JP2019074060A - State monitoring device of wind turbine generation windmill, state monitoring method and state monitoring system - Google Patents

Abstract

To provide a state monitoring device which can specify an abnormal region, can accurately and efficiently determine a level of damage and a progress status of the damage, and can predict a remaining life of the abnormal region while taking into consideration an influence of a wind condition, a method and a system.SOLUTION: A state monitoring device 10 comprises: a bearing vibration sensor 11; a nacelle vibration sensor 12; a filter processing part; a frequency analysis part; a precise diagnosis part for comparing a theoretical damage frequency resulting from the damage of a rolling bearing, and a frequency component included in spectrum data, and specifies the presence or absence of an abnormality of the rolling bearing and an abnormal region; a damage level diagnosis part for diagnosing a level of the damage of the region on the basis of a diagnosis value per frequency band; a remaining life prediction part for predicting a remaining life of the abnormal region from the abnormal region, the level of the damage of the region and an operation environment of the rolling bearing; and a remaining life correction part for correcting the predicted remaining life on the basis of a statistical value of the vibration of a nacelle 61 which is detected by the nacelle vibration sensor 12.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a state monitoring device, a state monitoring method, and a state monitoring system of a wind turbine for wind power generation, and more specifically, an abnormality of the wind turbine for wind power generation and an operation state causing an abnormality in advance. The present invention relates to a state monitoring device, a state monitoring method, and a state monitoring system of a wind turbine for wind power generation that can prevent serious damage.

  Conventionally, in order to detect damage of a wind turbine for wind power generation at an early stage and prevent serious damage, monitoring of the condition of the wind turbine is performed. In particular, abnormality diagnosis has been performed for bearings that can be relatively easily repaired, such as bearings and gears, and in many cases, vibration data that is easy to measure is statistically processed to monitor and process changes over time. An approach is taken to issue a warning when the value exceeds a preset threshold. In addition, when the processing value exceeds a preset threshold value, analysis by FFT or the like is performed to perform detailed diagnosis for specifying a damaged site (for example, see Patent Documents 1 to 3).

Unexamined-Japanese-Patent No. 2017-32520 International Publication No. 2015/015987 International Publication No. 2015/198793

  However, the vibration targeted at the specific part largely includes the influence of the load fluctuation due to the change of the wind condition, so there is a problem that it is difficult to set the threshold to an appropriate value. The state monitoring device and state monitoring method described in Patent Document 1 are directed to the presence / absence of abnormality of the bearing, identification of the damaged portion, and detection of the degree of damage, but load fluctuation due to change in wind conditions or deterioration of the wind turbine structure And the remaining life could not be calculated.

  In the technology described in Patent Document 2, using dynamics analysis, it is possible to simulate a damage signal of a bearing in vibration analysis in consideration of the entire structure of a wind turbine, and to set a threshold for determination with high accuracy. However, when the disturbance at the time of measurement is excessive, the effect may be limited.

  Further, in the technology described in Patent Document 3, the determination is performed while the nacelle is at rest, so that the determination accuracy is prevented from being lowered due to the vibration (disturbance) accompanying the rotation for the yaw control of the nacelle. However, since the influence by the wind conditions is received even when the nacelle is at rest, the effects of all the disturbances can not be removed, and there is room for improvement. Furthermore, Patent Documents 1 to 3 do not theoretically consider the effects of wind conditions that differ for each wind turbine, and there is a problem that it is not possible to make a repair plan until a change occurs in the monitored statistics. The

  This invention is made in view of the subject mentioned above, The objective grasps | ascertains the driving | running state by an actual wind condition about the windmill for wind power generation for every machine in a wind farm, Based on the information A condition monitoring device, a condition monitoring method, and a condition monitoring of a wind turbine for wind power generation which enables appropriate repair planning by evaluating the difference with the design life and which enables accurate diagnosis by improving the condition monitoring accuracy. To provide a system.

