JP2017161225A - Abnormality sign diagnosis system for wind power generation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an abnormality sign diagnosis system for a wind power generation system for detecting the initial stage of the damage with high accuracy.SOLUTION: An abnormality sign diagnosis system for a wind power generation system includes: a sensor 50 disposed near main shaft bearings 13 and 14; a storage unit that holds training data generated based on the output of the sensor 50; a control unit that generates evaluation data on the basis of the output of the sensor 50 obtained after the training data are generated; and a diagnosis unit that diagnoses the abnormality sign of the main shaft bearings by comparing the training data and the evaluation data. The main shaft bearings 13 and 14 include at least an outer ring portion that is rotatable, an inner ring portion that is hollow, and a rotor portion disposed between the outer ring portion and the inner ring portion. The sensor 50 is provided to an inner wall of the inner ring portion.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an abnormality sign diagnosis system for a wind power generation system.

Various systems for diagnosing abnormalities in equipment provided in a wind turbine generator have been proposed.
Patent Document 1 discloses a state monitoring system for diagnosing an abnormality of a device provided in a wind turbine generator, and the state monitoring system includes a monitor device including a sensor provided in the device, and the monitor device diagnoses an abnormality of the device. It is described that a monitoring-side control device that sets a threshold value to be used in order to diagnose a device abnormality based on the threshold value and a monitoring terminal that monitors the state of the device is described.

  Patent Document 2 discloses a state monitoring method for mechanical equipment including at least one of a rotating body and a sliding member, and analyzes a predetermined physical quantity of the mechanical equipment based on a signal generated from the mechanical equipment. The results of the analysis and the information that is the diagnostic criteria for the presence or absence of abnormalities in the mechanical equipment are compared and verified every first time to make a provisional diagnosis of the presence or absence of abnormalities in the mechanical equipment. Or, for every second time, based on the result of comparison and collation, if the number of times of tentative diagnosis of abnormality is equal to or greater than a threshold value, comprehensive evaluation is performed to determine whether there is an abnormality in the mechanical equipment and the abnormal part. Diagnose is described.

  Patent Document 3 discloses a monitoring system that monitors the state of a device of a wind turbine generator, and is arranged inside a casing that holds the device inside, and includes a detection unit that acquires sound information, and a detection unit. A data collection unit that collects acquired sound information and transfers it to a server on the Internet, a comparison analysis unit that compares and analyzes a reference value based on the sound information transferred from the data collection unit, and a comparison analysis unit And a monitoring terminal for displaying and monitoring the results of comparison and analysis according to the above.

  Further, Patent Document 4 describes a method for selecting the installation position of a vibration sensor with respect to a bearing device using an analysis result obtained by a vibration analysis method for analyzing the vibration of the bearing device by a computer in a rolling bearing state monitoring device. ing.

International Publication No. 2013/133002 JP 2005-17128 A JP2015-63916A JP2015-31626A

  With the increase in wind power generators that have been in operation for more than 10 years, wind turbine generators have been found to have long-term shutdowns due to the failure of the main shaft bearing, which is a major component of the wind power generator.

  In the department according to the present invention, which is in charge of maintenance, the service staff has performed periodic inspections of the wind power generator, and the service staff has made an abnormality determination by visually checking the noise of the spindle bearing and the grease of the bearing. However, the abnormality determination has a problem that it is greatly influenced by the experience of the service staff and that it is difficult to detect the abnormality until the degree of damage reaches the final stage.

  In Patent Documents 1, 2, and 4, a sensor is disposed in the bearing housing. Moreover, in patent document 3, the sensor is arrange | positioned in the nacelle. In any of the arrangement methods, there is a problem that sufficient abnormality cannot be detected in the verification result of this department.

  In particular, in the case of a synchronous wind power generator that does not have a gearbox, the measurement target is only a spindle bearing that rotates at a low speed, and in addition, the influence of disturbances such as wind vibration is significant in the low-speed rotation range. Become. Therefore, in order to detect abnormalities in the spindle bearings at an early stage, there has been a demand for a means for detecting the initial stage of damage (a sign of failure) quantitatively and with high accuracy.

  The present invention is an invention for solving the above-described problem, and an object thereof is to provide an abnormality sign diagnosis system for a wind power generation system capable of accurately detecting an initial stage of damage.

