070 sampling Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 6
- 238000009434 installation Methods 0.000 claims description 4
- 101100083446 Danio rerio plekhh1 gene Proteins 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 230000036541 health Effects 0.000 abstract description 4
- 238000000605 extraction Methods 0.000 abstract description 3
- 208000010392 Bone Fractures Diseases 0.000 description 12
- 206010017076 Fracture Diseases 0.000 description 12
- 238000010586 diagram Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000010248 power generation Methods 0.000 description 4
- 238000005336 cracking Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 239000004576 sand Substances 0.000 description 3
- 208000025274 Lightning injury Diseases 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 208000014674 injury Diseases 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 238000001931 thermography Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000010000 carbonizing Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
Abstract
The invention discloses a wind power blade damage monitoring method and system based on a Doppler radar. During monitoring, firstly, performing DC removal processing on the I, Q two paths of echo signals, and obtaining a time-frequency signal by adopting short-time Fourier transform; inputting a time-frequency signal, and obtaining the corresponding rotating speed of the blade in real time through a rotating speed monitoring device; and finally, establishing a blade damage monitoring algorithm based on time-frequency multi-feature extraction to obtain blade damage type judgment, wherein the damage type comprises fracture, surface corrosion and angle mismatch. The method can realize non-contact, comprehensive and accurate monitoring of the health condition of the blade by only a single radio frequency sensor, can discover the damage of the blade in time, and has the advantages of effectiveness, feasibility, reliable performance and high accuracy.
Description
Technical Field
The invention relates to the field of radars, in particular to a wind power blade damage monitoring method and system based on a Doppler radar.
Background
The wind energy has the characteristics of large reserve, reproducibility, cleanness, no pollution and the like, and has important significance in developing and utilizing good wind energy under the condition of meeting the energy crisis. Wind power generation is one of the most mature power generation modes with large-scale development conditions and commercial development prospects except water in the renewable energy field, the fund input for wind power generation is continuously increased all over the world, and the total capacity of the global wind power installation reaches 591GW by 2018. According to the global wind energy harnessing, by 2030, wind energy can supply more than 20% of the total global electric power, a wind power generator is an important device for converting wind energy into electric energy, and wind power generator blades (hereinafter referred to as "blades") are important components of the wind power generator. At present, the number of the blades of the wind driven generator is three, the blades are symmetrically arranged on a wind power tower, and rotate around a shaft under the action of wind power to drive the generator to generate electric energy, and the normal operation of the wind driven generator is the basis of the operation of the generator. The blades are manufactured by adopting a composite material and an integral forming process, along with the development of wind energy development industry, the large-scale characteristics of the wind turbine generator are more and more obvious, the probability of damage and defects is increased due to the increase of the structural size, the influence on the integral power generation capacity of the wind power plant after a single blade is damaged is also increased, and meanwhile, the transportation and replacement of the blades are more difficult.
Because the wind driven generator works in places with severe climatic conditions such as mountains, deserts, seasides and the like for a long time, the blades bear strong and complex natural force, and are extremely easy to generate fault damage. Common types of injury are: (1) cracking and cracking, if severe, may result in cracking. Low temperature and mechanical vibration are prone to such damage; and (2) carbonizing and puncturing the surface of the material. The surface of the blade is generally carbonized after the blade is subjected to slight lightning stroke, and the blade is broken down due to serious lightning stroke; (3) Surface wear, corrosion and sand holes are particularly severe in windward sides. In desert areas, sand holes are easily caused by the impact of wind and sand. In coastal areas, because more salt mist is carried in the air, corrosion is easy to cause; (4) The pneumatic unbalance, namely the problem of angle matching of three blades, can cause very large pressure on the cabin; (5) other damages such as icing.
The most common mode of monitoring blade health condition still is artifical inspection at present, and this mode inefficiency, risk are high, feedback cycle is long, and just can judge obvious damage. In recent years, methods including acoustic emission, ultrasonic wave, infrared thermal imaging, audio and video sensors and the like are sequentially proposed at home and abroad to detect defects and damage of blades, and the methods have the following defects and disadvantages: (1) Acoustic emission techniques require sensors to be placed internally during blade manufacture and do not allow for non-contact monitoring. (2) Ultrasonic wave and infrared thermal imaging all need the blade to dismantle to the mill when resting and detect, and timeliness is poor, and can not monitor three blades comprehensively. (3) Although the audio and video technology is a non-contact method, the technology is easily influenced by environmental factors, the post-processing is complex, and the reliability is low.