The above object of the present invention is achieved by the following constitution.
(1) A state monitoring device for a wind turbine for wind power generation, having a rotating part that rotates relative to a stationary member, comprising:
A first vibration sensor fixed to the stationary member or the rotating part;
A second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A filter processing unit that divides and extracts the waveform of the signal detected by the first vibration sensor into a plurality of damage filter frequency bands;
A frequency analysis unit that performs frequency analysis on the filtered waveform transferred from the filter processing unit to obtain spectral data;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotational speed signal of the rotating part, and the frequency component included in the spectrum data obtained by the frequency analysis unit And a precision diagnosis unit for comparing with the above and identifying the presence or absence of an abnormality of the rotating part and the site of the abnormality;
A damage level diagnosis unit that diagnoses the degree of damage to the site based on the diagnosis value for each frequency band calculated for each damage filter frequency band;
A remaining life prediction unit which predicts the remaining life of the abnormal part from the abnormal part, the degree of damage to the part, and the operating environment of the rotating part;
A remaining life correction unit that corrects the remaining life based on the statistical value of the vibration of the nacelle detected by the second vibration sensor;
An apparatus for monitoring the condition of a wind turbine for wind power generation, comprising:
(2) The remaining life correction unit obtains the load of the rotating part from the temporal change of the statistical value of the vibration of the nacelle based on the correlation between the vibration of the nacelle and the load of the rotating part set in advance. The state monitoring device for a wind turbine for wind power generation according to (1), wherein the remaining life is corrected by a load of the rotating component after the change with time.
(3) A method of monitoring the state of a wind turbine for wind power generation, which monitors the state of a wind power generation wind turbine having a rotating part that rotates relative to a stationary member, comprising:
A first detection step of detecting a signal from a first vibration sensor fixed to the stationary member or the rotating part;
A second detection step of detecting the vibration of the nacelle from a second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A filtering step of dividing and extracting the waveform of the signal detected by the first vibration sensor into a plurality of damage filter frequency bands;
A frequency analysis process for obtaining spectral data by frequency analysis of the filtered waveform transferred from the filtering process;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotation speed signal of the rotating part, and the frequency component included in the spectrum data obtained in the frequency analysis step And a precision diagnosis step of comparing the presence and absence of an abnormality of the rotating part and identifying the site of the abnormality,
A damage level diagnosis step of diagnosing the degree of damage to the site based on the diagnosis value for each frequency band calculated for each damage filter frequency band;
A remaining life predicting step of predicting the remaining life of the abnormal part from the abnormal part, the degree of damage to the part, and the operating environment of the rotating part;
A remaining life correcting step of correcting the remaining life by the remaining life predicting step on the basis of the statistical value of the vibration of the nacelle detected by the second vibration sensor;
A method of monitoring the condition of a wind turbine for wind power generation, comprising:
(4) In the remaining life correction step, the load of the rotating part is obtained from the temporal change of the statistical value of the vibration of the nacelle based on the correlation between the vibration of the nacelle and the load of the rotating part set in advance. The method according to (3), wherein the remaining life is corrected by the load of the rotating part after the change with time.
(5) A state monitoring device for a wind turbine for wind power generation, having a rotating part that rotates relative to a stationary member, comprising:
A first vibration sensor fixed to the stationary member or the rotating part;
A second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
An arithmetic processing unit that monitors the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
Equipped with
Comparing the statistical value of the vibration of the nacelle detected by the second vibration sensor for each wind direction to determine a wind direction having a low fluctuation of the statistical value or the statistical value;
The processing unit performs monitoring of the state of the rotating component based on a waveform of a signal detected by the first vibration sensor when the nacelle faces the wind direction. Wind turbine condition monitoring device.
(6) A filter processing unit for extracting a predetermined damage filter frequency band from the waveform of the signal detected by the first vibration sensor, further comprising:
The arithmetic processing unit
A frequency analysis unit that performs frequency analysis on the filtered waveform transferred from the filter processing unit to obtain spectral data;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotational speed signal of the rotating part, and the frequency component included in the spectrum data obtained by the frequency analysis unit And a precision diagnosis unit for comparing with the above and identifying the presence or absence of an abnormality of the rotating part and the site of the abnormality;
The wind turbine status monitoring apparatus for wind power generation according to (5), comprising:
(7) A method of monitoring the state of a wind turbine for wind power generation that monitors the state of a wind turbine for wind power generation having a rotating part that rotates relative to a stationary member, comprising:
A first detection step of detecting a signal from a first vibration sensor fixed to the stationary member or the rotating part;
A second detection step of detecting the vibration of the nacelle from a second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A state monitoring step of monitoring the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
Comparing the statistical value of the vibration of the nacelle detected by the second vibration sensor for each wind direction to determine a wind direction having a low fluctuation of the statistical value or the statistical value;
Equipped with
The state monitoring step performs state monitoring of the rotating component based on a waveform of a signal detected by the first vibration sensor when the nacelle faces the wind direction. Wind turbine condition monitoring method.
(8) A filtering process step of extracting a predetermined damage filter frequency band from the waveform of the signal detected by the first vibration sensor,
And further
The condition monitoring process
A frequency analysis process for obtaining spectral data by frequency analysis of the filtered waveform transferred from the filtering process;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotation speed signal of the rotating part, and the frequency component included in the spectrum data obtained in the frequency analysis step And a precision diagnosis step of comparing the presence and absence of an abnormality of the rotating part and identifying the site of the abnormality,
The wind turbine monitoring method according to (7), further comprising:
(9) A stationary member or a first vibration sensor fixed to a rotating part that rotates relative to the stationary member, and a second vibration sensor fixed to a nacelle of a wind turbine for wind power generation Several wind turbines for wind power generation,
An arithmetic processing unit that monitors the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
A plurality of wind turbine status monitoring systems for wind power generation,
A plurality of wind power generations characterized by comparing statistical values of the vibrations of the nacelle of the wind turbines for wind power generation detected by the second vibration sensor, and determining the degree of influence of the wind conditions in the wind turbines for wind power generation Wind turbine condition monitoring system.
(10) a first detection step of detecting a signal from a first vibration sensor fixed to a stationary member or a rotating part that rotates relative to the stationary member;
A second detection step of detecting vibration of the nacelle from a second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A state monitoring step of monitoring the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
A method of monitoring the state of a plurality of wind turbines for wind power generation,
A plurality of wind power generations characterized by comparing statistical values of the vibrations of the nacelle of the wind turbines for wind power generation detected by the second vibration sensor, and determining the degree of influence of the wind conditions in the wind turbines for wind power generation Monitoring method for wind turbines.

  According to the state monitoring device and state monitoring method of wind turbine for wind power generation of the present invention, with respect to rotating parts incorporated in the wind turbine for wind power generation, the presence or absence of abnormality, the determination of the location of abnormality and the degree of damage are accurately determined. It is possible to predict the remaining life of the abnormal part taking into account the influence of wind conditions.

  Further, according to the state monitoring device and state monitoring method of a wind turbine for wind power generation of the present invention, the nacelle monitors the state of rotating parts when the direction of the influence of the wind condition is small, and the wind power with high accuracy. It can monitor the power generation wind turbine.

  Further, according to the state monitoring system and the state monitoring method of wind turbine for wind power generation of the present invention, it is possible to monitor the state of a plurality of wind turbines for wind power generation while considering the degree of influence of the wind conditions.

It is a perspective view of a wind farm in which a plurality of wind turbines for wind power generation concerning an embodiment of the present invention were installed. BRIEF DESCRIPTION OF THE DRAWINGS It is a block diagram which shows schematic structure of the windmill for wind power generation concerning embodiment of this invention, and its state monitoring apparatus. It is a block diagram which shows the function structure of the arithmetic processing unit shown in FIG. It is a flowchart for demonstrating the operation | movement procedure of the state monitoring apparatus which concerns on embodiment of this invention. It is a table | surface which shows the relationship between the site | part of the damage | wound of the rolling bearing concerning embodiment of this invention, and the vibration frequency which arises due to a damage | wound.

First Embodiment
Hereinafter, a state monitoring device, a state monitoring method, and a state monitoring system of a wind turbine for wind power generation according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 5.

  As shown in FIG. 1, the wind farm F is configured of a plurality of wind power generation wind turbines 50 (50A to 50D) that generate electric power using wind power. The wind turbines 50A to 50D for wind power generation have approximately the same height, and are installed in the wind farm F at intervals of, for example, several hundred meters. In addition, each of the wind turbines 50A to 50D for wind power generation is operated under disturbance due to a local wind condition that changes according to the topography of the surrounding area and the like. The disturbance is generally referred to as wind shear or turbulent flow, and is a factor that reduces the power generation efficiency of the wind turbine 50 for wind power generation. For this reason, in each of the wind turbines for wind power generation 50A to 50D, the direction of the nacelle 61 is controlled so that the blades 52 described later face the wind direction so as to reduce the influence of the wind conditions, thereby generating electric power. Here, wind shear is the distribution of the wind speed in the vertical direction of the wind turbine.

