JP2015031626A - Device for monitoring state of rolling bearing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily and appropriately select an installation position of a vibration sensor to a bearing device, in a device for monitoring a state of a rolling bearing, by using the analysis result of a vibration analysis method for analyzing vibration of the bearing device by a computer.SOLUTION: A vibration analysis method includes a step (S10) of inputting damage data of a rolling bearing, a step (S40) of calculating a history of displacement between inner and outer rings, which occurs on the rolling bearing during rotation of a rotary shaft of the rolling bearing, by a kinetics analysis program, a step (S60) of calculating a vibration characteristic model of a bearing device by a mode analysis program, and a step (S80) of calculating a vibration waveform in a prescribed position of the bearing device by giving, to the vibration characteristic model, a history of excitation force obtained by multiplying the displacement between the inner and outer rings calculated in the step S40 by a bearing spring constant. The installation position of a vibration sensor is selected by using the vibration waveform calculated by the vibration analysis method.

Description

  The present invention relates to a rolling bearing state monitoring device, and more particularly, to a technique for analyzing vibrations of a rolling bearing and a bearing device including a casing thereof by a computer, and a rolling bearing state monitoring device using the analysis result.

  Japanese Patent Laying-Open No. 2006-234785 (Patent Document 1) discloses an abnormality diagnosis device for a rolling bearing. In this abnormality diagnosis apparatus, frequency analysis of an electric signal from a vibration sensor is performed, and a spectrum peak larger than a reference value calculated based on the spectrum obtained by frequency analysis is extracted. Then, the frequency between the peaks and the frequency component resulting from bearing damage calculated based on the rotational speed are compared and collated, and the presence / absence and abnormal part of the rolling bearing are determined based on the collation result. (See Patent Document 1).

JP 2006-234785 A

“Equipment Diagnosis by Practical Vibration Method to Answer Field Questions”, Noriaki Inoue, Japan Plant Maintenance Association, p. 65-71

  In the above-described abnormality diagnosis device for a rolling bearing, the vibration sensor detects the vibration of the bearing device including the rolling bearing and its housing. When the peak of the vibration waveform detected by the vibration sensor exceeds a predetermined threshold value, it is diagnosed that the rolling bearing is abnormal.

  In order to increase the accuracy of abnormality diagnosis, the vibration sensor is required to be able to detect vibration due to bearing damage with high sensitivity. Therefore, it is necessary to install a vibration sensor at a position where high detection sensitivity can be obtained. Regarding the installation position of such a vibration sensor, for example, Non-Patent Document 1 ("Equipment Diagnosis by Practical Vibration Method That Answers Questions in the Field", Noriaki Inoue, Japan Plant Maintenance Association, p. 65-71) It is described that it is desirable to install it in a location that is close to the target bearing and has high rigidity. This is because if the rigidity of the installation position of the vibration sensor is low, a high frequency component generated due to damage to the bearing is attenuated, and the abnormality of the bearing may not be detected.

  However, since the method for selecting the installation position of the vibration sensor shown in Non-Patent Document 1 is a qualitative method, the vibration detected by the vibration sensor when the vibration sensor is actually attached to the bearing device and the bearing is damaged. Until the waveform is confirmed, it cannot be determined whether or not the selected installation position is appropriate. Therefore, it becomes difficult to install the vibration sensor at a position with high detection sensitivity, and there is a problem that the detection sensitivity of the vibration sensor cannot be secured. In particular, in a large-scale facility such as a wind power generation facility, it is not easy to experimentally confirm by incorporating a damaged bearing, and thus the above selection method is not realistic. As a result, it has been difficult to detect bearing damage early.

  For such a problem, it is desired that the vibration state of the bearing device can be analyzed in advance by a computer. If the vibration state at the position where the vibration sensor is installed can be predicted in advance by computer analysis, an installation position with high detection sensitivity can be easily selected. Further, since it is possible to find an installation position with high detection sensitivity without attaching a vibration sensor to an actual machine, the installation position of the vibration sensor can be easily selected even in a large-scale facility such as a wind power generation facility.

  The present invention has been made to solve such a problem, and an object of the present invention is to provide a vibration analysis method for analyzing vibrations of a rolling bearing and a bearing device including the casing thereof by a computer in a rolling bearing state monitoring apparatus. Using the result, it is possible to easily and appropriately select the installation position of the vibration sensor with respect to the bearing device.

  According to the present invention, the rolling bearing state monitoring device has a vibration sensor for measuring the vibration of the rolling bearing and the bearing device including the housing thereof, and the magnitude of vibration measured using the vibration sensor is predetermined. And a determination unit that determines that the rolling bearing is abnormal when the threshold value is exceeded. The installation position of the vibration sensor is selected using a vibration waveform calculated by a vibration analysis method for analyzing the vibration of the bearing device by a computer. The vibration analysis method calculates the displacement between the inner ring and the outer ring of the rolling bearing that occurs in the rolling bearing due to the step of inputting the shape data of damage to the contact portion between the rolling element of the rolling bearing and the raceway surface. And a step of calculating a vibration characteristic model indicating the vibration characteristics of the bearing device by a mode analysis program for analyzing a vibration mode of the bearing device, and a displacement calculated by the step of calculating the displacement between the inner ring and the outer ring. Calculating a vibration waveform at an arbitrary position of the bearing device by giving the vibration characteristic model a history of forces generated in the rolling bearing obtained by multiplying the spring constant of

  Preferably, the step of calculating the displacement of the inner and outer rings is a step of calculating a history of displacement generated in the rolling bearing due to damage when the rotating shaft of the rolling bearing is rotated by a dynamic analysis program for performing dynamic analysis of the rolling bearing. including.

  Preferably, by using a vibration analysis method, a plurality of vibration waveforms are calculated corresponding to a plurality of candidate positions that can be set as the installation position of the vibration sensor. Of the plurality of candidate positions, the candidate position with the maximum vibration acceleration amplitude is selected as the installation position of the vibration sensor.

  Preferably, the bearing device includes a bearing device provided on a main shaft in the wind power generation facility. For each of the plurality of candidate positions, a ratio of the vibration acceleration amplitude calculated by the vibration analysis method to the vibration acceleration amplitude excited in the bearing device by the excitation force generated in the main shaft is calculated. Of the plurality of candidate positions, the candidate position having the maximum ratio is selected as the installation position of the vibration sensor.

  Preferably, the bearing device includes a bearing device provided in a speed increaser in a wind power generation facility. For each of the plurality of candidate positions, the ratio of the vibration acceleration amplitude calculated by the vibration analysis method to the vibration acceleration amplitude excited in the bearing device by the excitation force generated in the gear of the gear box is calculated. Of the plurality of candidate positions, the candidate position having the maximum ratio is selected as the installation position of the vibration sensor.

  Preferably, the bearing device includes a bearing device provided in a generator in a wind power generation facility. The generator is connected to a speed increaser in the wind power generation facility by a coupling unit. For each of the plurality of candidate positions, the ratio of the vibration acceleration amplitude calculated by the vibration analysis method to the vibration acceleration amplitude excited in the bearing device by the excitation force generated in the coupling unit is calculated. Of the plurality of candidate positions, the candidate position having the maximum ratio is selected as the installation position of the vibration sensor.

  Preferably, the rolling bearing state monitoring device further includes a selection unit that selects an installation position of the vibration sensor using a vibration waveform calculated by a vibration analysis method.

  Preferably, the predetermined threshold value is determined by using a vibration waveform assumed when an abnormality occurs in the rolling bearing at the selected installation position of the vibration sensor.

  In the present invention, the vibration waveform of the bearing device when damage occurs inside the bearing can be predicted in advance by the computer. Therefore, according to the present invention, it is possible to easily and appropriately select the installation position of the vibration sensor with respect to the bearing device in the rolling bearing state monitoring device using the above prediction result.