  In order to achieve the above object, an abnormality sign diagnosis system for a wind power generation system according to the present invention includes a sensor unit (for example, sensor 50) installed in the vicinity of a spindle bearing, and teacher data (for example, based on the output of the sensor unit). , A storage unit that holds normal data 31), a control unit that creates evaluation data based on the output of the sensor unit acquired after the teacher data is created (for example, processing in the abnormality degree calculation unit 23), a teacher A diagnostic unit (for example, processing in the abnormality degree determination unit 24) that diagnoses an abnormality sign of the spindle bearing by comparing the data and the evaluation data. The spindle bearing is at least a rotatable outer ring part (for example, outer ring 131), a hollow inner ring part (for example, inner ring 134), and a rotating body part (for example, an inner ring part, for example). Rolling element 132), and the sensor portion is installed on the inner wall of the inner ring portion. Other aspects of the present invention will be described in the embodiments described later.

  According to the present invention, the initial stage of damage can be detected with high accuracy.

BRIEF DESCRIPTION OF THE DRAWINGS It is a wind power generator of the wind power generation system which concerns on 1st Embodiment, (a) is a windmill external view, (b) is an enlarged view inside a nacelle containing a diagnostic apparatus. It is a figure which shows the structure of a diagnostic apparatus. It is a figure which shows the wind speed difference up and down a nacelle. It is a figure which shows the case where the sensor is arrange | positioned to the main axis | shaft. It is a figure which shows the example of a frequency analysis, (a) is a case where it is normal data, (b) is a case where a main shaft bearing is damaged. It is a figure which shows the example of AE measurement data, (a) is a case where it is normal data, (b) is a case where a spindle bearing is damaged. It is a flowchart which shows the process of the sign diagnosis of a diagnostic apparatus. It is a figure which shows the example of a display of a sign diagnosis result. It is a wind power generator concerning a 2nd embodiment, (a) is a windmill appearance figure, and (b) is an enlarged view inside a nacelle. It is a figure which shows the virtual plane in a spindle bearing.

<< first embodiment >>
DESCRIPTION OF EMBODIMENTS Embodiments for carrying out the present invention will be described in detail with reference to the drawings as appropriate.
FIG. 1 is a wind turbine generator 10 (wind turbine) of the wind turbine generator system according to the first embodiment, (a) is an external view of the wind turbine, and (b) is an enlarged view inside the nacelle 3 including the diagnostic device 100. is there. In the present embodiment, a direct drive wind power generator (direct drive wind power generator) will be described as an example. The wind power generator 10 includes a blade 2 mounted on the hub 1, a nacelle 3 (housing), and a tower 4 (support) that supports the nacelle 3. Inside the nacelle 3, the fixed main shaft 11 (first main shaft) that is hollow and supported by the support portion 19 of the tower 4, and the generator 5 is disposed on the outer periphery of the main shaft 11.

  The generator 5 includes a stator 6 and a generator rotor 7. The stator 6 is fixed to the stator casing 9. The generator rotor 7 includes a field magnet and a rotor plate 8 that supports the field magnet, and the rotor plate 8 is coupled to the main shaft 12 (second main shaft) on the rotation side.

  The main shaft 12 is supported on the main shaft 11 by two main shaft bearings 13 and 14 (main bearings) and is connected to the hub 1. Further, the protrusion 16 of the rotor plate 8 is supported by the stator casing 9 by a third bearing 15.

  A service person who performs periodic inspection of the wind turbine generator 10 installs the sensor 50 in the vicinity of the inner rings of the spindle bearings 13 and 14. More specifically, a sensor 50 for measuring and monitoring the state (operation state signal) of the main shaft bearings 13 and 14 is installed in a hollow portion of the main shaft 11 and located radially inward of the main shaft bearings 13 and 14. . The sensor 50 is good to be located on each virtual plane VP (refer FIG. 10) which passes along the axial center of the spindle bearings 13 and 14. The sensor 50 is at least one of an acoustic sensor, a vibration sensor, and an AE sensor. Note that AE (Acoustic Emission) is a phenomenon in which elastic energy stored inside is released as sound waves (elastic waves, AE waves) when a material is deformed or destroyed. The AE wave mainly has a high frequency component in the ultrasonic region (several tens of kHz to several MHz).