  As shown in FIG. 2, each wind turbine 50 for wind power generation mainly includes a main shaft 51, blades 52, a speed increasing gear 53, a generator 54, and a main bearing 55, and each member excluding the blades 52 has It is stored in a nacelle 61 supported by a tower 60. The blade 52 and the nacelle 61 are rotatably supported on the tower 60, and the yaw angle is controlled according to the wind direction.

  The blade 52 provided at the tip of the main shaft 51 converts wind power into rotational torque and transmits it to the main shaft 51. The main shaft 51 is rotatably supported by the main bearing 55, and transmits the rotational torque generated by the blade 52 receiving the wind force to the input shaft of the step-up gear 53. The speed increasing gear 53 accelerates the rotational speed of the main shaft 51 and outputs it to the generator 54. The speed increasing gear 53 is constituted by, for example, a gear speed increasing mechanism having a plurality of shafts rotatably supported by bearings.

  The generator 54 is connected to the output shaft of the speed increaser 53, and generates electric power by the rotational torque received from the speed increaser 53. The generator 54 is, for example, an induction generator, and a rotor (not shown) is rotatably supported by a bearing.

  The bearings disposed in the main bearing 55, the speed increasing gear 53, and the generator 54 are, for example, rolling bearings such as self-aligning roller bearings, tapered roller bearings, cylindrical roller bearings, and ball bearings. A plurality of rolling bearings (not shown) are disposed so as to be able to roll between an inner ring externally fitted to the rotary shaft of the wind turbine for wind power generation 50, an outer ring internally fitted to a housing or the like, and an inner ring and an outer ring. And a cage for rotatably holding the rolling element.

  As shown in FIG. 2, the state monitoring device 10 for a wind turbine for wind power generation according to this embodiment diagnoses an abnormality of a rolling bearing (not shown) which is a rotating part incorporated in the wind turbine 50 for wind power generation. A vibration sensor 11 for bearings as a first vibration sensor for detecting a vibration (signal) generated from a bearing, and for the nacelle as a second vibration sensor installed in the nacelle 61 to detect a vibration (signal) of the nacelle 61 Vibration sensor 12, wind condition measurement unit 13, rotation sensor (not shown) for detecting the rotational speed of the rolling bearing, wind condition determination unit 14, bearing vibration sensor 11, nacelle vibration sensor 12, wind condition The signal detected by the measurement unit 13 and the rotation sensor is received through the data transmission means (transmission means) 15 and signal processing is performed to diagnose the presence or absence of abnormality of the rolling bearing, to identify the location of the abnormality, and to damage degree And a controller 20 comprising an arithmetic processing unit 21 (arithmetic processing unit) that diagnoses the progress of damage and predicts the remaining life of the abnormal part in real time, and a controller 22 that drives and controls the wind turbine 50 for wind power generation And an output device 30 which receives a signal from the controller 20 through the data transmission means 31 and which comprises a monitor, an alarm and the like.

  An acceleration sensor is generally used as the vibration sensor 11 for bearings and the vibration sensor 12 for nacelle, but an AE (acoustic emission) sensor, an ultrasonic sensor, a shock pulse sensor, or a velocity, strain, stress, displacement sensor There is no particular limitation as long as vibration can be converted into an electrical signal. When mounted on a noisy machine, it is preferable to use the insulation type as it is less susceptible to noise.

The bearing vibration sensor 11 is fixed to a housing load area of an outer ring which is a fixed ring of the rolling bearing. Methods of fixing the bearing vibration sensor 11 include bolt fixation, adhesion, combined use of bolt fixation and adhesion, and embedding with a resin material. When the bearing vibration sensor 11 is fixed to the inner ring which is a rolling ring of a rolling bearing, it may be fixed to the side surface or may be fixed to the inner diameter surface using a slip ring.
In the case of bolt fixing, a detent function may be provided. In addition, by molding the bearing vibration sensor 11 with a resin material, it is possible to prevent the entry of moisture, and the vibration isolation against external vibration is improved, so the reliability of the sensor itself is dramatically improved. It can be improved.

  The nacelle vibration sensor 12 is installed on the outer shell of the nacelle 61 to detect the nacelle 61 vibration. The nacelle vibration sensor 12 may be installed on the inner wall surface of the outer shell of the nacelle 61 or the beam or column inside.

  The wind condition measurement unit 13 measures wind conditions such as the wind direction and the wind speed, and a wind direction and anemometer provided in the nacelle 61 can be used. In addition, the wind condition measurement unit 13 can also use a meteorological radar (Doppler lidar) that can measure the moving speed of the measurement object as a wind condition measurement device, and install it on the ground near the wind turbines. May be measured.

  The wind condition determination unit 14 calculates, for example, wind condition values such as wind speed direction appearance frequency distribution, disturbance degree distribution, wind power, blowing angle, wind shear, etc., based on the wind conditions obtained in real time by the wind condition measurement unit 13 .

The control device 22 drives and controls the wind turbine 50 for wind power generation so that the diagnosis result obtained by the arithmetic processor 21 is fed back to the operating condition (for example, the number of revolutions is decreased). Further, the controller 20 is configured by a microcomputer (IC chip, CPU, MPU, DSP, etc.). Therefore, since each process to be described later can be executed by the program of the microcomputer, the apparatus can be simplified, miniaturized and configured inexpensively.
Further, in the case of the present embodiment, one controller 20 including the arithmetic processor 21 and the controller 22 may be provided for each of the plurality of wind turbines 50A to 50D for wind power generation, or a plurality of wind turbines for wind power generation One unit may be provided for 50A to 50D. Alternatively, one controller 20 may be provided for each of the plurality of wind turbines 50A to 50D for wind power generation, and may further include another controller 20 that transmits and receives data of each controller 20.

FIG. 3 is a block diagram showing the main functional configuration of the arithmetic processing unit 21 according to the present embodiment.
As shown in FIG. 3, the arithmetic processing unit 21 includes a data collection / distribution unit 211, a rotation analysis unit 212, a filter processing unit 213, a frequency analysis unit 214, a comparison determination unit 215, and an internal memory 216. .
As described above, the arithmetic processing unit 21 is configured by the microcomputer as described above, that is, each processing unit such as the data collection and distribution unit 211 is executed by executing the program recorded and held in the microcomputer. Will execute each process as follows.

The data collection / distribution unit 211 converts the signal sent from the bearing vibration sensor 11 into a digital signal by the A / D converter, and at the same time collects and temporarily accumulates the signal related to the rotational speed. In accordance with this, it is distributed to either the rotation analysis unit 212 or the filter processing unit 213.
The A / D converter may be integrated with the bearing vibration sensor 11, and the digital signal may be received through the data transmission unit 15 described above.