It is the figure which showed the model of the bearing apparatus analyzed by the vibration analysis method by Embodiment 1 of this invention. FIG. 3 is a block diagram illustrating a main part of a hardware configuration of the vibration analysis apparatus according to the first embodiment. It is a functional block diagram which shows the structure of the vibration analyzer shown in FIG. 2 functionally. It is a flowchart for demonstrating the process sequence of the vibration-analysis method performed with the vibration-analysis apparatus shown in FIG. It is a flowchart for demonstrating the process sequence of the selection method of the installation position of the vibration sensor using the vibration-analysis method shown in FIG. It is the figure which showed roughly the structure of the wind power generation equipment with which the state monitoring apparatus of a rolling bearing is applied. It is a flowchart for demonstrating the process sequence of the selection method of the installation position of the vibration sensor with respect to the bearing apparatus shown in FIG. It is a functional block diagram which shows the structure of the data processor shown in FIG. 6 functionally. It is the figure which showed the vibration waveform of the bearing apparatus when abnormality has not generate | occur | produced in the bearing apparatus. It is the figure which showed the vibration waveform of the bearing apparatus seen when the surface roughness of the bearing ring of a bearing apparatus and poor lubrication generate | occur | produce. It is the figure which showed the vibration waveform of the bearing apparatus in the initial stage when peeling generate | occur | produces in the bearing ring of a bearing apparatus. It is the figure which showed the vibration waveform of the bearing apparatus seen in the last stage of peeling abnormality. Changes in the effective value of the vibration waveform of the bearing device and the effective value of the alternating current component of the envelope waveform when separation occurs on a part of the bearing ring and then the separation moves across the entire ring. FIG. It is the figure which showed the time change of the effective value of the vibration waveform of a bearing device, and the effective value of the alternating current component of an envelope waveform when the surface roughness and poor lubrication of the bearing ring of a bearing device generate | occur | produce. It is a functional block diagram which shows the other structure of a data processor functionally. It is a flowchart for demonstrating the process sequence of the vibration-analysis method performed with the vibration-analysis apparatus by a modification. 6 is a flowchart for explaining a processing procedure of a vibration analysis method executed by the vibration analysis apparatus according to the second embodiment. FIG. 10 is a functional block diagram functionally showing the configuration of a vibration analyzing apparatus according to a third embodiment. 10 is a flowchart for explaining a processing procedure of a vibration analysis method executed by the vibration analysis apparatus according to the third embodiment. It is a flowchart for demonstrating the process sequence of the selection method of the installation position of the vibration sensor with respect to the bearing apparatus for gearboxes. It is a flowchart for demonstrating the process sequence of the selection method of the installation position of the vibration sensor with respect to the generator bearing apparatus.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

[Embodiment 1]
FIG. 1 is a diagram showing a model of a bearing device 10 analyzed by the vibration analysis method according to Embodiment 1 of the present invention. With reference to FIG. 1, the bearing device 10 includes a rolling bearing 20 and a housing 30. In this Embodiment 1, the case where the rolling bearing 20 is a ball bearing is demonstrated. The rolling bearing 20 includes an inner ring 22, a plurality of rolling elements 24, and an outer ring 26.

  The inner ring 22 is fitted on the rotary shaft 12 and rotates integrally with the rotary shaft 12. The outer ring 26 is a fixed ring that is provided on the outer peripheral side of the inner ring 22 and is fitted into the housing 30. Each of the plurality of rolling elements 24 is a spherical ball, and is interposed between the inner ring 22 and the outer ring 26 while maintaining a distance from each other by a cage (not shown). The housing 30 is fixed to the base 40 with bolts (not shown).

  In this model, the outer ring 26 that is a fixed ring is connected to the housing 30 via linear springs kF1 to kF3 in the bearing radial direction at the position of the rolling element within the load range among the plurality of rolling elements 24. Shall. Moreover, about the coupling | bond part of the housing 30 and the base 40, mass m1, m2 shall act on the linear springs kH1, kH2 imitating bolt coupling etc., respectively.

(Description of vibration analysis method of bearing device 10)
FIG. 2 is a block diagram showing a main part of the hardware configuration of the vibration analyzing apparatus according to the first embodiment. Referring to FIG. 2, vibration analysis apparatus 100 includes an input unit 110, an interface (I / F) unit 120, a CPU (Central Processing Unit) 130, a RAM (Random Access Memory) 140, and a ROM (Read Only). Memory) 150 and an output unit 160.

  The CPU 130 implements a vibration analysis method that will be described in detail later by executing various programs stored in the ROM 150. The RAM 140 is used as a work area by the CPU 130. The ROM 150 records a program including steps of a flowchart (described later) showing the procedure of the vibration analysis method. The input unit 110 is a means for reading data from the outside, such as a keyboard, a mouse, a recording medium, and a communication device. The output unit 160 is a means for outputting a calculation result by the CPU 130 such as a display, a recording medium, or a communication device.

  FIG. 3 is a functional block diagram functionally showing the configuration of the vibration analysis apparatus 100 shown in FIG. Referring to FIG. 3 together with FIG. 2, the vibration analysis apparatus 100 includes an approach amount change amount calculation unit 205, a dynamic analysis model setting unit 210, a displacement calculation unit 220, a vibration characteristic calculation unit 230, and a vibration waveform calculation. Unit 240, sensor position selection unit 250, and input unit 110 and output unit 160 described above.

  The vibration prediction of the bearing device 10 by the vibration analysis device 100 is roughly divided into two parts. That is, one is that when the rotary shaft of the rolling bearing 20 rotates due to damage caused in the contact portion between the rolling element 24 of the rolling bearing 20 and the raceway surface (the outer peripheral surface of the inner ring 22 or the inner peripheral surface of the outer ring 26). A history of displacement between the inner and outer rings (between the inner ring 22 and the outer ring 26) generated in the rolling bearing 20 is predicted. The other is to predict the vibration waveform generated at the installation location of a vibration sensor (not shown) installed in the bearing device 10 by the excitation force based on the displacement generated in the rolling bearing 20 being transmitted through the housing 30. is there.

  From the input unit 110, specification data of the rolling bearing 20, data on the shape of damage to the contact portion between the rolling element 24 and the raceway surface (hereinafter also referred to as “damage data”), lubrication conditions, operating conditions (rotational speed, etc.) ), Specification data and the like of the rotary shaft 12 and the housing 30 coupled to the rolling bearing 20 are input. The input unit 110 may be a data input unit using a Web interface, a reading unit that reads data from a recording medium on which the data is recorded according to a predetermined format, or an external unit that is transmitted according to a predetermined format. The communication apparatus etc. which receive the said data which come may be sufficient.

  The approach amount change amount calculation unit 205 receives data relating to the rolling element 24 and its trajectory, and damage data from the input unit 110. Then, the approach amount change amount calculation unit 205 calculates the amount of change in the approach amount between the rolling element 24 and the raceway surface caused by the given damage by a contact analysis program for analyzing the contact between the rolling element 24 and the raceway surface. calculate. The contact analysis program calculates the contact pressure distribution of the contact portion using, for example, a finite element method (FEM), and calculates the amount of change in the approach amount between the rolling element 24 and the raceway surface according to the presence or absence of damage. Is.

  For the contact analysis, it is preferable to use an elasto-plastic analysis taking into account the plastic deformation of the contact portion. This is because when the contact portion between the rolling element 24 and the raceway surface is damaged, the surface pressure may increase to the extent that local plastic deformation occurs. In order to simplify the calculation, the approach amount change amount calculation unit 205 may calculate the amount of change in the elastic approach amount of the rolling element 24 caused by damage using an elastic analysis for the contact analysis.

  The dynamic analysis model setting unit 210 receives the various data input from the input unit 110 and receives the change amount of the approach amount calculated by the approach amount change amount calculation unit 205. Then, the dynamic analysis model setting unit 210 sets a dynamic analysis model of the rolling bearing 20 for performing dynamic analysis in consideration of the dynamic characteristics of the rolling bearing 20. The dynamic analysis is a method in which an equation of motion is established for each component of the rolling bearing 20 (inner ring 22, rolling element 24, and outer ring 26), and simultaneous ordinary differential equations are integrated along the time axis. By the analysis, it is possible to simulate in real time the interference force between the constituent elements that change every moment, the behavior of each constituent element, and the like.

  In this dynamic analysis model, an approach amount change amount calculated by the approach amount change amount calculation unit 205 is given. Each component of the rolling bearing 20 is a rigid body, and the rotating shaft 12 and the housing 30 coupled to the rolling bearing 20 are elastic bodies having a predetermined mass and vibration mode. The inertial force of the rotating body (inner ring, rolling element 24 and rotating shaft 12) and the influence of gravity acting on each component are also reflected in the model.

  The displacement calculation unit 220 uses the dynamic analysis model set by the dynamic analysis model setting unit 210 to change the approach amount between the track and the rolling element given by the damage data input from the input unit 110. A history of the displacement between the inner and outer rings generated in the rolling bearing 20 when the rolling bearing 20 rotates is calculated. More specifically, the displacement calculation unit 220 uses the dynamic analysis model to calculate the amount of approach between the rolling element 24 and the raceway surface when the rotary shaft 12 is operated according to the operation conditions input from the input unit 110. A history of displacement between the inner and outer rings generated in the rolling bearing 20 due to the change is calculated.