  FIG. 10 is a diagram illustrating a virtual plane VP in the spindle bearing. FIG. 10 is a perspective explanatory view. Reference is made to FIG. 4 as appropriate. The sensor 50 includes an outer peripheral portion OP connecting a plurality of axial center points C1 (first center points) of the outer wall of the outer ring 131 of the spindle bearing 13 and a center point C2 of the spindle bearing 13 in the vertical direction of the center point C1. It is good to arrange | position on the virtual plane VP which ties (2nd center point). More preferably, the center point C3 (not shown) of the sensor 50 is disposed on the virtual plane VP. The sensor 50 may be installed so that the inner wall of the inner ring 134 (see FIG. 4) contacts the detection surface of the sensor 50 and faces the outer wall 131 (see FIG. 4) of the main shaft bearing. The same applies to the main shaft bearing 14.

  More specifically, the main shaft bearing 13 has a predetermined length in the axial direction. Since the center of this length is the center point C1 and the main shaft is generally cylindrical, the center point C1 exists infinitely along the circumferential direction of the cylinder inner surface. A “ring” consisting of a set of center points C1 can be formed. The point where the sum of the distances from all the points constituting the “ring” is the smallest is the “center point C2”. Here, if the “ring” is a circle, the center point C2 is the center of the circle. A region that is closed by the “ring” and includes the center point C2 is a virtual plane VP. In the present embodiment, it is desirable that the center point of the sensor 50 is on this virtual plane.

  Returning to FIG. 1, the diagnostic device 100 acquires sensor data from the sensor 50 via the amplifier 51, and the acquired sensor data (evaluation data) and normal stored in advance in the storage unit 30 (see FIG. 2). The data (teacher data) is compared to diagnose whether or not the spindle bearings 13 and 14 are abnormal. Specifically, the diagnostic device 100 detects a sign of failure of the spindle bearings 13 and 14. In addition, although the diagnostic apparatus 100 is arrange | positioned in the tower 4 in FIG.1 (b), it is not necessarily limited to this, You may arrange | position in the management ridge on the ground.

  FIG. 2 is a diagram illustrating a configuration of the diagnostic apparatus 100. As illustrated in FIG. 2, the diagnostic device 100 includes a processing unit 20, a storage unit 30, an input unit 41, a display unit 42, and a communication unit 43. The processing unit 20 acquires a detection signal from the sensor 50 at a normal time (when the spindle bearings 13 and 14 are operating normally) and a sensor data acquisition unit 21 that acquires a detection signal (sensor data) from the sensor 50. Normal data creation unit 22 stored in the storage unit 30 as normal data, an abnormality degree calculation unit 23 that calculates a deviation rate (abnormality) between the sensor data and the normal data, and an abnormality calculated by the abnormality degree calculation unit 23 An abnormality degree determination unit 24 that determines whether the degree is abnormal as compared with a predetermined threshold value. The storage unit 30 stores normal data 31, an abnormality degree 32, a determination result 33, and the like.

  When configured as an abnormality sign diagnosis system for a wind power generation system, a sensor unit (for example, sensor 50) installed in the vicinity of the spindle bearing and teacher data (for example, normal data 31) created based on the output of the sensor unit are used. A stored storage unit (for example, storage unit 30), a control unit that creates evaluation data based on the output of the sensor unit acquired after the creation time of the teacher data (for example, processing in the abnormality degree calculation unit 23), A diagnosis unit (for example, processing in the abnormality degree determination unit 24) that diagnoses an abnormality sign of the spindle bearing by comparing the teacher data and the evaluation data may be included.

  Usually, in the abnormality sign diagnosis system, data of a state in which the target device is operating normally is collected and “normal data” is held before diagnosis. Next, after this point of time, that is, after the diagnosis is started, the collected data is compared with normal data, and for example, the rotational speed of the windmill is calculated to determine whether the speed is abnormal or not. At this time, the data determined as “not abnormal” as a result of diagnosis can be newly reflected in “normal data”, and “normal data” can be updated.

  FIG. 3 is a diagram showing the difference in wind speed between the upper and lower nacelles 3. As shown in FIG. 3, the upper and lower wind speeds of the nacelle 3 are not constant but have a considerable difference in wind speed. In this case, a starting point of damage is generated in the spindle bearings 13 and 14 due to a moment or the like caused by the difference in wind speed between the upper and lower sides. The diagnostic apparatus 100 can accurately detect the initial stage of this damage by acquiring sensor data from the sensor 50 installed in the vicinity of the spindle bearings 13 and 14, particularly on the inner surface of the spindle 11 (first spindle). It becomes possible.