Disclosure of Invention
The invention aims to provide a wind power blade damage monitoring method and system based on Doppler radar.
The technical solution for realizing the purpose of the invention is as follows: a wind power blade damage monitoring method based on Doppler radar comprises the following steps:
step 1, collecting echo signals of blades of a wind driven generator in real time, and sampling the echo signals;
step 2, performing DC removal processing on the echo signals, and obtaining time-frequency signals by adopting short-time Fourier transformation;
step 3, acquiring the corresponding rotating speed of the blade in real time;
and 4, judging the damage type of the blade to obtain a judging result.
The wind power blade damage monitoring system based on Doppler radar comprises a signal acquisition device, a time-frequency analysis device, a rotation speed monitoring device and a blade damage monitoring device which are connected in sequence,
the signal acquisition device is used for acquiring echo signals of the wind driven generator blade and sampling the echo signals;
the time-frequency analysis device is used for carrying out DC removal on the echo signals, obtaining time-frequency signals by adopting short-time Fourier transformation, and then transmitting the time-frequency signals to the rotating speed monitoring device;
the rotating speed monitoring device is used for acquiring the rotating speed corresponding to the blade in real time;
the blade damage monitoring device is used for judging the type of blade damage, and the type of damage comprises fracture, surface corrosion and angle mismatch.
Compared with the prior art, the invention has the remarkable advantages that: 1) According to the invention, the non-contact measurement of the rotating speed of the blade can be realized by using a single Doppler radar, compared with a traditional sensor in the cabin, the operation is more convenient, the feedback time is shorter, and the rotating speed information can be obtained in a short time by only holding the radar to aim at the blade; 2) According to the invention, the real-time detection of three types of damage to the blade can be realized by using a single Doppler radar under the condition of not contacting the blade and not affecting the normal operation of the blade, so that the influence of artificial factors of manual observation is avoided, and the method is more objective and accurate; 3) The radar is used as one of the radio frequency sensors, is little affected by the environment, can work in all weather and under all weather conditions, and is stable and reliable; 4) The method is simple and effective, reliable in performance and convenient to implement.
The invention is described in further detail below with reference to the accompanying drawings.
Drawings
FIG. 1 is a schematic diagram of the system principle of the present invention.
FIG. 2 is a flow chart of a system and method for monitoring blade speed and blade damage in real time in accordance with the present invention.
Fig. 3 shows the time-frequency signal of the blade echo obtained by the time-frequency analysis device according to the invention.
Fig. 4 is a schematic diagram of a time-frequency signal processed by the rotation speed monitoring device of the present invention.
FIG. 5 is a schematic diagram of a time-frequency signal of a broken blade processed by the blade damage monitoring device of the present invention.
FIG. 6 is a schematic diagram of a time-frequency signal of a blade damage monitoring device of the present invention for treating a surface corroded blade.
FIG. 7 is a schematic diagram of a time-frequency signal of a blade with a non-matching processing angle according to the blade damage monitoring device of the present invention.
Detailed Description
With reference to the accompanying drawings, the wind power blade damage monitoring method based on the Doppler radar comprises the following steps of:
step 1, collecting echo signals of blades of a wind driven generator in real time, and sampling the echo signals; the method comprises the following steps:
step 1-1, erecting a single Doppler radar under a wind driven generator, aligning a radar antenna upwards to the rotation center of a blade, and continuously recording blade echo signals;
step 1-2, sampling the recorded signals in a I, Q two-way mode.
Step 2, performing DC removal processing on the echo signals, and obtaining time-frequency signals by adopting short-time Fourier transformation; the method comprises the following steps:
step 2-1, performing DC removal processing on the I, Q two paths of echo signals, namely firstly obtaining the average value of the I, Q two paths of echo signals, and then subtracting the average value by using the I, Q two paths of echo signals respectively;
and 2-2, storing the I, Q echo signals after DC removal processing into echo signals as a real part and an imaginary part respectively, segmenting the echo signals according to m sampling points of each segment, overlapping two adjacent segments by n%, wherein m and n are natural numbers, windowing each segment of echo signals by using a Hamming window function, and performing fast Fourier transform to obtain the frequency spectrum of the echo signals, namely the time-frequency signal.
Preferably, m has a value of 256, 512 or 1024, and n has a value in the range of 30 to 70.