  On the other hand, the vibration characteristic calculation unit 230 calculates a vibration characteristic model indicating the vibration transfer characteristic of the bearing device 10 by a mode analysis program for analyzing the vibration mode of the bearing device 10. In the first embodiment, a vibration mode indicating the vibration characteristic of the bearing device 10 is calculated as a vibration characteristic model by so-called theoretical mode analysis. In the mode analysis, various vibrations are obtained by synthesizing a plurality of natural modes, and the natural modes and natural frequencies are obtained. The theoretical mode analysis is to mathematically determine what vibration mode (eigenvalue information) the structure (elastic body) has. Specifically, in the theoretical mode analysis, a model of an object (bearing device 10) is created by determining the shape, mass distribution, rigidity distribution and constraint conditions of the object, and a mass matrix and stiffness representing the mass characteristics of the model. A natural frequency and a natural mode of an object are obtained by solving a natural value and a natural value vector from a stiffness matrix representing characteristics by theoretical analysis or numerical calculation.

  The vibration characteristic calculation unit 230 receives specification data such as the shape, material density, Young's modulus, Poisson's ratio, and the like of the bearing device 10 from the input unit 110. Further, the rolling element 24 is treated as a linear spring, and the vibration characteristic calculation unit 230 has spring information for treating each of the rolling element 24 and the coupling portion (for example, bolt coupling portion) between the housing 30 and the base 40 as a linear spring. Is received from the input unit 110. And the vibration characteristic calculation part 230 calculates the vibration characteristic model (vibration mode) of the bearing apparatus 10 with a mode analysis program (theoretical mode analysis program).

  In the above vibration characteristic model, in order to reproduce the actual vibration characteristics more faithfully, only the connection surface near the bolt among the connection surfaces of the bolt connection portion (for example, the connection portion between the housing 30 and the base 40) is connected. It is good to be in the state where it was done. This is because the actual bolt coupling portion is coupled with a high compressive force on the coupling surface near the bolt, and is coupled with a relatively small force on the coupling surface away from the bolt. This tendency becomes more prominent as the rigidity of the parts joined by the bolt is lower. As an example of the shape in the vicinity of the bolt to be coupled, a ring shape concentric with the rotation axis of the bolt, in which the bolt hole diameter is the inner diameter and the bolt head diameter is the outer diameter, can be used.

  The vibration waveform calculation unit 240 performs a transient response analysis using the vibration characteristic model calculated by the vibration characteristic calculation unit 230, thereby generating a vibration waveform at a specified position on the bearing device 10 (for example, a planned installation location of the vibration sensor). calculate. More specifically, the vibration waveform calculation unit 240 receives a history of displacement between the inner and outer rings that occurs in the rolling bearing 20 from the displacement calculation unit 220, and the received spring-constant between the inner and outer rings (hereinafter referred to as “bearing”). Multiplying by "spring constant"), the history of the excitation force generated in the rolling bearing 20 is calculated. Then, the vibration waveform calculation unit 240 gives a history of the excitation force to the vibration characteristic model calculated by the vibration characteristic calculation unit 230, and calculates the vibration waveform at that time.

  The bearing spring constant is calculated as follows, for example. That is, using the Hertz theory, the spring constant at the contact portion between the inner ring 22 and the rolling element 24 (referred to as “first spring constant”) and the spring constant at the contact portion between the outer ring 26 and the rolling element 24 ( (Referred to as “second spring constant”). Next, the first spring constant and the second spring constant are used as the spring constant of the series spring, and the spring constant (referred to as “rolling body spring constant”) for one rolling element 24 is calculated. And a bearing spring constant is calculated by synthesize | combining the rolling element spring constant about the rolling element in the load range.

  In the first embodiment, the history of the excitation force obtained by multiplying the displacement between the inner and outer rings by the bearing spring constant in the vibration waveform calculation unit 240 is the vibration characteristic model calculated by the vibration characteristic calculation unit 230. , At least one point on the central axis of the inner ring 22 of the rolling bearing 20 is given. As a result, a combination of the prediction of the displacement between the inner and outer rings of the rolling bearing 20 using the dynamic analysis program and the excitation force based thereon and the prediction of the vibration transfer characteristics of the bearing device 10 using the mode analysis program are combined. Vibration analysis can be performed.

  The output unit 160 outputs the vibration waveform calculated by the vibration waveform calculation unit 240. The output unit 160 may be a display for displaying the vibration waveform, writing means for writing the vibration waveform data to a recording medium according to a predetermined format, or externally transmitting the vibration waveform data according to a predetermined format. It may be a communication device or the like that transmits to.

  As described above, in the first embodiment, the history of the excitation force generated in the rolling bearing 20 is obtained by multiplying the displacement between the inner and outer rings of the rolling bearing 20 calculated using the dynamic analysis program by the bearing spring constant. And a response analysis (vibration analysis) is performed by giving the history of the calculated excitation force to the vibration characteristic model. By such a method, the inner and outer rings that are actually generated even when the excitation force generated when the rolling element passes through the damaged part is small, for example, because the damaged part shows a gradual shape change such as damage due to wear. Response analysis (vibration analysis) can be performed based on the displacement between the two.

  FIG. 4 is a flowchart for explaining the processing procedure of the vibration analysis method executed by the vibration analysis apparatus 100 shown in FIG. Referring to FIG. 4, first, vibration analysis apparatus 100 inputs data relating to rolling element 24 and its raceway (the outer peripheral surface of inner ring 22 or the inner peripheral surface of outer ring 26) and damage data given to rolling bearing 20. (Step S10).

  Next, the vibration analyzing apparatus 100 calculates a change in the approach amount between the rolling element 24 and the raceway surface caused by the damage input in step S10 in accordance with a contact analysis program prepared in advance (step S20). Next, the vibration analyzing apparatus 100 reads various data for executing dynamic analysis for the rolling bearing 20 (step S30). Specifically, the vibration analysis apparatus 100 reads data such as the specification data and operating conditions of the rolling bearing 20 and the mass and spring characteristics of the rotating shaft 12 and the housing 30 from the input unit 110, and is further calculated in step S20. Read the amount of change in approaching distance.

  Subsequently, the vibration analysis apparatus 100 sets a dynamic analysis model of the rolling bearing 20 based on the various data read in step S30. Then, the vibration analysis apparatus 100 is generated in the rolling bearing 20 by the change in the approach amount when the rotating shaft 12 is operated according to the operation condition input in step S10 according to the dynamic analysis program using the dynamic analysis model. A history of displacement between the inner and outer rings is calculated (step S40).

  Next, the vibration analysis device 100 reads various data for executing the theoretical mode analysis of the bearing device 10 (step S50). Specifically, the vibration analysis device 100 reads data such as the shape of the bearing device 10, the material density, Young's modulus, and Poisson's ratio. The vibration analysis apparatus 100 also reads spring information for handling each of the rolling elements 24 and the coupling portions (for example, bolt coupling portions) between the housing 30 and the base 40 as linear springs. These pieces of data may be read from the input unit 110 or may be previously stored as internal data.

  When each data is input in step S50, vibration analysis device 100 calculates mode information of bearing device 10 in accordance with a theoretical mode analysis program prepared in advance (step S60). Specifically, the vibration analysis device 100 calculates a vibration mode (natural frequency and natural mode) of the bearing device 10 using a theoretical mode analysis program based on each data input in step S50.

  Next, the vibration analysis device 100 reads various data for executing a transient response analysis (mode analysis method) of the bearing device 10 (step S70). Specifically, the vibration analysis apparatus 100 reads the mode information calculated in step S60, the change amount of the approach amount calculated in step S20, the history of the displacement between the inner and outer rings calculated in step S40, and the like. .

  Then, the vibration analysis device 100 calculates a vibration waveform of the bearing device 10 in accordance with a transient response analysis (mode analysis method) program prepared in advance (step S80). Specifically, the vibration analyzing apparatus 100 has the vibration mode calculated in step S60, the history of the excitation force obtained by multiplying the displacement (history) between the inner and outer rings calculated in step S40 by the bearing spring constant. By giving to at least one point on the central axis of the inner ring 22 of the bearing device 10, a vibration waveform generated in the bearing device 10 due to the displacement between the inner and outer rings calculated in step S40 is calculated.