  FIG. 4 is a diagram showing a case where a sensor is arranged on the main shaft 11. The main shaft bearing 13 includes an outer ring 131, a rolling element 132 (for example, a roller), a cage 133 that determines the position of the rolling element 132, and an inner ring 134. The main shaft 11 is fitted to the inner ring 134, and the main shaft 12 is fitted to the outer ring 131. The sensor 50 is installed on the inner surface of the main shaft 11 via a sensor jig 52. The sensor 50 detects a sound (shock wave) generated when a defect occurs in the raceway surface of the outer ring 131 or the inner ring 134 and the rolling element 132 contacts the defect as a signal. The main shaft bearing 14 has the same configuration as the main shaft bearing 13.

  In FIG. 4, the case where the main shaft 11 and the inner ring 134 are independent members has been described. However, the inner ring 134 of the main shaft bearing 13 may serve as the main shaft. In that case, the inner ring 134 (inner ring portion) may be arranged so that the detection surface of the sensor 50 contacts the inner wall and faces the outer ring outer wall side of the spindle bearing 13.

Next, a diagnostic method using the diagnostic apparatus 100 will be described.
FIG. 5 is a diagram showing an example of frequency analysis, where (a) shows normal data and (b) shows a case where the spindle bearing is damaged. In the case of the normal data in FIG. 5A, when the frequency of the detection signal from the sensor 50 is analyzed, there is no frequency component having a singular point in amplitude. However, when the spindle bearings 13 and 14 shown in FIG. 5B are damaged, a frequency analysis of the detection signal generated from the damaged portion shows that the amplitude includes a harmonic component.

Generally, the damage frequency of a main shaft bearing (rolling bearing) is expressed by the following equation.
Outer ring damage: fout = Z / 2 · fr · (1-d / D · cosα) (1)
Inner ring damage: fin = Z / 2 · fr · (1 + d / D · cosα) (2)
Rolling element damage: fb = D / d · fr · (1− (d / D · cosα) ² ) (3)
Where fr: rotational frequency (Hz)
D: Pitch circle diameter (mm)
d: Rolling element diameter (mm)
Z: Number of rolling elements (pieces)
α: Contact angle (deg)

  6A and 6B are diagrams showing actual examples of AE measurement data. FIG. 6A shows the case of normal data, and FIG. 6B shows the case where the spindle bearing is damaged. PZT (lead zirconate titanate) is mainly used as the detection element of the AE sensor. Piezoelectric materials such as PZT have the property of generating an electric charge when a force is applied. The AE wave that has propagated through the surface of metal or the like is transmitted to the PZT inside the AE sensor, and the PZT is distorted and converted into an electrical signal.

  In the case of normal data in FIG. 6A, when the detection signal from the sensor 50 is subjected to frequency analysis, there is no frequency component having a singular point in intensity. However, when there is damage to the spindle bearings 13 and 14 in FIG. 6B, when the detection signal generated from the damaged portion is subjected to frequency analysis, it can be seen that the intensity includes a high intensity component around 8 Hz and 16 Hz. .

  FIG. 7 is a flowchart showing the predictive diagnosis process of the diagnostic apparatus 100. This will be described with reference to FIG. When the sensor data acquisition unit 21 receives the detection signal from the sensor 50 (processing S71), the abnormality degree calculation unit 23 analyzes the frequency of the detection signal of the detection signal for a predetermined period (see FIG. 6B) and stores it. The normal data 31 stored in the unit 30 is compared (processing S72). Specifically, the frequency analysis result shown on the right side of FIG. 6A is compared with the frequency analysis result on the right side of FIG. 6B to calculate the degree of abnormality. The degree-of-abnormality calculation unit 23 determines whether or not the frequency analysis result of the detection signal is within the specified range (processing S73). If the result is within the specified range (processing S73, within the specified range), the spindle bearing 13 and 14 are stored together with the determination time in the storage unit 30 as the determination result 33 and displayed on the display unit 42 (processing S75), and the process returns to the processing S71. On the other hand, if the abnormality degree calculation unit 23 is outside the specified range (processing S73, out of the specified range), the fact that the spindle bearings 13 and 14 are abnormal is stored in the storage unit 30 as the determination result 33 together with the determination time. Is displayed on the display unit 42 (process S74), and the process is terminated.

  The diagnosis apparatus 100 may diagnose whether or not the main bearing is abnormal by comparing a frequency characteristic obtained by performing a Fourier transform on normal data (teacher data) and a frequency characteristic obtained by performing a Fourier transform on evaluation data. Further, the diagnosis device 100 diagnoses that the main bearing is abnormal when the deviation rate between the frequency characteristic obtained by Fourier transforming normal data (teacher data) and the frequency characteristic obtained by Fourier transform of the evaluation data exceeds a predetermined value. May be.