Step 3, acquiring the corresponding rotating speed of the blade in real time; the method comprises the following steps:
step 3-1, marking the time-frequency signal, wherein the positive frequency part of the marked time-frequency signal is marked with a first complete leaf-shaped frequency spectrum l along the time axis direction from a certain time t 0 A second complete leaf-shaped frequency spectrum in the direction of the time axis is l 1 The third complete leaf-shaped frequency spectrum along the time axis direction is l 2 From l 0 Three leaf-shaped frequency spectrums in the reverse time axis direction are respectively marked as l -1 、l -2 And l -3 The method comprises the steps of carrying out a first treatment on the surface of the The positive frequency part of the time-frequency signal comprises zero frequency;
step 3-2, calculating a third leaf-like spectrum l in the reverse time axis direction -3 And a first leaf-like spectrum l in the time axis direction 0 The interval of the starting point on the time axis is T 0 A second leaf-like spectrum l in the reverse time axis direction -2 And a second foliated spectrum l along the time axis direction 1 The interval of the starting point on the time axis is T 1 Inverse time axis squareFirst leaf-like spectrum of direction l -1 And a third leaf-like spectrum l in the time axis direction 2 The interval of the starting point on the time axis is T 2 ；
Step 3-3, calculating T 0 、T 1 And T 2 The formula used is:
the rotation speed of the blade at the moment t is as follows:
step 4, judging the damage type of the blade to obtain a judging result, specifically:
step 4-1, dividing a time-frequency signal with positive frequency into a plurality of periods; the specific method comprises the following steps:
the positive frequency part of the time-frequency signal is marked as a first period from a first complete foliate frequency spectrum and zero frequency intersection point to a fourth foliate frequency spectrum and zero frequency intersection point along the time axis direction, the period is the period of blade rotation, three foliate frequency spectrums are contained in the period, and the three foliate frequency spectrums are respectively marked as a blade 1, a blade 2 and a blade 3 from left to right correspondingly;
marking the second period from the intersection point of the fourth leaf-shaped frequency spectrum and the zero frequency to the intersection point of the seventh leaf-shaped frequency spectrum and the zero frequency along the time axis direction, marking the third period from the intersection point of the seventh leaf-shaped frequency spectrum and the zero frequency to the intersection point of the tenth leaf-shaped frequency spectrum and the zero frequency along the time axis direction, and so on, dividing a time-frequency signal with positive frequency into a plurality of periods; the positive frequency part of the time-frequency signal comprises zero frequency;
step 4-2, extracting the duration T of the kth period based on the division of step 4-1 k Extracting the maximum Doppler frequency f of the leaf spectrum corresponding to the blade to be detected Dop,max1 The light speed is c, and the radar working frequency is f c Then the length of the blade to be tested in the period is testedThe values are:
determining the deviation between the test value of the blade to be tested and the standard value of the installation machine, if the test value meets the judging condition of fracture damage, considering that the blade has fracture damage, and executing the same operation on the rest blades to determine whether the blade has fracture damage or not;
the judgment conditions of the fracture damage are as follows: the deviation between the true value and the standard value of the blade to be tested in more than 8 periods in 10 continuous periods is 1m or more;
step 4-3, extracting energy mean values of three leaf-shaped spectrums in the mth period to be E respectively on the basis of the division of the step 4-1 1 、E 2 、E 3 Calculate the mean mu e ＝(E 1 +E 2 +E 3 ) And 3, the mean square error is:
the blade energy variation coefficient for this period is:
if the judging conditions of the surface corrosion damage are met, the surface corrosion damage of the blade is considered;
the judging conditions of the surface corrosion damage are as follows: blade energy variation coefficients of more than 8 periods in 10 continuous periods are more than 5%;
step 4-4, based on the division of step 4-1, at a duration of T n In the nth period of (2), the intersection point of the foliated spectrum corresponding to the blades 2 and 3 and the zero frequency divides one period into three periods of time, and the three periods of time are recorded as t from left to right on the time axis 1 、t 2 And t 3 Calculate the mean μ of these three periods t ＝T n And 3, the mean square error is:
the time coefficient of variation for this period is:
if the judging condition of the angle mismatch damage is met, the three blades are considered to have the angle mismatch damage;
the judging conditions of the angle mismatch damage are as follows: the time variation coefficient of more than 8 periods in 10 continuous periods is more than 5%.
The method can realize non-contact, comprehensive and accurate monitoring of the health condition of the blade by only a single radio frequency sensor, can discover the damage of the blade in time, and has the advantages of effectiveness, feasibility, reliable performance and high accuracy.