  As described above, in the vibration analyzing apparatus 100, it is possible to simulate the vibration waveform of the bearing apparatus 10 when damage occurs inside the rolling bearing 20. Accordingly, as described below, in a state monitoring device (Condition Monitoring System) for monitoring the state (abnormality) of the rolling bearing using the vibration analysis device 100, the installation position of the vibration sensor (hereinafter referred to as “sensor”). Can also be selected easily and appropriately. Furthermore, since the threshold value for determining the abnormality of the bearing can be determined, it is possible to determine the abnormality of the rolling bearing in the state monitoring device using the threshold value.

(Description of how to set the installation position of the vibration sensor)
Hereinafter, the selection method of the installation position of the vibration sensor with respect to the bearing device 10 using the vibration analysis method according to the first embodiment will be described. This selection method can be realized by the sensor position selection unit 250 in the vibration analysis apparatus 100 shown in FIG. Alternatively, the position of the vibration sensor may be selected outside the state monitoring device using the vibration waveform data output from the output unit 160 of the vibration analysis device 100. In the first embodiment, as an example, a configuration in which the sensor position selection unit 250 selects an installation position of a vibration sensor will be described.

  Referring to FIG. 3, data of a plurality of candidate positions that can be set as vibration sensor installation positions are input from input unit 110. The vibration analysis apparatus 100 predicts the vibration waveform generated at the candidate position due to the damage of the rolling bearing 20 for each of the plurality of candidate positions by executing the vibration analysis method illustrated in FIG. 4. The vibration waveform calculation unit 240 outputs a plurality of vibration waveforms calculated corresponding to a plurality of candidate positions.

  The sensor position selection unit 250 receives data of a plurality of candidate positions from the input unit 110. Further, the vibration waveform calculation unit 240 receives a plurality of vibration waveforms calculated corresponding to the plurality of candidate positions. And the sensor position selection part 250 selects the candidate position corresponding to the vibration waveform from which the acceleration amplitude of vibration becomes the largest among several vibration waveforms as an installation position of a vibration sensor. The sensor position selection unit 250 outputs the selected installation position to the output unit 160.

  Through the processing as described above, the installation position of the vibration sensor with respect to the bearing device 10 is selected. These processes can be summarized in the following process flow.

  FIG. 5 is a flowchart for explaining the processing procedure of the method for selecting the installation position of the vibration sensor using the vibration analysis method shown in FIG. Note that the flowchart shown in FIG. 5 can be realized by executing a program stored in advance in the vibration analysis apparatus 100. Alternatively, it may be executed outside the state monitoring device using vibration waveform data output from the vibration analysis device 100.

  Referring to FIG. 5, first, data of a plurality of candidate positions that can be set as vibration sensor installation positions are input from input unit 110 (step S1000).

  Next, when the vibration analysis apparatus 100 reads data of a plurality of candidate positions from the input unit 110, each candidate of the bearing apparatus 10 when damage occurs inside the rolling bearing 20 by executing a vibration analysis method. A vibration waveform generated at the position is calculated (step S1100).

  The vibration analyzing apparatus 100 extracts a vibration waveform that maximizes the acceleration amplitude of vibration from a plurality of vibration waveforms calculated corresponding to a plurality of candidate positions. Then, a candidate position corresponding to the extracted vibration waveform is selected as a vibration sensor installation position (step S1200).

  As described above, in the first embodiment, the installation position of the vibration sensor with respect to the bearing device can be selected using the analysis result obtained by the vibration analysis method. Thereby, compared with the conventional qualitative selection method, the installation position of a vibration sensor can be selected easily and appropriately.

  Furthermore, since a threshold value for determining a bearing abnormality can be determined using the vibration waveform at the installation position of the selected vibration sensor, the state monitoring device uses the threshold value to determine the rolling bearing. Abnormality determination can be performed.

  In the following, as an example, regarding the rolling bearing state monitoring device in which the vibration sensor installation position and the abnormality determination threshold value determined using the analysis result of the vibration analysis device 100 are used, the rolling bearing in the wind power generation facility is used. A state monitoring apparatus will be described representatively.

(Whole structure of wind power generation equipment)
FIG. 6 is a diagram schematically illustrating a configuration of a wind power generation facility to which the rolling bearing state monitoring device is applied. Referring to FIG. 6, wind power generation facility 310 includes a main shaft 320, a blade 330, a speed increaser 340, a power generator 350, and a main shaft bearing device (hereinafter also simply referred to as “bearing device”) 360. A vibration sensor 370 and a data processing device 380. The speed increaser 340, the generator 350, the bearing device 360, the vibration sensor 370, and the data processing device 380 are stored in the nacelle 390, and the nacelle 390 is supported by the tower 400.

  The main shaft 320 enters the nacelle 390, is connected to the input shaft of the speed increaser 340, and is rotatably supported by the bearing device 360. The main shaft 320 transmits the rotational torque generated by the blade 330 that receives wind force to the input shaft of the speed increaser 340. The blade 330 is provided at the tip of the main shaft 320 and converts wind force into rotational torque and transmits it to the main shaft 320.

  The bearing device 360 is fixed in the nacelle 390 and supports the main shaft 320 rotatably. The bearing device 360 is composed of a rolling bearing and a housing, and here, the rolling bearing is composed of a ball bearing. The vibration sensor 370 is fixed to the bearing device 360. The vibration sensor 370 detects the vibration of the bearing device 360 and outputs the detected value to the data processing device 380. The vibration sensor 370 is constituted by, for example, an acceleration sensor using a piezoelectric element.

  The speed increaser 340 is provided between the main shaft 320 and the power generator 350, and increases the rotational speed of the main shaft 320 and outputs it to the power generator 350. As an example, the speed increaser 340 is configured by a gear speed increasing mechanism including a planetary gear, an intermediate shaft, a high speed shaft, and the like. Although not specifically shown, a plurality of bearings for rotatably supporting a plurality of shafts are also provided in the speed increaser 340. The generator 350 is connected to the output shaft of the speed increaser 340 by a coupling portion (joint), and generates electric power by rotational torque received from the speed increaser 340. The generator 350 is constituted by, for example, an induction generator. The generator 350 is also provided with a bearing that rotatably supports the rotor.

  The data processing device 380 is provided in the nacelle 390 and receives a detection value of vibration of the bearing device 360 from the vibration sensor 370. Then, the data processing device 380 diagnoses an abnormality of the bearing device 360 using a vibration waveform of the bearing device 360 according to a method described later according to a preset program.

(Description of the method for selecting the installation position of the vibration sensor in the spindle bearing device)
Hereinafter, a method of selecting the installation position of the vibration sensor 370 with respect to the main shaft bearing device 360 will be described.

  In the wind power generation facility 310 shown in FIG. 6, when the blade 330 rotates by receiving wind force, a load from the blade 330 is applied to the main shaft 320. This load includes a load generated in the axial force of the main shaft 320 when the blade 330 cuts the wind, an alternating load due to rotation of the rotating portion including the blade 330, an unbalanced load, and the like. These loads also fluctuate due to constantly changing winds. The main shaft 320 vibrates using the load applied from the blade 330 as an excitation force. The excitation force generated in the main shaft 320 is transmitted to the bearing device 360 to excite vibrations in the bearing device 360.

  As described above, the bearing device 360 is excited by vibration (noise) by receiving the excitation force generated in the main shaft 320. The vibration generated by the excitation force from the main shaft 320 is superimposed on the vibration generated due to the damage of the rolling bearing. Therefore, when the excitation force generated in the main shaft 320 increases due to the application of a large load, the detection sensitivity of the vibration sensor 370 decreases due to the influence of noise, so that the bearing device 360 when the rolling bearing is damaged occurs. It becomes difficult to detect the vibration of the. As a result, the state monitoring device may not be able to accurately diagnose the abnormality of the bearing device 360.

  Therefore, when selecting the installation position of the vibration sensor 370, the magnitude of vibration generated at the installation position of the vibration sensor 370 is added to the vibration waveform of the bearing device 360 calculated by the above-described vibration analysis method, The calculation is performed in consideration of the influence of vibration (noise) generated by the excitation force on the detection sensitivity of the vibration sensor 370. If the vibration sensor 370 can be installed at a position where the influence of noise is small, the vibration sensor 370 can detect vibration due to damage to the bearing with high sensitivity.