  The diagnostic apparatus 100 may display the degree of abnormality based on the frequency characteristics obtained by performing Fourier transform on normal data (teacher data) and the frequency characteristics obtained by performing Fourier transform on evaluation data on the display unit. Moreover, the diagnostic apparatus 100 may display the degree of abnormality based on the frequency characteristic obtained by performing Fourier transform on normal data and the frequency characteristic obtained by performing Fourier transform on evaluation data on the display unit as a variation value with respect to the time axis.

  FIG. 8 is a diagram illustrating a display example of the sign diagnosis result. FIG. 8 is an example of the predictive diagnosis result displayed on the display unit 42 of the diagnostic apparatus 100, where the horizontal axis indicates the operation time and the vertical axis indicates the degree of abnormality. The operation time is obtained from a control device (not shown) of the wind power generator 10 through a communication line. Since the operation time is the time-series cumulative operation time data, when the determination time is adjusted to the time series, the degree of abnormality calculated from the comparison between the normal data and the detection signal (sensor data) is expressed with respect to the time axis. A graph can be displayed as a variation value. The diagnosis apparatus 100 displays the sign detection 81 when the degree of abnormality exceeds a predetermined threshold value for abnormality determination, and indicates the estimated operation time when the failure 82 occurs if the abnormality continues thereafter. Also good.

<< Second Embodiment >>
In the first embodiment, the sensor 50 is disposed on the inner surface of the main shaft 11 on the fixed side, but the present invention is not limited to this. In the second embodiment, a case where the sensor 50 is arranged on the inner surface of the rotation-side main shaft 11a will be described.

  FIG. 9 is a wind turbine generator 10 (windmill) according to the second embodiment, (a) is an external view of the windmill, and (b) is an enlarged view inside the nacelle 3. Also in the second embodiment, a direct drive wind power generator (direct drive wind power generator) will be described as an example. The same elements as those in the first embodiment shown in FIG.

  Inside the nacelle 3 of the wind turbine generator 10, there is a fixed main shaft 12 a (second main shaft) that is hollow and supported by the support portion 19 a of the tower 4. The rotation-side main shaft 11a (first main shaft) connected to the hub 1 is supported on the main shaft 12a by two main shaft bearings 13 and 14 (main bearing). A generator 5 is disposed at one end of the main shaft 11a.

  The generator 5 includes a stator 6 and a generator rotor 7. The stator 6 is fixed to the stator casing 9. The generator rotor 7 includes a field magnet and a rotor plate 8 that supports the field magnet, and the rotor plate 8 is coupled to a rotation-side main shaft 11a (first main shaft). The protrusion 16a of the rotor plate 8 is supported on the stator casing 9 by a third bearing 15a.

  A service person who performs periodic inspection of the wind turbine generator 10 installs the sensor 50 in the vicinity of the inner ring of the spindle bearings 13 and 14 through the opening 121 at the center of the spindle 12a and the opening 111 of the spindle 11a. More specifically, a sensor 50 that measures and monitors the state of the main shaft bearings 13 and 14 is installed in a hollow portion of the main shaft 11a and positioned radially inward of the main shaft bearings 13 and 14. The sensor 50 is good to be located on each virtual surface which passes along the axial center of the spindle bearings 13 and 14. An amplifier 51 that amplifies the output signal from the sensor 50 and a communication device 55 that transmits sensor data to the diagnostic device 100 are installed. The sensor 50, the amplifier 51, and the communication device 55 may be installed via a powerful removable magnet.

  The diagnostic device 100 acquires sensor data from the sensor 50 transmitted from the communication device 55 via a communication device (not shown), and is stored in the storage unit 30 (see FIG. 2) in advance with the acquired sensor data. It is diagnosed whether the spindle bearings 13 and 14 are abnormal by comparing with normal data. Specifically, the diagnostic device 100 detects a sign of failure of the spindle bearings 13 and 14. In addition, the diagnostic apparatus 100 is good to arrange | position in the tower 4 or the management ridge on the ground.

  The spindle 12a of the second embodiment may also be used as a bearing housing that holds a spindle bearing. Further, the main shaft 12a does not need to hold the main shaft bearings 13 and 14 as one body, but a bearing housing for the main shaft bearing 13 supported by the support portion 19a and a bearing housing for the main shaft bearing 14 supported by the support portion 19a. You may have.