Referring to fig. 1, the wind power blade damage monitoring system based on the doppler radar provided by the invention comprises a signal acquisition device, a time-frequency analysis device, a rotation speed monitoring device and a blade damage monitoring device which are connected in sequence, wherein:
the signal acquisition device comprises: collecting echo signals (I, Q two paths of echo signals) of a blade of the wind driven generator by using a Doppler radar, and sampling the echo signals;
time-frequency analysis device: performing DC removal processing on the I, Q two paths of echo signals, and obtaining a time-frequency signal by adopting short-time Fourier transform;
rotation speed monitoring device: inputting a time-frequency signal, and obtaining the corresponding rotating speed of the blade in real time through a rotating speed monitoring device;
blade damage monitoring device: inputting time-frequency signals, establishing a blade damage monitoring algorithm based on time-frequency multi-feature extraction, and obtaining blade damage type judgment, wherein the damage type comprises fracture, surface corrosion and angle mismatch.
Further, the signal acquisition device: the Doppler radar is utilized to collect echo signals (I, Q two paths of echo signals) of the wind driven generator blade, and the echo signals are sampled, specifically:
step 1-1, erecting a single Doppler radar at a position which is far away from a machine column by x meters and is right below a wind driven generator, wherein the value of x is 5-10 m, and a radar antenna is upwards aligned with the rotation center of a blade and continuously records echo signals of the blade for more than 10 minutes;
step 1-2, sampling the recorded signals in a I, Q two-way mode, and inputting the signals into a computer.
Further, the time-frequency analysis device: performing DC removal processing on the I, Q two paths of echo signals, and obtaining a time-frequency signal by adopting short-time Fourier transform, wherein the method specifically comprises the following steps:
step 2-1, taking the average value of the I, Q echo signals, namely DC I ＝mean(signal_I)，DC Q =mean (signal_q), and the average value of the I, Q echo signals is subtracted from each other, i.e., sig_i=signal_i-DC I ，Sig_Q＝signal_I-DC Q Namely, direct current removal treatment is carried out;
and 2-2, respectively storing I, Q paths of echo signals after DC removal into the echo signals as a real part and an imaginary part, segmenting the echo signals according to m sampling points of each segment, overlapping two adjacent segments by n%, wherein m is 256, 512, 1024 and the like, and n is 30-70. And windowing each section by using a Hamming window function, and performing fast Fourier transform to obtain the frequency spectrum of the echo signal, namely the time-frequency signal.
Further, the rotation speed monitoring device: inputting a time-frequency signal, and obtaining the corresponding rotating speed of the blade in real time through a rotating speed monitoring device, wherein the method comprises the following steps of:
step 3-1, referring to fig. 4, the positive frequency part (including zero frequency) of the marked time-frequency signal is marked with a first complete leaf-like frequency spectrum l in the time axis direction from a certain time t 0 The second bar is l 1 The third is l 2 From l 0 Three leaf-shaped frequency spectrums in the reverse time axis direction are respectively marked as l -1 、l -2 And l -3 ；
Step 3-2, calculating a leaf-like frequency spectrum l -3 And l 0 The interval of the starting point on the time axis is T 0 ，l -2 And l 1 The interval of the starting point on the time axis is T 1 ，l -1 And l 2 The interval of the starting point on the time axis is T 2 ；
Step 3-3, calculating T 0 、T 1 And T 2 The average value of (2) is:
the rotation speed of the blade at the moment t is as follows:
further, the blade damage monitoring device: inputting time-frequency signals, establishing a blade damage monitoring algorithm based on time-frequency multi-feature extraction, and obtaining blade damage type judgment, wherein the damage type comprises fracture, surface corrosion and angle mismatch, and specifically comprises the following steps:
and step 4-1, marking a first period from a first complete leaf-shaped frequency spectrum and zero frequency intersection point to a fourth leaf-shaped frequency spectrum and zero frequency intersection point along the time axis direction in a positive frequency part (including zero frequency) of the time-frequency signal as a period of blade rotation, wherein the period comprises three leaf-shaped frequency spectrums corresponding to three blades, and marking the three leaf-shaped frequency spectrums as a blade 1, a blade 2 and a blade 3 from left to right. Marking the second period from the intersection point of the fourth leaf-shaped frequency spectrum and the zero frequency to the intersection point of the seventh leaf-shaped frequency spectrum and the zero frequency along the time axis direction, marking the third period from the intersection point of the seventh leaf-shaped frequency spectrum and the zero frequency to the intersection point of the tenth leaf-shaped frequency spectrum and the zero frequency along the time axis direction, and so on, dividing a time-frequency signal with positive frequency into a plurality of periods;