  In the first embodiment, as an example, the magnitude of the influence of noise on the detection sensitivity of the vibration sensor 370 is evaluated using a so-called SN ratio (Signal to Noise ratio). The SN ratio in the main shaft bearing device 360 is the acceleration amplitude of vibration excited in the bearing device 360 due to damage to the bearing with respect to the acceleration amplitude of vibration (noise) excited in the bearing device 360 by the excitation force generated in the main shaft 320. Calculated as a ratio. Then, the installation position of the vibration sensor 370 is selected based on the calculated SN ratio.

  FIG. 7 is a flowchart for explaining a processing procedure of a method for selecting an installation position of the vibration sensor 370 with respect to the main shaft bearing device 360. The flowchart shown in FIG. 7 realizes step S1200 (selection of sensor position) in the flowchart shown in FIG. 5 by the processing of steps S1210 to S1230.

  The flowchart shown in FIG. 7 can be realized by executing a program stored in advance in the data processing device 380. Alternatively, the vibration waveform data output from the data processing device 380 may be used to execute it outside the state monitoring device. In this embodiment, as an example, a configuration in which the data processing device 380 has the function of the sensor position selection unit 250 (FIG. 3) will be described.

  Referring to FIG. 7, first, data of a plurality of candidate positions that can be set as installation positions of vibration sensor 370 with respect to bearing device 360 are input from an input unit (not shown) of data processing device 380 (step). S1000).

  Next, when the data processing device 380 reads data of a plurality of candidate positions from the input unit, the data processing device 380 calculates a vibration waveform generated at each candidate position of the bearing device 360 when damage occurs inside the rolling bearing (step S1100). ).

  Subsequently, the data processing device 380 calculates a vibration waveform excited on the bearing device 360 by the excitation force generated on the main shaft 320 (step S1210). For example, the data processing device 380 calculates a load (excitation force) applied to the main shaft 320 using data such as the shape of the blade 330, the wind speed, and the rotation speed. Then, the data processing device 380 calculates a vibration waveform generated in the bearing device 360 by receiving the calculated excitation force by performing a response analysis using the vibration characteristic model of the bearing device 360.

  Next, the data processing device 380 uses the vibration waveform generated in each candidate position of the bearing device 360 calculated in step S1100 and the vibration waveform of the bearing device 360 calculated in step S1210 to generate the SN at each candidate position. The ratio is calculated (step S1220). Thereby, a plurality of SN ratios are calculated in correspondence with a plurality of candidate positions.

  Next, the data processing device 380 extracts the one with the largest value from the plurality of calculated SN ratios. Then, the candidate position corresponding to the extracted SN ratio is selected as the installation position of the vibration sensor 370 (step S1230).

  In the processing flow of FIG. 7, as described above, as a method of calculating the SN ratio, the vibration waveform of the bearing device due to the bearing damage and the vibration waveform of the bearing device due to the excitation force of the main shaft 320 are separately obtained. Although the structure to calculate was illustrated, it is good also as a structure which calculates the vibration waveform of the bearing apparatus by each excitation force in the state which gave the excitation force produced by the damage of a bearing, and the excitation force from the main shaft 20 to the bearing apparatus simultaneously. .

  As described above, a candidate position that minimizes the influence of noise among the plurality of candidate positions is selected as the installation position of the vibration sensor 370. Thereby, the vibration sensor 370 can detect the vibration of the bearing device 360 when the rolling bearing is damaged with high sensitivity.

  In addition, since the installation position of the vibration sensor 370 can be selected using the analysis result by the vibration analysis method and the response analysis result using the excitation force generated in the main shaft 320, the vibration sensor 370 is compared with the conventional qualitative selection method. Can be selected easily and appropriately.

  Next, the abnormality diagnosis of the rolling bearing in the bearing device 360 in which the vibration sensor 370 is installed by executing the selection method described above will be described.

(Configuration of data processing device)
FIG. 8 is a functional block diagram functionally showing the configuration of the data processing device 380 shown in FIG. Referring to FIG. 8, data processing device 380 includes high-pass filters (hereinafter referred to as “HPF (High Pass Filter)”) 410, 450, effective value calculation units 420, 460, envelope processing unit 440, A storage unit 480 and a diagnosis unit 490 are included.

  The HPF 410 receives the detection value of the vibration of the bearing device 360 from the vibration sensor 370. Then, the HPF 410 passes a signal component higher than a predetermined frequency for the received detection signal, and blocks the low frequency component. The HPF 410 is provided to remove a direct current component included in the vibration waveform of the bearing device 360. Note that the HPF 410 may be omitted if the output from the vibration sensor 370 does not include a DC component.

  The effective value calculation unit 420 receives the vibration waveform of the bearing device 360 from which the DC component is removed from the HPF 410. And the effective value calculating part 420 calculates the effective value (it is also called "RMS (Root Mean Square) value") of the vibration waveform of the bearing apparatus 360, and memorize | stores the effective value of the calculated vibration waveform in a memory | storage part. Output to 480.

  Envelope processing unit 440 receives a vibration detection value of bearing device 360 from vibration sensor 370. Envelope processing unit 440 generates an envelope waveform of the vibration waveform of bearing device 360 by performing envelope processing on the received detection signal. Note that various known methods can be applied to the envelope processing calculated by the envelope processing unit 440. As an example, the vibration waveform of the bearing device 360 measured using the vibration sensor 370 is rectified to an absolute value. By passing through a low pass filter (LPF), an envelope waveform of the vibration waveform of the bearing device 360 is generated.

  The HPF 450 receives the envelope waveform of the vibration waveform of the bearing device 360 from the envelope processing unit 440. Then, the HPF 450 passes a signal component higher than a predetermined frequency with respect to the received envelope waveform, and blocks a low frequency component. The HPF 450 is provided to remove a DC component contained in the envelope waveform and extract an AC component of the envelope waveform.

  The effective value calculator 460 receives from the HPF 450 the envelope waveform from which the DC component has been removed, that is, the AC component of the envelope waveform. Then, the effective value calculation unit 460 calculates the effective value (RMS value) of the AC component of the received envelope waveform, and outputs the calculated effective value of the AC component of the envelope waveform to the storage unit 480.

  The storage unit 480 synchronizes the effective value of the vibration waveform of the bearing device 360 calculated by the effective value calculation unit 420 with the effective value of the alternating current component of the envelope waveform calculated by the effective value calculation unit 460 from time to time. To do. The storage unit 480 is configured by, for example, a readable / writable nonvolatile memory.

  The diagnosis unit 490 reads the effective value of the vibration waveform of the bearing device 360 and the effective value of the alternating current component of the envelope waveform, which are stored in the storage unit 480 every moment, from the storage unit 480, and based on the two read effective values. Thus, the abnormality of the bearing device 360 is diagnosed. Specifically, the diagnosis unit 490 diagnoses an abnormality of the bearing device 360 based on the temporal change of the effective value of the vibration waveform of the bearing device 360 and the effective value of the AC component of the envelope waveform.

  That is, since the effective value of the vibration waveform of the bearing device 360 calculated by the effective value calculation unit 420 is the effective value of the raw vibration waveform that has not been subjected to the envelope processing, for example, separation occurs in a part of the raceway ring. However, although the increase in the value is small for the impulse-like vibration in which the amplitude increases only when the rolling element passes through the peeled portion, it occurs when the contact surface between the race and the rolling element is rough or poorly lubricated. For continuous vibrations, the value increases.

  On the other hand, the effective value of the AC component of the envelope waveform calculated by the effective value calculation unit 460 is small and increases in some cases with respect to the continuous vibration that occurs when the raceway surface is rough or poorly lubricated. Although not, the increase in value increases for impulse vibration. Therefore, by using the effective value of the vibration waveform of the bearing device 360 and the effective value of the AC component of the envelope waveform, it is possible to detect an abnormality that cannot be detected by only one of the effective values, and to realize a more accurate abnormality diagnosis. .

  9 to 12 are diagrams showing vibration waveforms of the bearing device 360 measured using the vibration sensor 370. FIG. 9 to 12 show vibration waveforms when the rotation speed of the main shaft 320 (FIG. 6) is constant.

  FIG. 9 is a diagram illustrating a vibration waveform of the bearing device 360 when no abnormality has occurred in the bearing device 360. Referring to FIG. 9, the horizontal axis indicates time, and the vertical axis is an index indicating the magnitude of vibration, and here, as an example, acceleration of vibration.