Rotation analysis unit 212 calculates the rotation speed of the inner ring based on the signal output from the rotation sensor, and outputs the calculated rotation speed to comparison and determination unit 215.
When the rotational speed detection means is configured by an encoder attached to the inner ring and a magnet or magnetic detection element attached to the outer ring, the output signal is a pulse signal according to the shape and rotational speed of the encoder Become. Therefore, the rotation analysis unit 212 has a predetermined conversion function or conversion table corresponding to the shape of the encoder, and calculates the rotation speed of the inner ring from the pulse signal.

The filter processing unit 213 has a band pass filter function, divides and extracts the output signal of the bearing vibration sensor 11 into a plurality of damage filter frequency bands, and removes other unnecessary frequency bands. The damage filter frequency band is set according to the natural frequency band in each bearing device. The natural frequency can be easily obtained by vibrating the object to be measured by an impact method using an impulse hammer or the like, and frequency analyzing the sound generated by the vibration detector attached to the object or the impact. it can.
When the object to be measured is a rolling bearing, a natural frequency caused by any of the inner ring, the outer ring, the rolling element, the housing, etc. is given. In general, there are a plurality of natural frequencies of machine parts, and the amplitude level at the natural frequency is high, so the sensitivity of measurement is good.

  Further, the filter processing unit 213 of the present embodiment may be any one as long as it extracts a plurality of damage filter frequency bands to be frequency analyzed from the waveform of the signal detected by the vibration sensor. What is necessary is just to be comprised by one side.

The frequency analysis unit 214 analyzes the frequency of the vibration signal generated from the rolling bearing based on the output signal (measured data) from the bearing vibration sensor 11. The frequency analysis unit 214 is an FFT operation unit that calculates the frequency spectrum of the vibration signal, and calculates the frequency spectrum of the vibration signal based on the FFT algorithm and the envelope analysis. The calculated frequency spectrum is output to the comparison / determination unit 215 as spectrum data.
The frequency analysis unit 214 may perform absolute value conversion processing or envelope processing as pre-processing to perform FFT, and may convert only frequency components necessary for diagnosis of abnormality. In addition, the spectrum data after envelope processing (envelope frequency spectrum) is also output to the comparison / determination unit 215 as necessary.

  The comparison determination unit 215 periodically monitors the state of the rolling bearing based on the measured vibration and rotational speed of the rolling bearing. Specifically, the comparison / determination unit 215 configures a simple diagnosis unit, a precision diagnosis unit, a damage level diagnosis unit, a remaining life prediction unit, and a remaining life correction unit shown below.

  The simplified diagnosis unit calculates at least one simplified diagnostic value of an effective value, a peak value, and a crest factor obtained from the waveform of the signal detected by the bearing vibration sensor 11 and compares the calculated value with each threshold value. And when it is "effective value, peak value, crest factor> each threshold value", it diagnoses simply as abnormality existence of a rolling bearing.

When the simple diagnosis unit diagnoses that there is an abnormality, the precision diagnosis unit pre-calculates the theoretical damage frequency due to the damage of each part of the rolling bearing using a predetermined relational expression shown in FIG. Based on the spectral data output from the analysis unit 214, the occurrence of an abnormality such as a flaw in a bearing and the location thereof are identified by checking each theoretical damage frequency (success or failure of "peak frequency = theoretical damage frequency").
The theoretical damage frequency may be calculated using past data stored in the internal memory 216 when the same diagnosis is performed before.

  The damage level diagnostic unit determines the degree of damage to the site based on the diagnosis value for each frequency band calculated for each damage filter frequency band when or after identifying the damage site (degree of damage: initial / progressive / end stage) To diagnose In addition, as a diagnostic method of the extent of damage, the following two are mentioned, for example.

(1) The diagnosis value for each frequency band is used as the vibration effective value, and the damage level diagnosis unit compares the vibration effective value calculated for each damage filter frequency band with the normal product value or the normal value. Then, the value of the ratio of the vibration effective value to the normal product or the normal value is determined for each damage filter frequency band to diagnose the degree of damage of the rotating parts or the progress of damage.
(2) Assuming that the diagnostic value for each frequency band is the envelope vibration effective value calculated from the spectrum data obtained by the frequency analysis unit 214, the damage level diagnostic unit calculates the envelope vibration effective value calculated for each damage filter frequency band The degree of damage to the rotating parts or the progress of damage is diagnosed by comparison with the normal product or the normal value.

Next, the remaining life prediction unit predicts the remaining life of the abnormal part from the abnormal part, the degree of damage to the part, and the operating environment of the rotating part. Specifically, the remaining life of the rotating part is determined using an expansion prediction analysis of the damaged portion based on the determination result of the degree of damage.
It is possible to predict by the fracture mechanics method by regarding "expansion of damage" as pseudo "damage of crack". In the fatigue life of metal materials, small cracks are generated at the very early stage of fatigue, and the process of small cracks growing by repeated stress is mostly occupied, and this phenomenon can be used as a model for bearing damage Make predictions. The crack growth behavior in fatigue phenomena is governed by the linear fracture mechanics parameter K (stress intensity factor) represented by the crack length and the stress acting on the crack, and in the region where the crack stably propagates, the variation range of K It is known as a Paris law that ΔK (stress intensity factor range) and crack growth rate da / dN (crack growth amount per stress cycle) have a double logarithmic linear relationship (Equation (1)).

  The relationship of equation (1) is important data representing the fatigue crack growth characteristics and is the basis of the fatigue life design of the structural material. The development of rolling fatigue cracks found in rolling bearings is characterized by mode II, and for example, the data described in the following non-patent documents (1) to (3) are effective. Further, it has been clarified by Okazaki et al. That the stress intensity factor of mode II is expressed as the following equation (2) (non-patent document (4) below).