step 4-2, referring to FIG. 5, the duration T of the kth cycle is extracted according to step 4-1 k Extracting the maximum Doppler frequency f of the leaf 1 corresponding to the leaf-like frequency spectrum Dop,max1 The light speed is c, and the radar working frequency is f c The test value of the length of the blade 1 in this cycle is:
and calculating the deviation of the test value of the blade 1 and the installed standard value. If the blade 1 deviates from the standard value by 1m or more over 8 cycles in 10 consecutive cycles, the blade 1 is considered to have fracture damage. The same is performed for the blades 2 and 3 to determine if they have fracture damage;
step 4-3, referring to FIG. 6, according to step 4-1, extracting energy average values of three leaf spectrums in the mth period as E respectively 1 、E 2 、E 3 Calculate the mean mu e ＝(E 1 +E 2 +E 3 ) And 3, the mean square error is:
the blade energy variation coefficient for this period is:
if the energy variation coefficient of the blade in more than 8 periods in 10 continuous periods is more than 5%, the blade is considered to have surface corrosion damage;
step 4-4, referring to FIG. 7, according to step 4-1, at a duration T n In the nth period of (2), the intersection point of the foliated spectrum corresponding to the blades 2 and 3 and the zero frequency divides one period into three periods of time, and the three periods of time are recorded as t from left to right on the time axis 1 、t 2 And t 3 Calculate the mean μ of these three periods t ＝T n And 3, the mean square error is:
the time coefficient of variation for this period is:
if the time variation coefficient of more than 8 periods in 10 continuous periods is more than 5%, the three blades are considered to have angle mismatch damage.
According to the invention, the non-contact measurement of the rotating speed of the blade can be realized by using a single Doppler radar, compared with a traditional sensor in the cabin, the operation is more convenient, the feedback time is shorter, and the rotating speed information can be obtained in a short time by only holding the radar to aim at the blade; according to the invention, the real-time detection of three types of damage to the blade can be realized by using a single Doppler radar under the condition of not contacting the blade and not affecting the normal operation of the blade, so that the influence of artificial factors of manual observation is avoided, and the method is more objective and accurate; the invention takes the radar as one of the radio frequency sensors, has little influence by the environment, can work in all weather and under all weather conditions, and is stable and reliable.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of:
step 1, collecting echo signals of blades of a wind driven generator in real time, and sampling the echo signals;
step 2, performing DC removal processing on the echo signals, and obtaining time-frequency signals by adopting short-time Fourier transformation;
step 3, acquiring the corresponding rotating speed of the blade in real time;
and 4, judging the damage type of the blade to obtain a judging result.
The specific steps are consistent with those described above, and will not be described again here.
The method is simple and effective, reliable in performance and convenient to implement.
The present invention will be described in further detail with reference to examples.
Examples
In this embodiment, the system and the method for monitoring the rotation speed and damage of the blade in real time are adopted to test the wind driven generator, and the system and the method comprise the following steps:
1. and (3) rotating speed test:
ten wind generators were tested, and the actual and test values of the blade rotational speeds are shown in table 1.
Table 1 results of rotational speed test
| Actual value (rpm) | 5 | 5 | 5 | 10 | 10 |
| Test value (rpm) | 4.90 | 5.02 | 4.95 | 10.15 | 10.02 |
| Test error | 2％ | 0.2％ | 1％ | 1.5％ | 0.2％ |
| Actual value (rpm) | 10 | 10 | 15 | 15 | 15 |
| Test value (rpm) | 9.89 | 9.97 | 15.11 | 14.85 | 15.14 |
| Test error | 1.1％ | 0.3％ | 0.73％ | 1％ | 0.93％ |
The error between the blade rotating speed test value and the actual value obtained by the invention is within 2% and can be calculated by the table 1, so that the blade rotating speed can be accurately monitored.
2. Blade fracture damage test:
ten wind power generators were tested, wherein four wind power generators had blades with fracture damage, the standard value of the blade length installation was 50m, and the test results are shown in table 2.
TABLE 2 fracture injury test results
As can be seen from Table 2, the judging accuracy of the invention reaches 90%, and the damaged fan with the number of 3 is judged to be undamaged due to one judging error.
3. Blade surface corrosion damage test:
ten wind power generators were tested, three of which had blade surface corrosion damage, and the test results are shown in table 3.
TABLE 3 Corrosion damage test results
As can be seen from Table 3, the judging accuracy of the invention reaches 80%, and the judging error is twice, the damaged fan with the number of 7 is judged to be undamaged, and the non-damaged cold machine with the number of 10 is judged to be damaged.