  FIG. 10 is a diagram showing a vibration waveform of the bearing device 360 that is seen when surface roughness or poor lubrication of the bearing ring of the bearing device 360 occurs. Referring to FIG. 10, when the raceway surface roughness or poor lubrication occurs, the acceleration increases and the state in which the acceleration increases continuously occurs. There are no noticeable peaks in the vibration waveform. Therefore, with respect to such a vibration waveform, the effective value of the vibration waveform (output of the effective value calculation unit 420 (FIG. 8)) and the effective value of the AC component of the envelope waveform (effective) when no abnormality occurs in the bearing device 360. As compared with the output of the value calculation unit 460 (FIG. 8), the effective value of the raw vibration waveform not subjected to the envelope processing increases, and the effective value of the AC component of the envelope waveform does not increase so much.

  FIG. 11 is a diagram illustrating a vibration waveform of the bearing device 360 at an initial stage when separation occurs on the raceway of the bearing device 360. Referring to FIG. 11, the initial stage of the peeling abnormality is a state in which peeling occurs in a part of the raceway, and a large vibration is generated when the rolling element passes through the peeling portion. Vibration occurs periodically according to the rotation of the shaft. When the rolling element passes through other than the peeled portion, the increase in acceleration is small. Therefore, when such a vibration waveform is compared with the effective value of the vibration waveform and the effective value of the alternating current component of the envelope waveform when no abnormality has occurred in the bearing device 360, the effective value of the alternating current component of the envelope waveform increases. The effective value of the raw vibration waveform does not increase so much.

  FIG. 12 is a diagram showing a vibration waveform of the bearing device 360 seen in the final stage of the peeling abnormality. Referring to FIG. 12, the final stage of the separation abnormality is a state where the separation is transferred to the entire area of the raceway, and the acceleration increases as a whole compared with the initial stage of the abnormality, and the fluctuation with respect to the amplitude of the acceleration. Will weaken. Therefore, when compared with the effective value of the vibration waveform and the effective value of the alternating current component of the envelope waveform at the initial stage of the separation abnormality, the effective value of the raw vibration waveform increases, and the alternating current component of the envelope waveform increases. The effective value decreases.

  FIG. 13 shows the effective value of the vibration waveform of the bearing device 360 and the effective of the alternating current component of the envelope waveform when the separation occurs in a part of the bearing ring of the bearing device 360 and then the separation shifts to the entire region of the bearing ring. It is the figure which showed the time change of a value. In FIG. 13 and FIG. 14 described below, the temporal change of each effective value when the rotation speed of the main shaft 320 is constant is shown.

  Referring to FIG. 13, a curve L1 indicates a temporal change in the effective value of the vibration waveform that is not subjected to the envelope process, and a curve L2 indicates a temporal change in the effective value of the AC component of the envelope waveform. At time t1 before separation occurs, both the effective value (L1) of the vibration waveform and the effective value (L2) of the AC component of the envelope waveform are small. The vibration waveform at time t1 is as shown in the above-described FIG.

  When separation occurs in a part of the bearing ring of the bearing device 360, as described with reference to FIG. 11, the effective value (L2) of the alternating current component of the envelope waveform is greatly increased, while the vibration waveform not subjected to the envelope processing is increased. The effective value (L1) does not increase that much (near time t2).

  After that, when the separation shifts to the entire area of the raceway, as described with reference to FIG. 12, the effective value (L1) of the vibration waveform not subjected to the envelope processing is greatly increased, while the effective value of the AC component of the envelope waveform is increased. (L2) decreases (near time t3).

  FIG. 14 is a diagram showing temporal changes in the effective value of the vibration waveform of the bearing device 360 and the effective value of the AC component of the envelope waveform when surface roughness or poor lubrication of the bearing ring of the bearing device 360 occurs. is there. Referring to FIG. 14, similarly to FIG. 13, a curve L <b> 1 indicates a temporal change in the effective value of the vibration waveform that is not subjected to the envelope processing, and a curve L <b> 2 indicates a temporal change in the effective value of the AC component of the envelope waveform. Showing change.

  At time t11 before the raceway surface roughness or lubrication failure occurs, both the effective value (L1) of the vibration waveform and the effective value (L2) of the AC component of the envelope waveform are small. Note that the vibration waveform at time t11 is as shown in FIG.

  When the surface roughness or poor lubrication of the bearing ring of the bearing device 360 occurs, the effective value (L1) of the vibration waveform not subjected to the envelope processing increases as described with reference to FIG. There is no increase in the effective value (L2) (near time t12).

  As described above, the effective value of the vibration waveform of the bearing device 360 measured using the vibration sensor 370 and the AC component of the envelope waveform generated by performing envelope processing on the vibration waveform measured using the vibration sensor 370. By diagnosing the abnormality of the bearing device 360 based on the effective value of the above, it is possible to realize an accurate abnormality diagnosis as compared with the conventional frequency analysis technique.

  When performing such an abnormality diagnosis, it is necessary to appropriately set a vibration threshold value for performing the abnormality diagnosis. Such a threshold value is determined using the vibration analysis apparatus 100 described above. be able to. That is, the vibration analysis apparatus 100 can provide a model of the bearing apparatus 360 and also provide bearing damage data, and can predict a vibration waveform at a place where the vibration sensor 370 is installed in the bearing apparatus 360. Then, by calculating the effective value of the vibration waveform and the effective value of the AC component of the envelope waveform generated by performing envelope processing on the vibration waveform, the threshold value for abnormality determination in the state monitoring device is appropriately determined. be able to.

  Note that when the rotational speed of the main shaft 320 (FIG. 6) changes, the magnitude of vibration of the bearing device 360 changes. In general, the vibration of the bearing device 360 increases as the rotational speed of the main shaft 320 increases. Therefore, each of the effective value of the vibration waveform of the bearing device 360 and the effective value of the AC component of the envelope waveform is normalized by the rotation speed of the main shaft 320, and the abnormality diagnosis of the bearing device 360 is performed using each normalized effective value. You may make it perform.

  FIG. 15 is a functional block diagram functionally showing another configuration of the data processing apparatus. Referring to FIG. 15, data processing device 380A further includes a modified vibration degree calculation unit 430, a modified modulation degree calculation unit 470, and a speed function generation unit 500 in the configuration of data processing device 380 shown in FIG. Including.

  The speed function generation unit 500 receives a detection value of the rotation speed of the main shaft 320 by the rotation sensor 510 (not shown in FIG. 6). The rotation sensor 510 may output a detection value of the rotation position of the main shaft 320, and the speed function generation unit 500 may calculate the rotation speed of the main shaft 320. Then, the speed function generator 500 normalizes the effective value of the vibration waveform of the bearing device 360 calculated by the effective value calculator 420 with the rotational speed N of the main shaft 320, and the effective value. A speed function B (N) for normalizing the effective value of the alternating current component of the envelope waveform calculated by the calculation unit 460 with the rotational speed N of the main shaft 320 is generated. As an example, the speed functions A (N) and B (N) are expressed by the following equations.

A (N) = a × N ^−0.5 (1)
B (N) = b × N ^−0.5 (2)
Here, a and b are constants determined in advance by experiments or the like, and may be different values or the same values.

  The corrected vibration degree calculation unit 430 receives the effective value of the vibration waveform of the bearing device 360 from the effective value calculation unit 420 and receives the speed function A (N) from the speed function generation unit 500. Then, the corrected vibration degree calculation unit 430 uses the speed function A (N) to normalize the effective value of the vibration waveform calculated by the effective value calculation unit 420 with the rotation speed of the main shaft 320 (hereinafter referred to as “correction vibration”). Called "degree"). Specifically, using the effective value Vr of the vibration waveform calculated by the effective value calculation unit 420 and the speed function A (N), the corrected vibration degree Vr * is calculated by the following equation.

Here, Vra indicates an average value of Vr at time 0 to T.
Then, the corrected vibration degree calculation unit 430 outputs the corrected vibration degree Vr * calculated by Expression (3) to the storage unit 480.

  The modified modulation degree calculation unit 470 receives the effective value of the AC component of the envelope waveform from the effective value calculation unit 460 and receives the speed function B (N) from the speed function generation unit 500. Then, the modified modulation degree calculation unit 470 uses the speed function B (N) to normalize the effective value of the alternating current component of the envelope waveform calculated by the effective value calculation unit 460 with the rotation speed of the main shaft 320 (hereinafter, referred to as “the actual value”) (Referred to as “modified modulation degree”). Specifically, using the effective value Ve of the alternating current component of the envelope waveform and the speed function B (N) calculated by the effective value calculation unit 460, the modified modulation degree Ve * is calculated by the following equation.