Non-patent literature (1): material, vol. 51, no. 8 (2002), pp. 918-925, "Measurement of Mode II fatigue crack growth lower limit value ΔK _IIth "
Non-patent literature (2): material, vol. 50, no. 10 (2001), pp. 1108-1113, "A new test method for determining mode II fatigue crack growth characteristics of high hardness materials"
Non-patent literature (3): material, vol. 54, no. 12 (2005), pp. 1295-1300, "Experimental evaluation of mode II fatigue crack growth characteristics and crack driving force"
Nonpatent literature (4): Theoretical and Applied Fracture Mechanics, vol. 73 (2014), pp. 161-169

  In equation (2), τ is a shear stress acting on the crack, and F is a coefficient determined by the shape and Poisson's ratio of the material. √ (area) is the square root of the area of the elliptical crack that depends on the infinite field, that is, it can be regarded as a parameter representing the area of the defect determined by the above-mentioned damage determination. In rolling bearings, damage defects develop in a direction parallel to the raceway (the circumferential direction of the race) due to shear stress generated by contact pressure. At this time, since the axial size of the damage defect can be regarded as a constant determined by the length of the rolling element in contact with the raceway surface, the defect area parameter √ (area) is substantially the circumferential length It can be regarded as a function of Accordingly, the defect area parameter √ (area) can be expressed as the following equation (3) using the circumferential defect length 2a.

In equation (3), C 'is a coefficient determined by the dimensional specifications of the bearing.
The shear stress τ is a far shear stress of a defect acting on a plane parallel to the rolling surface (that is, also parallel to the defect surface). The shear stress can be determined as a stress proportional to the maximum contact surface pressure P _max by the exact solution of Lundberg and Hanson et al., And the maximum value is used here.

In the case of rolling contact fatigue, shear stress occurs in both directions as the rolling elements pass. Assuming that the amplitude is a nominal shear stress amplitude τ _a , the shear stress range is Δτ = 2τ _a , so the mode II stress intensity factor range ΔK _II is expressed by the following equation (4) using equation (3) Ru.

  Using equation (4), crack growth rate equation (1) can be expressed as a function of defect length a as in the following equation (5).

In order to obtain the fatigue life N _f from the rolling fatigue crack growth rate equation, the reciprocal of the crack growth rate (dN / da) ranges from the initial crack length a ₀ to the final crack length a _f at fatigue failure , Obtained by integration at the crack length a. That is, it is calculated | required like the following formula (6).

ここで、初期き裂長さａ_０は、損傷診断によって得られた損傷部の大きさ（サイズ）と考えることができ、最終き裂長さａ_ｆは、運転上許容できる最大のサイズ（許容限界サイズ）を意味する。運転上許容できる最大のサイズとは、例えば音響性能の限界値に達する損傷部サイズ、もしくは軸受の回転トルクの許容限界値に達する損傷部サイズなどである。いずれの場合も対象にしても、任意の損傷部サイズをａ_ｆとすることができる。

Here, the initial crack length a ₀ can be considered to be the size (size) of the damaged portion obtained by the damage diagnosis, and the final crack length a _f is the maximum size allowable for operation (allowable limit size Means). The maximum operation allowable size is, for example, the damage size reaching the limit value of the acoustic performance or the damage size reaching the allowable limit value of the rotational torque of the bearing. In any case, any lesion size can be a _f .

As described above, based on the size of the damaged part obtained from the damage diagnostic data, it is possible to determine the cycle (that is, the remaining life) from the time of damage diagnosis to the development of the damaged part to the allowable limit size. In addition, by using the rotation speed n (min ^-1 ) of the corresponding bearing and the stress repetition number e (cycle / rev) per shaft rotation, the remaining life (day) can be easily obtained as in the following formula (7) Can be calculated.

  Under wind shear or turbulent flow, the load (shear stress τ) acting on the rolling bearing differs depending on the balance of wind received by the blades 52 of the wind turbine for wind power generation 50, and the support structure of the equipment in the wind turbine deviates from the design condition Load, which greatly affects the prediction of the remaining life of the rolling bearing.

  As described above, the yaw angle of the wind turbine for wind power generation 50 is controlled so as to be directed to a direction in which the influence of the wind condition is small, but it is difficult to completely eliminate the influence of the wind condition. Therefore, depending on the wind conditions, correction may be necessary to predict the remaining life. In addition, deterioration of the wind turbine structure may also affect the remaining life, and a correction is necessary to predict the remaining life.

  In the remaining life correction unit, the change over time in the statistical value of the vibration of the nacelle 61 detected by the nacelle vibration sensor 12 is obtained, and the change over time in the statistical value of the vibration is regarded as a change in wind conditions or deterioration of the windmill structure. Make life corrections. That is, when a rise is seen in the temporal change of the statistical value of the vibration of the nacelle 61, deterioration of the wind turbine structure is concerned.

  For remaining life correction, the correlation between the vibration of the nacelle 61 and the load (shear stress τ) of the rolling bearing is obtained in advance by numerical analysis means such as structural analysis, and the vibration of the nacelle 61 detected by the nacelle vibration sensor 12 Determine the load of the rolling bearing. Then, the remaining life is recalculated and corrected according to the above-described remaining life prediction procedure. This makes it possible to predict the remaining life with high accuracy in consideration of the influence of the wind conditions.

  The diagnosis result of the rolling bearing determined in this manner is stored in the internal memory 216, and is output to the control device 22 that controls the operation of the wind turbine 50, and feeds back a control signal according to the diagnosis result. When it is diagnosed that an abnormality has occurred in the bearing, the control device 22 feedbacks to the operating conditions of the application in which the bearing is incorporated (load load is reduced, rotational speed is reduced, etc.) to achieve the desired life. Make adjustments automatically or manually. Furthermore, the diagnosis result is sent to the output device 30 by the data transmission means 31 using wire communication or wireless considering a network. The output device 30 may be a remote location.

  The internal memory 216 is composed of, for example, a memory or an HDD, etc., and design specification data of each rotating part used for calculation of the theoretical damage frequency, and diagnosis of presence / absence of abnormality of the rolling bearing determined by the comparison determination unit 215 Store data on site identification and residual life of abnormal site.

The output device 30 displays the diagnosis result of the rolling bearing on a monitor or the like in real time. In addition, when it is diagnosed that an abnormality has occurred in the bearing, a warning is given along with a predicted value of the remaining life (including the corrected remaining life). The warning and the method of notifying the remaining life prediction value are displayed directly on the indicator light and display monitor provided in the bearing unit or the application in which they are incorporated, and an alarm for the user using a warning device such as a light or buzzer. You may be reminded of a certain thing.
Further, since the data transmission means 31 of the signal may be able to transmit and receive the signal properly, it may be wired or may use wireless considering a network.

  Next, the operation of the state monitoring device 10 configured as described above will be described. FIG. 4 is a flowchart for explaining the operation procedure of the state monitoring device 10.