4. Blade mismatch damage test:
ten wind generators were tested, of which two wind generator blades had angle mismatch damage, and the test results are shown in table 4.
Table 4 mismatch damage test results
As can be seen from Table 4, the judging accuracy of the invention reaches 90%, and the nondestructive cold machine with the number of 9 is judged to be damaged due to one-time judging error.
In summary, the system and the method for monitoring the rotating speed and the damage of the blade in real time can realize the monitoring of the rotating speed of the blade and the monitoring and the judgment of the damage condition of the blade in real time. Compared with the existing blade damage monitoring method, the system provided by the invention can realize non-contact, comprehensive and accurate monitoring of the health condition of the blade by only a single radio frequency sensor, can be timely found when the blade is damaged, and is effective, feasible, reliable in performance and high in accuracy.
Claims (8)
1. The wind power blade damage monitoring method based on the Doppler radar is characterized by comprising the following steps of:
step 1, collecting echo signals of blades of a wind driven generator in real time, and sampling the echo signals;
step 2, performing DC removal processing on the echo signals, and obtaining time-frequency signals by adopting short-time Fourier transformation;
step 3, acquiring the corresponding rotating speed of the blade in real time; the method comprises the following steps:
step 3-1, marking the time-frequency signal, wherein the positive frequency part of the marked time-frequency signal is marked with a first complete leaf-shaped frequency spectrum l along the time axis direction from a certain time t 0 A second complete leaf-shaped frequency spectrum in the direction of the time axis is l 1 The third complete leaf-shaped frequency spectrum along the time axis direction is l 2 From l 0 Three leaf-shaped frequency spectrums in the reverse time axis direction are respectively marked as l -1 、l -2 And l -3 The method comprises the steps of carrying out a first treatment on the surface of the The positive frequency part of the time-frequency signal comprises zero frequency;
step 3-2, calculating a third leaf-like spectrum l in the reverse time axis direction -3 And a first leaf-like spectrum l in the time axis direction 0 The interval of the starting point on the time axis is T 0 A second leaf-like spectrum l in the reverse time axis direction -2 And a second foliated spectrum l along the time axis direction 1 The interval of the starting point on the time axis is T 1 A first leaf-like spectrum l in the reverse time axis direction -1 And a third leaf-like spectrum l in the time axis direction 2 The interval of the starting point on the time axis is T 2 ；
Step 3-3, calculating T 0 、T 1 And T 2 The formula used is:
the rotation speed of the blade at the moment t is as follows:
step 4, judging the damage type of the blade to obtain a judging result; judging the damage type of the blade, wherein the judging result is specifically:
step 4-1, dividing a time-frequency signal with positive frequency into a plurality of periods;
step 4-2, extracting the duration T of the kth period based on the division of step 4-1 k Extracting the maximum Doppler frequency f of the leaf spectrum corresponding to the blade to be detected Dop,max1 The light speed is c, and the radar working frequency is f c The test value of the length of the blade to be tested in the period is:
determining the deviation between the test value of the blade to be tested and the standard value of the installation machine, if the test value meets the judging condition of fracture damage, considering that the blade has fracture damage, and executing the same operation on the rest blades to determine whether the blade has fracture damage or not;
step 4-3, extracting energy mean values of three leaf-shaped spectrums in the mth period to be E respectively on the basis of the division of the step 4-1 1 、E 2 、E 3 Calculate the mean mu e ＝(E 1 +E 2 +E 3 ) And 3, the mean square error is:
the blade energy variation coefficient for this period is:
if the judging conditions of the surface corrosion damage are met, the surface corrosion damage of the blade is considered;
step 4-4, based on the division of step 4-1, at a duration of T n In the nth period of (2), the intersection point of the foliated spectrum corresponding to the blades 2 and 3 and the zero frequency divides one period into three periods of time, and the three periods of time are recorded as t from left to right on the time axis 1 、t 2 And t 3 Calculate the mean μ of these three periods t ＝T n And 3, the mean square error is:
the time coefficient of variation for this period is:
if the judging condition of the angle mismatch damage is met, the three blades are considered to have the angle mismatch damage.
2. The method for monitoring wind power blade damage based on Doppler radar as claimed in claim 1, wherein step 1 is to collect echo signals of wind power generator blades in real time and sample the echo signals, specifically:
step 1-1, erecting a single Doppler radar under a wind driven generator, aligning a radar antenna upwards to the rotation center of a blade, and continuously recording blade echo signals;
step 1-2, sampling the recorded signals in a I, Q two-way mode.