  Here, Vea shows the average value of Ve in time 0-T. The modified modulation degree calculation unit 470 outputs the modified modulation degree Ve * calculated by Expression (4) to the storage unit 480.

  Then, the corrected vibration level Vr * and the corrected modulation degree Ve * stored in the storage unit 480 are read by the diagnosis unit 490 from time to time, and temporal changes in the read corrected vibration level Vr * and corrected modulation level Ve * are read. Based on this transition, the diagnosis unit 490 performs an abnormality diagnosis of the bearing device 360.

  In the above description, the rotation sensor 510 may be attached to the main shaft 320, or a bearing with a rotation sensor in which the rotation sensor 510 is incorporated in the bearing device 360 may be used for the bearing device 360.

  With such a configuration, the modified vibration degree Vr * obtained by normalizing the effective value of the vibration waveform of the bearing device 360 by the rotational speed, and the modified modulation degree Ve * obtained by normalizing the effective value of the AC component of the envelope waveform by the rotational speed. Therefore, a more accurate abnormality diagnosis is realized by removing the disturbance due to the fluctuation of the rotation speed.

  As described above, in the first embodiment, in the vibration analysis apparatus 100, damage data of the rolling bearing 20 to be analyzed is input, and damage is caused when the rotating shaft of the rolling bearing 20 is rotated by the dynamic analysis program. Thus, the displacement history between the inner and outer rings generated in the rolling bearing is calculated. Then, the history of the excitation force obtained by multiplying the calculated displacement between the inner and outer rings by the bearing spring constant is given to the vibration characteristic model of the bearing device 10 calculated by the mode analysis program. A vibration waveform at an arbitrary position (for example, an installation position of the vibration sensor) is calculated. Thereby, the vibration waveform of the bearing device 10 when damage occurs inside the bearing can be predicted in advance. Therefore, according to the first embodiment, using the above prediction result, for example, in the state monitoring device for the rolling bearing (bearing device 360) applied to the wind power generation facility 310, the vibration sensor 370 with respect to the bearing device 360 is used. The installation position can be easily and appropriately selected. Further, it is possible to appropriately determine a threshold value for determining the abnormality of the rolling bearing.

  As an example, let Vr * 0 and Ve * 0 be the modified vibration degree and the modified modulation degree measured in the initial normal state of the wind power generation facility. For the determination of the occurrence of peeling described with reference to FIG. 13, the rate of increase Ie (= Ve * / Ve * 0) from the initial state of the corrected modulation exceeded the threshold CeFlake for determining the peeling of the corrected modulation. Later, the occurrence of peeling is determined by confirming that the rate of increase Ir (= Vr * / Vr * 0) from the initial state of the corrected vibration exceeds the threshold CrFlake for determining the corrected vibration. Is done. For the determination of surface roughness and poor lubrication described with reference to FIG. 14, the increase rate of the correction vibration degree in a state where the increase rate Ie of the correction modulation degree does not exceed the threshold value CeSurf for determining the surface roughness of the correction modulation degree. By confirming that Ir has exceeded the threshold value CrSurf for determining the surface roughness of the corrected vibration, occurrence of surface roughness or poor lubrication is determined. In this case, there are four threshold values, CeFlake, CrFlake, CeSurf, and CrSurf.

  Note that the above threshold value and abnormality determination using the threshold value are merely examples, and more complicated pattern recognition may be used. Moreover, you may make it distinguish a surface roughness and poor lubrication by using a temperature sensor together. In any case, as described above, it is necessary to use a threshold value for performing abnormality determination in an increased vibration state.

  In the vibration analyzing apparatus 100, the history of the excitation force obtained by multiplying the displacement between the inner and outer rings caused by the given damage by the bearing spring constant is at least one point on the central axis of the inner ring 22 of the rolling bearing 20. Given to. As a result, a combination of the prediction of the displacement between the inner and outer rings of the rolling bearing 20 using the dynamic analysis program and the excitation force based thereon and the prediction of the vibration transfer characteristics of the bearing device 10 using the mode analysis program are combined. Vibration analysis can be performed.

[Modification]
In the above description, the mode analysis method is used for the transient response analysis when analyzing the vibration characteristics of the bearing device 10. However, a direct integration method may be used instead of the mode analysis method. The direct integration method is a method in which the calculated change in the approach amount between the rolling element and the raceway surface and the history of displacement between the inner and outer rings are given to the finite element model of the bearing device 10 for successive integration. This is effective when the vibration analysis apparatus 100 has sufficient arithmetic processing capability.

  FIG. 16 is a flowchart for explaining a processing procedure of a vibration analysis method executed by the vibration analysis apparatus 100 according to this modification. Referring to FIG. 16, this flowchart includes steps S90 to S94 instead of steps S50 to S80 in the flowchart shown in FIG.

  That is, when the history of displacement between the inner and outer rings generated in the rolling bearing 20 is calculated in step S40, the vibration analyzing apparatus 100 calculates the specification data of the bearing apparatus 10 (the shape and material density of the bearing apparatus 10, the Young's modulus, Based on the Poisson's ratio and the like), a finite element model of the bearing device 10 is calculated (step S90). Next, the vibration analysis device 100 reads various data for executing a transient response analysis (direct integration method) of the bearing device 10 (step S92). Specifically, the vibration analysis apparatus 100 reads the finite element model calculated in step S90, the change amount of the approach amount calculated in step S20, the displacement history between the inner and outer rings calculated in step S40, and the like. Include.

  Then, the vibration analysis device 100 calculates the vibration waveform of the bearing device 10 in accordance with a transient response analysis (direct integration method) program prepared in advance (step S94). Specifically, the vibration analysis apparatus 100 uses the finite element model calculated in step S90 to obtain the excitation force history obtained by multiplying the displacement (history) between the inner and outer rings calculated in step S40 by the bearing spring constant. By giving to at least one point on the central axis of the inner ring 22 of the bearing device 10 shown, a vibration waveform generated in the bearing device 10 is calculated based on the displacement history between the inner and outer rings calculated in step S40.

[Embodiment 2]
In the first embodiment and the modification thereof, the rolling bearing 20 is constituted by a ball bearing. In the second embodiment, a case where the rolling bearing 20 is constituted by a roller bearing will be described.

  When the rolling bearing 20 is constituted by a roller bearing, a displacement history between the inner and outer rings generated in the rolling bearing 20 is calculated by a dynamic analysis model using a so-called slice method. The slicing method is characterized in that the contact load is calculated for each minute width portion obtained by dividing the contact portion between the roller and the raceway surface into a minute width along the axial direction of the roller.

  FIG. 17 is a flowchart for explaining a processing procedure of a vibration analysis method executed by vibration analysis apparatus 100 according to the second embodiment. Referring to FIG. 17, first, vibration analysis apparatus 100 reads data relating to rolling element 24 and its orbit, and damage data applied to rolling bearing 20 from input unit 110 (step S110).

  Next, the vibration analysis apparatus 100 reads various data for executing the dynamic analysis (step S120). Specifically, the vibration analysis apparatus 100 reads from the input unit 110 data such as specification data of the rolling bearing 20, damage shape and operating conditions, and mass and spring characteristics of the rotating shaft 12 and the housing 30.

  When various data are read in step S120, the vibration analysis apparatus 100 changes from the damaged shape to the approach amount change amount between the roller and the raceway surface for each slice of the contact portion between the roller and the raceway surface. Conversion is performed (step S125). Here, the above conversion is performed using the load in the slice and the width of the damaged shape in the rolling direction as main variables. For this conversion, the effect of the width of the recess in the rolling direction on the relationship between the load and the amount of change in the approach amount in contact between the cylindrical objects may be examined in advance and converted into a function. Then, the vibration analyzing apparatus 100 operates the rotating shaft 12 according to the operation condition input in step S120 in accordance with the dynamic analysis program using the slice method in which the approach amount change amount in each slice is input. Then, a history of displacement between the inner and outer rings generated in the rolling bearing 20 due to the damage input in step S110 is calculated (step S130).

  The subsequent processes in steps S140 to S170 are basically the same as the processes in steps S50 to S80 shown in FIG. 4, and thus description thereof will not be repeated.