  First, in step S1, the rotational speed of the rolling bearing is detected by the vibration and rotation sensor generated from the rolling bearing by the vibration sensor 11 for bearings, and the detected vibration signal and rotational speed signal are transmitted through the data transmission means 15. It is input to the data collection / distribution unit 211 of the arithmetic processing unit 21.

  The data collection / distribution unit 211 amplifies the input analog vibration signal as necessary, and converts it into a digital signal by an A / D converter.

  Next, in step S2, based on the data stored in the data collection / distribution unit 211 and the internal memory 216, diagnostic parameters to be used for the subsequent diagnosis are calculated. Specifically, as diagnostic parameters, at least one simple diagnostic value threshold for theoretical value (RMS), peak value (PEAK) and crest factor (CF) used for simple diagnosis, theoretical damage frequency, damage filter frequency band are calculated Be done.

  Next, in step S3, a simple diagnosis is performed to compare the effective value (RMS), peak value (PEAK), and crest factor (CF) obtained from the waveform of the signal detected by the bearing vibration sensor 11 with each threshold. If "effective value (RMS), peak value (PEAK), crest factor (CF)> each threshold value", it is determined that there is an abnormality in the rolling bearing and the process proceeds to step S4, and if each value is less than the threshold It returns to step S1 as nothing, and performs step S1 at the next timing.

  In step S4, the arithmetic processing unit 21 divides and extracts the waveform of the signal detected by the bearing vibration sensor 11 for each damage filter frequency band calculated in step S2.

  Further, in step S5, the arithmetic processing unit 21 performs envelope processing and frequency analysis on the waveform after filter processing transferred from the filter processing unit 213 to obtain spectral data.

Then, in step S6, in the damage filter frequency band extracted by the comparison / determination unit 215, the theoretical damage frequency due to the damage of the rolling bearing calculated based on the rotation speed signal of the rolling bearing and the calculation processing unit The frequency component contained in the acquired spectral data is compared, and in step S7, the portion of the rolling bearing abnormality is identified.
That is, the theoretical damage frequency component of the rolling bearing includes the bearing damage component Sx, that is, the inner ring damage component Si, the outer ring damage component So, the rolling element damage component Sb and the cage component Sc, and the level of each frequency component is extracted It will be done. Then, it is specified whether the abnormal part is any of the inner ring, the outer ring, the rolling element, and the cage.

  Next, in step S8, the degree of damage to the site is diagnosed by any method using the diagnostic value classified by frequency band described above, which is calculated for each damage filter frequency band.

  Furthermore, in step S9, the remaining life of the abnormal part is predicted from the abnormal part, the degree of damage to the part, and the operating environment of the rotating part.

  In step S10, the load of the rolling bearing is determined from the temporal change of the statistical value of the vibration of the nacelle 61 detected by the nacelle vibration sensor 12 according to the change of the wind condition, and the remaining life is corrected. If the change with time of the statistical value is small, the remaining life correction can be omitted.

  Through such a procedure, it is possible to diagnose the presence or absence of an abnormality in the rolling bearing which is a rotating part, to identify the site of the abnormality, to diagnose the degree of damage to the site, and to predict the remaining life of the site of the abnormality.

  The result of the diagnosis performed in this manner is output to the control device 22 that controls the operation of the wind turbine for wind power generation 50, and in step S11, a control signal according to the diagnosis result is fed back. Further, the data is transmitted to the output device 30 by the data transmission means 31 using a wired connection or a wireless in consideration of a network.

  As described above, according to the state monitoring device 10 and the state monitoring method according to the present embodiment, the vibration sensor 11 for bearings fixed to the rolling bearing and the nacelle 61 of the wind power generation wind turbine 50 are fixed. Filter processing unit 213 for dividing and extracting the waveform of the signal detected by the vibration sensor 12 for bearings, the vibration sensor for bearings 11 into a plurality of damage filter frequency bands, and filter processing transferred from the filter processing unit 213 The frequency analysis unit 214 analyzes the frequency of the subsequent waveform and obtains spectral data, and the theoretical damage frequency attributable to the damage of the rolling bearing calculated based on the rotation speed signal of the rolling bearing in the extracted damage filter frequency band, The frequency component contained in the spectrum data obtained by the frequency analysis unit 214 is compared, and the presence or absence of abnormality of the rolling bearing and Damage level diagnostic unit for diagnosing the degree of damage to a site based on a precision diagnostic unit for identifying a regular site, and a diagnostic value for each frequency band calculated for each damage filter frequency band; A remaining life prediction unit that predicts the remaining life of the abnormal part from the degree of damage and the operating environment of the rolling bearing, and the remaining life predicted based on the statistical value of the vibration of the nacelle 61 detected by the nacelle vibration sensor 12 And a remaining life correction unit that corrects As a result, with respect to the rolling bearing incorporated in the wind turbine 50 for wind power generation, it is possible to accurately determine the presence or absence of an abnormality, the identification of an abnormal part, and the degree of damage, and the abnormal part taking into account the influence of wind conditions. It is possible to predict the remaining life of

  Further, the remaining life correction unit obtains the load of the rolling bearing from the temporal change of the statistical value of the vibration of the nacelle 61 based on the correlation between the vibration of the nacelle 61 and the load of the rolling bearing set in advance. Since the remaining life is corrected by the load of the rolling bearing, the remaining life of the abnormal part can be corrected in consideration of the influence of the wind condition.

Second Embodiment
Next, a state monitoring device and a state monitoring method according to a second embodiment of the present invention will be described with reference to the state monitoring device 10 of the first embodiment.
In the second embodiment, the vibration of the nacelle 61 is detected by the nacelle vibration sensor 12 installed in each of the wind turbines 50 for wind power generation, and the waveform of the detected signal is at least one of effective value, peak value and crest factor. Statistical values are calculated and compared for each wind direction. When the direction of the nacelle 61 is controlled so that the blade 52 faces the wind direction with the wind direction having a low value of the statistical value or the fluctuation of the statistical value as the wind direction with little influence of the wind condition, the bearing vibration sensor 11 is used. The state of the rolling bearing is monitored based on the detected signal. Alternatively, the nacelle 61 validates the state monitoring result when the blade 52 is directly facing the wind direction in which the statistical value or the fluctuation of the statistical value is low. As a result, it is possible to perform highly accurate state monitoring with reduced influence of wind conditions.