3. The method for monitoring wind power blade damage based on Doppler radar as claimed in claim 1, wherein the step 2 is to perform DC removal processing on the echo signal, and to obtain a time-frequency signal by short-time Fourier transform, specifically:
step 2-1, performing DC removal processing on the I, Q two paths of echo signals, namely firstly obtaining the average value of the I, Q two paths of echo signals, and then subtracting the average value by using the I, Q two paths of echo signals respectively;
and 2-2, storing the I, Q echo signals after DC removal processing into echo signals as a real part and an imaginary part respectively, segmenting the echo signals according to m sampling points of each segment, overlapping two adjacent segments by n%, wherein m and n are natural numbers, windowing each segment of echo signals by using a Hamming window function, and performing fast Fourier transform to obtain the frequency spectrum of the echo signals, namely the time-frequency signal.
4. A method for monitoring damage to a wind turbine blade based on doppler radar as claimed in claim 3, wherein in step 2-2, m is 256, 512 or 1024, and n is 30-70.
5. The method for monitoring wind power blade damage based on Doppler radar according to claim 1, wherein the specific method for dividing the time-frequency signal with positive frequency in step 4-1 is as follows:
the positive frequency part of the time-frequency signal is marked as a first period from a first complete foliate frequency spectrum and zero frequency intersection point to a fourth foliate frequency spectrum and zero frequency intersection point along the time axis direction, the period is the period of blade rotation, three foliate frequency spectrums are contained in the period, and the three foliate frequency spectrums are respectively marked as a blade 1, a blade 2 and a blade 3 from left to right correspondingly; marking the second period from the intersection point of the fourth leaf-shaped frequency spectrum and the zero frequency to the intersection point of the seventh leaf-shaped frequency spectrum and the zero frequency along the time axis direction, marking the third period from the intersection point of the seventh leaf-shaped frequency spectrum and the zero frequency to the intersection point of the tenth leaf-shaped frequency spectrum and the zero frequency along the time axis direction, and so on, dividing a time-frequency signal with positive frequency into a plurality of periods; the positive frequency portion of the time-frequency signal includes a zero frequency.
6. The method for monitoring damage to a wind turbine blade based on doppler radar according to claim 5, wherein the condition for determining the fracture damage in step 4-2 is: the deviation between the true value and the standard value of the blade to be tested in more than 8 periods in 10 continuous periods is 1m or more;
the judging conditions of the surface corrosion damage in the step 4-3 are as follows: blade energy variation coefficients of more than 8 periods in 10 continuous periods are more than 5%;
the judging conditions of the angle mismatch damage in the step 4-4 are as follows: the time variation coefficient of more than 8 periods in 10 continuous periods is more than 5%.
7. A Doppler radar-based wind power blade damage monitoring system based on the method of any one of claims 1 to 6 is characterized by comprising a signal acquisition device, a time-frequency analysis device, a rotation speed monitoring device and a blade damage monitoring device which are sequentially connected, wherein,
the signal acquisition device is used for acquiring echo signals of the wind driven generator blade and sampling the echo signals;
the time-frequency analysis device is used for carrying out DC removal on the echo signals, obtaining time-frequency signals by adopting short-time Fourier transformation, and then transmitting the time-frequency signals to the rotating speed monitoring device;
the rotating speed monitoring device is used for acquiring the rotating speed corresponding to the blade in real time;
the blade damage monitoring device is used for judging the type of blade damage, and the type of damage comprises fracture, surface corrosion and angle mismatch.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 6 when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911199627.3A CN112882017B (en) | 2019-11-29 | 2019-11-29 | Wind power blade damage monitoring method and system based on Doppler radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911199627.3A CN112882017B (en) | 2019-11-29 | 2019-11-29 | Wind power blade damage monitoring method and system based on Doppler radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112882017A CN112882017A (en) | 2021-06-01 |
| CN112882017B true CN112882017B (en) | 2023-11-21 |
Family