  According to the diagnosis method of the wind turbine generator 10 of the present embodiment, the sensor 50 for quantitatively measuring and monitoring the state of the spindle bearings 13 and 14 is installed in the vicinity of the bearing, and the initial stage of damage that has not been found so far can be detected. It can be detected with high accuracy. Based on the sensor information from the sensor 50, it is possible to diagnose whether the spindle bearing 13 needs to be replaced before the wind turbine generator 10 is stopped for a long time.

DESCRIPTION OF SYMBOLS 1 Hub 2 Blade 3 Nacelle 4 Tower 5 Generator 6 Stator 7 Generator rotor 8 Rotor plate 9 Stator casing 10 Wind power generator 11, 11a Main shaft (first main shaft)
12, 12a Main shaft (second main shaft)
13 Spindle bearing (main bearing, first spindle bearing)
14 Spindle bearing (main bearing, second spindle bearing)
15, 15a Third spindle bearing 16 Projection 19 Stator casing 20 Processing unit 21 Sensor data acquisition unit 22 Normal data creation unit 23 Abnormality calculation unit (control unit)
24 Abnormality determination unit (diagnosis unit)
30 storage unit 31 normal data (teacher data)
32 Abnormality 33 Judgment result 41 Input unit 42 Display unit 43 Communication unit 50 Sensor (sensor unit)
51 Amplifier 100 Diagnostic Device

Claims (11)

An anomaly prediction system for wind power generation systems,
A sensor unit installed near the spindle bearing;
A storage unit holding teacher data created based on the output of the sensor unit;
A control unit that creates evaluation data based on the output of the sensor unit acquired after the creation time of the teacher data;
An abnormality sign diagnosis system for a wind power generation system, comprising: a diagnosis unit that compares the teacher data with the evaluation data to diagnose an abnormality sign of the spindle bearing. The spindle bearing is at least
A rotatable outer ring part, a hollow inner ring part, and a rotating body part installed between the outer ring part and the inner ring part,
The abnormal sign diagnosis system for a wind power generation system according to claim 1, wherein the sensor unit is installed on an inner wall of the inner ring unit. The sensor unit is
An outer peripheral portion connecting a plurality of first center points in the axial direction of the outer ring outer wall of the main shaft bearing and a virtual plane connecting the second center point of the main shaft bearing in the vertical direction of the first center point The wind power according to claim 2, wherein a center point of the sensor part is arranged, and the detection surface of the sensor part is brought into contact with the inner wall of the inner ring part so as to face the outer ring outer wall side of the spindle bearing. Abnormality sign diagnosis system for power generation systems. The abnormal sign diagnosis system for a wind power generation system according to any one of claims 1 to 3, wherein the sensor unit detects an operation state signal of the main shaft bearing. The main shaft bearing is composed of a plurality of main shaft bearings in the axial direction,
The abnormal sign diagnosis system for a wind power generation system according to claim 3, wherein the sensor units are respectively arranged on the virtual plane. The abnormal sign diagnosis system for a wind power generation system according to any one of claims 1 to 5, wherein the sensor unit is at least one of an acoustic sensor, a vibration sensor, and an AE sensor. . The said diagnostic part memorize | stores the said operation state signal acquired from the said sensor part in the memory | storage part as the said teacher data, when the said wind power generation system is working normally. An anomaly sign diagnosis system for wind power generation systems in Japan. The diagnosis unit calculates an abnormality degree based on the teacher data and the evaluation data, and diagnoses that the spindle bearing is abnormal when the abnormality degree exceeds a predetermined value. An anomaly predictive diagnosis system for wind power generation system as described in 1. The diagnosis unit compares the frequency characteristic obtained by Fourier transforming the teacher data with the frequency characteristic obtained by Fourier transform of the evaluation data, and diagnoses whether or not the spindle bearing is abnormal. The abnormality sign diagnosis system of the wind power generation system of 7. The wind turbine generation system according to claim 7, wherein the diagnosis unit displays a degree of abnormality based on a frequency characteristic obtained by performing a Fourier transform on the teacher data and a frequency characteristic obtained by performing a Fourier transform on the evaluation data on a display unit. Abnormal sign diagnosis system. 8. The diagnosis unit displays an abnormality degree based on a frequency characteristic obtained by performing a Fourier transform on the teacher data and a frequency characteristic obtained by performing a Fourier transform on the evaluation data on the display unit as a variation value with respect to a time axis. An anomaly predictive diagnosis system for wind power generation system as described in 1.

Images