As described above, according to the state monitoring device 10 and the state monitoring method according to the present embodiment, the vibration sensor 11 for bearings fixed to the rolling bearing and the vibration for nacelle fixed to the nacelle 61 of the wind turbine 50 for wind power generation The statistical value of the vibration of the nacelle 61 detected by the nacelle vibration sensor 12 includes the sensor 12 and the arithmetic processing unit 21 that monitors the state of the rolling bearing based on the waveform of the signal detected by the bearing vibration sensor 11. Since the calculation processing unit 21 monitors the state of the rolling bearing when the nacelle 61 faces the wind direction by determining the wind direction in which the statistical value or the fluctuation of the statistical value is low and comparing the Can monitor the wind turbine 50 for wind power generation.
The other configurations and actions are similar to those of the first embodiment.

  In the state monitoring step of monitoring the state of the rolling bearing based on the waveform of the signal detected by the bearing vibration sensor 11 in the present embodiment, the degree of damage or the abnormal portion as in the first embodiment The present invention is not limited to the identification of the remaining life, but may be the presence or absence of an abnormality by a simple diagnosis or the identification of a site by a precise diagnosis.

  For example, if the identification is performed by precise diagnosis, the filter processing unit 213 extracts a predetermined damage filter frequency band from the waveform of the signal detected by the bearing vibration sensor 11, and the frequency analysis unit 214 The filtered waveform transferred from the processing unit 213 is frequency analyzed to obtain spectral data. Then, in the damage filter frequency band extracted by the precision diagnosis unit, the theoretical damage frequency due to the damage of the rolling bearing calculated based on the rotation speed signal of the rolling bearing and the spectral data obtained by the frequency analysis unit are included. The frequency components may be compared with each other to identify the presence or absence of an abnormality in the rolling bearing and the location of the abnormality.

Third Embodiment
Next, a state monitoring system and a state monitoring method according to a third embodiment of the present invention will be described with reference to the state monitoring device 10 of the first embodiment.
In the third embodiment, as shown in FIG. 1, in the wind farm F in which a plurality of wind turbines for wind power generation 50A to 50D are installed, statistical values of vibrations of each nacelle 61 (effective value, peak value, crest factor, etc.) Is detected by the nacelle vibration sensor 12. Then, by comparing the statistical values of the vibrations of the respective nacelle 61 among the wind turbines 50A to 50D for wind power generation, the degree of influence of the wind conditions is obtained for each of the wind turbines 50A to 50D for wind power generation.

  That is, the statistical value of the vibration of the nacelle 61 detected by the nacelle vibration sensor 12 is compared with the numerical value of the other wind turbines 50 for wind power generation in the same wind farm F, and the numerical value, the temporal change, etc. Then, the wind power generation wind turbines 50A to 50D in the same wind farm F are ranked regarding the degree of influence of the wind conditions. Then, when an abnormality is found in one of the wind turbines 50 for wind power generation in the wind farm F, a repair plan is created based on this ranking. For example, for a wind turbine that is more susceptible to wind conditions than the wind turbine for wind power generation 50 in which an abnormality has been found, damage may be made by planning the same repair equipment as the abnormality location of the wind turbine in which the abnormality is generated. The repair plan of the entire wind farm F including the undetected wind power generation wind turbine 50 is optimized.

  As described above, according to the state monitoring system and the state monitoring method according to the present embodiment, the plurality of wind turbines 50A to 50D for wind power generation are the vibration sensor 11 for bearings fixed to the rolling bearings, and the wind turbine 50 for wind power generation The vibration sensor 12 for nacelle fixed to the nacelle 61 of each of the above, and the arithmetic processing unit 21 for monitoring the state of the rolling bearing based on the waveform of the signal detected by the vibration sensor 11 for bearing Since the statistical value of the vibration of the nacelle 61 of each of the wind turbines 50A to 50B for wind power generation detected by the sensor 12 is compared to determine the degree of influence of the wind condition on each of the wind turbines 50A to 50B, a plurality of wind farms F The wind turbines 50A to 50B for wind power generation can be efficiently monitored with high accuracy and managed.

In the state monitoring system of the present embodiment, the arithmetic processing unit 21 is shared by the plurality of wind turbines 50 for wind power generation, and each nacelle vibration sensor 12 of each of the wind turbines 50A to 50D for wind power generation detects The statistical value of the vibration of each nacelle 61 is acquired through the data transmission means 15, and each statistical value is compared to determine the degree of influence of the wind conditions in the wind turbines 50A to 50D for wind power generation.
Therefore, the state of each rolling bearing is monitored based on the waveform of the signal detected by the bearing vibration sensor 11 of each of the wind turbines 50A to 50D by one arithmetic processing unit 21 provided outside. It is also good.
Other configurations and actions are similar to those of the first and second embodiments.

  The state monitoring device, the state monitoring method, and the state monitoring system of the present invention are not limited to the embodiments described above, and appropriate modifications, improvements, and the like can be made.

  For example, the method of predicting the remaining life of the present invention is not limited to that of the present embodiment. In that case, the degree of damage to the abnormal part, which is a parameter for determining the remaining life of the abnormal part, will be described. When judging the rolling fatigue life of a rolling bearing, the peeling life starting from the surface or its vicinity is often used. Taking this as an example, in the abnormal initial stage of the rolling bearing, wear occurs on the surface, and local stress concentration occurs due to the formation of indentations due to foreign substances and changes in the metallographic structure. When the stress concentration portion is repeatedly stressed, it leads to peeling, and the abnormality of the rolling bearing progresses.

  In addition, since the mounting location of the rolling bearing varies depending on the application, the operating environment may vary depending on whether the lubrication condition is good or poor. The poor lubrication state includes, for example, a thin lubrication state, a lubrication state in which foreign matter is mixed, a lubrication state in which water is mixed, and a high temperature lubrication state, and is not particularly limited in the present invention.