ID=76038472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911199627.3A Active CN112882017B (en) | 2019-11-29 | 2019-11-29 | Wind power blade damage monitoring method and system based on Doppler radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112882017B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201020408D0 (en) * | 2010-12-01 | 2011-08-17 | Bae Systems Plc | Radar system and method of providing a blade count of a rotating bladed target |
| CN105891784A (en) * | 2016-04-11 | 2016-08-24 | 中国民航大学 | Wind turbine orientation estimation method during radar wind power farm clutter suppression |
| CN106291482A (en) * | 2016-09-29 | 2017-01-04 | 三峡大学 | A kind of wind turbine blade radar echo signal Doppler frequency spectrum method for solving |
| CN106646397A (en) * | 2017-01-21 | 2017-05-10 | 三峡大学 | Radar echo solving method of wind motor in terrestrial background |
| CN108387881A (en) * | 2018-02-01 | 2018-08-10 | 三峡大学 | A kind of accurate simulation algorithm of wind turbine blade echo |
| CN109557514A (en) * | 2019-01-14 | 2019-04-02 | 三峡大学 | A kind of accurate method for solving of wind turbine blade echo |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2663886A2 (en) * | 2011-01-11 | 2013-11-20 | Ophir Corporation | Monitoring complex flow fields for wind turbine applications |
-
2019
- 2019-11-29 CN CN201911199627.3A patent/CN112882017B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201020408D0 (en) * | 2010-12-01 | 2011-08-17 | Bae Systems Plc | Radar system and method of providing a blade count of a rotating bladed target |
| CN105891784A (en) * | 2016-04-11 | 2016-08-24 | 中国民航大学 | Wind turbine orientation estimation method during radar wind power farm clutter suppression |
| CN106291482A (en) * | 2016-09-29 | 2017-01-04 | 三峡大学 | A kind of wind turbine blade radar echo signal Doppler frequency spectrum method for solving |
| CN106646397A (en) * | 2017-01-21 | 2017-05-10 | 三峡大学 | Radar echo solving method of wind motor in terrestrial background |
| CN108387881A (en) * | 2018-02-01 | 2018-08-10 | 三峡大学 | A kind of accurate simulation algorithm of wind turbine blade echo |
| CN109557514A (en) * | 2019-01-14 | 2019-04-02 | 三峡大学 | A kind of accurate method for solving of wind turbine blade echo |
Non-Patent Citations (2)
| Title |
|---|
| 基于动态RCS的风电机叶片多普勒特性;唐波 等;《高电压技术》;第43卷(第10期);全文 * |
| 风轮机叶片雷达散射特性计算与解析式拟合;何炜坤 等;《仪器仪表学报》;第38卷(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112882017A (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Status and problems of wind turbine structural health monitoring techniques in China | |
| CN107796611B (en) | Alarm system for detecting abnormal work of wind driven generator | |
| CN102798529B (en) | Method and system for diagnosing bearing faults of large-size wind turbine bearing | |
| CN105547698A (en) | Fault diagnosis method and apparatus for rolling bearing | |
| CN103940611B (en) | Rolling bearing self adaptation method for detecting abnormality under a kind of Wind turbines variable working condition | |
| Refaat et al. | ANN-based for detection, diagnosis the bearing fault for three phase induction motors using current signal | |
| CN107782443B (en) | Automatic extraction method for natural frequency of wind driven generator blade | |
| CN103760243A (en) | Microcrack nondestructive detecting device and method | |
| CN104677623A (en) | On-site acoustic diagnosis method and monitoring system for wind turbine blade failure | |
| CN104101652B (en) | Audio signal based wind power blade damage monitoring method and system | |
| CN111400961B (en) | Wind generating set blade fault judging method and device | |
| CN107781118A (en) | Blade of wind-driven generator health status monitoring system based on multi-sensor information | |
| CN109236587B (en) | Alarm system for detecting abnormal work of wind driven generator | |
| CN104596766A (en) | Early fault determining method for bearing | |
| CN111120388B (en) | Fan state combined monitoring method and system | |
| CN111890126B (en) | Early turning flutter early warning and monitoring method based on sound pressure energy kurtosis index | |
| Lin et al. | A review and strategy for the diagnosis of speed-varying machinery | |
| CN112882017B (en) | Wind power blade damage monitoring method and system based on Doppler radar | |
| CN114565571A (en) | Fan blade defect detection method and device based on computer vision | |
| CN212454697U (en) | Wind generating set blade running state detection device | |
| CN110469460B (en) | Fault detection method and system for wind driven generator | |
| CN111173687A (en) | On-line monitoring device and method for crack damage of wind power fan blade | |
| CN112734001A (en) | Wind power transmission chain intelligent fault diagnosis method based on order spectrum migration | |
| CN110633686A (en) | Equipment rotating speed identification method based on vibration signal data driving | |
| CN113323803B (en) | Variable pitch bearing detection method based on dynamic control of fan |
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 |