10 State monitoring device for wind turbine for wind power generation 11 Vibration sensor for bearings (first vibration sensor)
12 Vibration sensor for nacelle (second vibration sensor)
21 Arithmetic processor (arithmetic processor)
50, 50A-50D Wind turbine for wind power generation 61 Nacelle 213 Filter processing unit F Wind farm Sb Rolling element flaw component (bearing damage frequency)
Sc cage component (bearing damage frequency)
Si inner ring flaw component (bearing damage frequency)
So outer ring flaw component (bearing damage frequency)

Claims (10)

A state monitoring device for a wind turbine for wind power generation, having a rotating part that rotates relative to a stationary member, comprising:
A first vibration sensor fixed to the stationary member or the rotating part;
A second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A filter processing unit that divides and extracts the waveform of the signal detected by the first vibration sensor into a plurality of damage filter frequency bands;
A frequency analysis unit that performs frequency analysis on the filtered waveform transferred from the filter processing unit to obtain spectral data;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotational speed signal of the rotating part, and the frequency component included in the spectrum data obtained by the frequency analysis unit And a precision diagnosis unit for comparing with the above and identifying the presence or absence of an abnormality of the rotating part and the site of the abnormality;
A damage level diagnosis unit that diagnoses the degree of damage to the site based on the diagnosis value for each frequency band calculated for each damage filter frequency band;
A remaining life prediction unit which predicts the remaining life of the abnormal part from the abnormal part, the degree of damage to the part, and the operating environment of the rotating part;
A remaining life correction unit that corrects the remaining life based on the statistical value of the vibration of the nacelle detected by the second vibration sensor;
An apparatus for monitoring the condition of a wind turbine for wind power generation, comprising:   The remaining life correction unit obtains the load of the rotating component from the temporal change of the statistical value of the vibration of the nacelle based on the correlation between the vibration of the nacelle and the load of the rotating component set in advance, and The state monitoring device for a wind turbine for wind power generation according to claim 1, wherein the remaining life is corrected by the load of the rotating part after change. A method of monitoring the state of a wind turbine for wind power generation, which monitors the state of a wind turbine for wind power generation having a rotating part that rotates relative to a stationary member, comprising:
A first detection step of detecting a signal from a first vibration sensor fixed to the stationary member or the rotating part;
A second detection step of detecting the vibration of the nacelle from a second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A filtering step of dividing and extracting the waveform of the signal detected by the first vibration sensor into a plurality of damage filter frequency bands;
A frequency analysis process for obtaining spectral data by frequency analysis of the filtered waveform transferred from the filtering process;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotation speed signal of the rotating part, and the frequency component included in the spectrum data obtained in the frequency analysis step And a precision diagnosis step of comparing the presence and absence of an abnormality of the rotating part and identifying the site of the abnormality,
A damage level diagnosis step of diagnosing the degree of damage to the site based on the diagnosis value for each frequency band calculated for each damage filter frequency band;
A remaining life predicting step of predicting the remaining life of the abnormal part from the abnormal part, the degree of damage to the part, and the operating environment of the rotating part;
A remaining life correcting step of correcting the remaining life by the remaining life predicting step on the basis of the statistical value of the vibration of the nacelle detected by the second vibration sensor;
A method of monitoring the condition of a wind turbine for wind power generation, comprising:   In the remaining life correction step, the load of the rotating component is obtained from the temporal change of the statistical value of the vibration of the nacelle based on the correlation between the vibration of the nacelle and the load of the rotating component set in advance, The method for monitoring the state of a wind turbine for wind power generation according to claim 3, wherein the remaining life is corrected by the load of the rotating part after change. A state monitoring device for a wind turbine for wind power generation, having a rotating part that rotates relative to a stationary member, comprising:
A first vibration sensor fixed to the stationary member or the rotating part;
A second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
An arithmetic processing unit that monitors the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
Equipped with
Comparing the statistical value of the vibration of the nacelle detected by the second vibration sensor for each wind direction to determine a wind direction having a low fluctuation of the statistical value or the statistical value;
The processing unit performs monitoring of the state of the rotating component based on a waveform of a signal detected by the first vibration sensor when the nacelle faces the wind direction. Wind turbine condition monitoring device. It further comprises a filter processing unit for extracting a predetermined damage filter frequency band from the waveform of the signal detected by the first vibration sensor,
The arithmetic processing unit
A frequency analysis unit that performs frequency analysis on the filtered waveform transferred from the filter processing unit to obtain spectral data;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotational speed signal of the rotating part, and the frequency component included in the spectrum data obtained by the frequency analysis unit And a precision diagnosis unit for comparing with the above and identifying the presence or absence of an abnormality of the rotating part and the site of the abnormality;
The wind turbine status monitoring device for wind power generation according to claim 5, comprising: A method of monitoring the state of a wind turbine for wind power generation, which monitors the state of a wind turbine for wind power generation having a rotating part that rotates relative to a stationary member, comprising:
A first detection step of detecting a signal from a first vibration sensor fixed to the stationary member or the rotating part;
A second detection step of detecting the vibration of the nacelle from a second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A state monitoring step of monitoring the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
Comparing the statistical value of the vibration of the nacelle detected by the second vibration sensor for each wind direction to determine a wind direction having a low fluctuation of the statistical value or the statistical value;
Equipped with
The state monitoring step performs state monitoring of the rotating component based on a waveform of a signal detected by the first vibration sensor when the nacelle faces the wind direction. Wind turbine condition monitoring method. A filtering step of extracting a predetermined damage filter frequency band from the waveform of the signal detected by the first vibration sensor;
And further
The condition monitoring process
A frequency analysis process for obtaining spectral data by frequency analysis of the filtered waveform transferred from the filtering process;
In the extracted damage filter frequency band, the theoretical damage frequency due to the damage of the rotating part calculated based on the rotation speed signal of the rotating part, and the frequency component included in the spectrum data obtained in the frequency analysis step And a precision diagnosis step of comparing the presence and absence of an abnormality of the rotating part and identifying the site of the abnormality,
The method for monitoring the state of a wind turbine for wind power generation according to claim 7, comprising: A plurality of wind powers each having a stationary member or a first vibration sensor fixed to a rotating part that rotates relative to the stationary member, and a second vibration sensor fixed to a nacelle of a wind turbine for wind power generation A wind turbine for power generation,
An arithmetic processing unit that monitors the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
A plurality of wind turbine status monitoring systems for wind power generation,
A plurality of wind power generations characterized by comparing statistical values of the vibrations of the nacelle of the wind turbines for wind power generation detected by the second vibration sensor, and determining the degree of influence of the wind conditions in the wind turbines for wind power generation Wind turbine condition monitoring system. A first detection step of detecting a signal from a first vibration sensor fixed to the stationary member or a rotating part that rotates relative to the stationary member;
A second detection step of detecting vibration of the nacelle from a second vibration sensor fixed to the nacelle of the wind turbine for wind power generation;
A state monitoring step of monitoring the state of the rotating component based on the waveform of the signal detected by the first vibration sensor;
A method of monitoring the state of a plurality of wind turbines for wind power generation,
A plurality of wind power generations characterized by comparing statistical values of the vibrations of the nacelle of the wind turbines for wind power generation detected by the second vibration sensor, and determining the degree of influence of the wind conditions in the wind turbines for wind power generation Monitoring method for wind turbines.

Images