  As described above, according to the second embodiment, even when the rolling bearing 20 is constituted by a roller bearing, the vibration waveform of the bearing device 10 when the inside of the bearing is damaged can be predicted in advance. it can. As a result, it is possible to easily and appropriately select the installation position of the vibration sensor with respect to the bearing device in the rolling bearing state monitoring device applied to wind power generation facilities and the like. Furthermore, the abnormality determination threshold value of the rolling bearing can be appropriately determined.

[Embodiment 3]
In the third embodiment, the vibration analysis apparatus calculates up to a threshold value for determining the abnormality of the rolling bearing in the rolling bearing state monitoring apparatus. That is, the vibration analyzing apparatus shown in the third embodiment determines up to a vibration threshold value for abnormality determination in the state monitoring apparatus based on the predicted vibration waveform.

  FIG. 18 is a functional block diagram functionally showing the configuration of the vibration analyzing apparatus according to the third embodiment. In the following, a representative description will be given based on the first embodiment, but the same function can be added to the second embodiment.

  Referring to FIG. 18, vibration analysis device 100A further includes an abnormal threshold setting unit 260 and a base vibration input unit 270 in the configuration of vibration analysis device 100 according to the first embodiment shown in FIG.

  The base vibration input unit 270 generates a base vibration waveform indicating a vibration waveform when the rolling bearing 20 is normal. As the base vibration waveform, an actual measurement value of a bearing of the same type is preferable, but an expected value from a measurement value of the same type of device may be used. A vibration waveform assumed when an abnormality occurs is obtained by adding the vibration waveform received from the vibration waveform calculation unit 240 to the base vibration waveform.

  Abnormal threshold setting unit 260 receives from vibration waveform calculation unit 240 the calculated value of the vibration waveform at the place where the vibration sensor is attached in bearing device 10. Then, the abnormal threshold value setting unit 260 uses the vibration waveform received from the vibration waveform calculation unit 240 and the base vibration waveform received from the base vibration input unit 270 to determine the magnitude of vibration for determining the rolling bearing 20 as abnormal. Determine the threshold. As an example, the abnormal threshold setting unit 260 calculates the effective value of the vibration waveform and the effective value of the alternating current component of the envelope waveform from the data of the vibration waveform assumed when an abnormality occurs, and the calculated result Based on this, the threshold value is determined.

  Although not particularly illustrated, the abnormality threshold setting unit 260 gives damage data of various sizes to the displacement calculation unit 220 and calculates a vibration waveform for each damage data to determine a threshold for abnormality determination. The value may be determined.

  FIG. 19 is a flowchart for explaining a processing procedure of a vibration analysis method executed by vibration analysis apparatus 100A according to the third embodiment. Referring to FIG. 19, this flowchart further includes step S82 in the flowchart shown in FIG.

  In the third embodiment, in step S80, the vibration waveform at the place where the vibration sensor is attached in the bearing device 10 is calculated. When the vibration waveform is calculated in step S80, the vibration analysis apparatus 100A uses the calculated vibration waveform to determine the vibration magnitude threshold for determining that the rolling bearing 20 is abnormal in the state monitoring apparatus. A value is determined (step S82).

  Although not shown in particular, the threshold value of vibration for abnormality determination in the state monitoring device is determined based on the vibration waveform predicted by the vibration analysis device in the same manner as in the second embodiment. Can do.

  As described above, according to the third embodiment, in vibration analysis apparatus 100 (100A), it is possible to determine a threshold value for determining abnormality of a rolling bearing in a rolling bearing state monitoring apparatus.

  In each of the above embodiments, a dynamic analysis is used to calculate the history of displacement between the inner and outer rings generated in the rolling bearing, and the response analysis is performed by multiplying the displacement between the inner and outer rings by the bearing spring constant. The history of excitation force given to (vibration analysis) is calculated, but to simplify the calculation, the history of excitation force is calculated based on the calculation result of static contact analysis without using dynamic analysis. It may be calculated.

  Specifically, the amount of change in the amount of approach between the rolling element 24 calculated by contact analysis and the raceway surface is defined as the geometric shape collapse amount (for example, the amount of dent) of the raceway surface, and this shape collapse amount is taken into consideration. The amount of displacement between the inner and outer rings is calculated based on a static force balance analysis of the entire rolling bearing. The value obtained by multiplying the amount of displacement between the inner and outer rings by the bearing spring constant is taken as the maximum value of the excitation force generated in the rolling bearing, and this value is the excitation during the time that the rolling element passes through the damaged part. A waveform (history) of the excitation force is calculated as the maximum value of the force. Note that various shapes such as a sine wave shape, a triangular wave shape, a trapezoidal wave shape, and a rectangular wave shape can be adopted as the waveform shape.

  Further, in each of the above embodiments, as shown in FIG. 1, the outer ring 26 that is a fixed ring is a linear spring in the bearing radial direction at the position of the rolling element within the load range among the plurality of rolling elements 24. Although it is assumed that it is connected to the housing 30 via kF1 to kF3, it may be spring-connected to the housing 30 in the bearing radial direction at the center between the rolling elements in the load range.

  In the first and second embodiments and the third embodiment based thereon, the history of the excitation force obtained by multiplying the displacement between the inner and outer rings caused by the given damage by the bearing spring constant is: Although given to at least one point on the central axis of the inner ring 22 of the rolling bearing 20, the above-mentioned is applied to the rolling elements 24 in the load range according to the distribution of the force supported by the rolling elements 24 in the load range. A history of excitation force may be given.

[Other application examples of rolling bearing condition monitoring devices]
In each of the above embodiments, the state monitoring device for the main shaft bearing device 360 in the wind power generation facility 310 (FIG. 6) has been described, but the state monitoring device for the rolling bearing according to the present invention is the other in the wind power generation facility 310. The present invention can also be applied to a bearing device. For example, a plurality of bearings provided in the speed increaser 340 and rotatably supporting a plurality of shafts of the gear speed increasing mechanism (hereinafter also referred to as “speed increaser bearing device”), or in the generator 350. The state monitoring device for a rolling bearing according to the present invention can also be applied to a bearing that is provided and supports a rotor in a freely rotatable manner (hereinafter also referred to as a “generator bearing device”).

  Hereinafter, a method of selecting the installation position of the vibration sensor with respect to each bearing device when the rolling bearing state monitoring device according to the present invention is applied to the bearing device for the speed increaser and the bearing device for the generator will be described.

(Explanation of the selection method of the installation position of the vibration sensor in the bearing device for gearbox)
In the gear speed increasing mechanism that constitutes the speed increaser 340, due to variations in gear accuracy, gear assembly errors, fluctuations in tooth rigidity at the tooth meshing portions, and the like, meshing transmission errors occur in the pair of gears. . The meshing transmission error of the gear pair vibrates the meshing portion of the gear. And the excitation force which generate | occur | produces in the meshing part of a gearwheel is excited to a bearing apparatus by transmitting to a bearing apparatus. As described above, in the speed increasing bearing device, vibration (noise) is excited by the excitation force generated in the meshing portion of the gear, and therefore the detection sensitivity of the vibration sensor may be reduced due to the influence of noise.

  Therefore, when selecting the installation position of the vibration sensor, the speed increaser in which the magnitude of vibration generated at the installation position of the vibration sensor is calculated by the vibration analysis method by the same method as in the above-described main shaft bearing device. In addition to the vibration waveform of the bearing device for the motor, the calculation is performed in consideration of the influence of the vibration (noise) generated by the excitation force from the gear meshing portion on the detection sensitivity of the vibration sensor. Specifically, the SN ratio of the gearbox bearing device is calculated, and the installation position of the vibration sensor is selected based on the calculated SN ratio.

  FIG. 20 is a flowchart for explaining the processing procedure of the method for selecting the installation position of the vibration sensor with respect to the speed increaser bearing device. The flowchart shown in FIG. 20 realizes step S1200 (selection of sensor position) in the flowchart shown in FIG. 5 by the processing of steps S1240 to S1260.

  The flowchart shown in FIG. 20 can be realized by executing a program stored in advance in the data processing device 380 (FIG. 6). Alternatively, the vibration waveform data output from the data processing device 380 may be used to execute it outside the state monitoring device. In this embodiment, as an example, a configuration in which the data processing device 380 has the function of the sensor position selection unit 250 (FIG. 3) will be described.

  Referring to FIG. 20, first, data of a plurality of candidate positions that can be set as the installation positions of the vibration sensor with respect to the speed increasing gear bearing device are input from an input unit (not shown) of data processing device 380. (Step S1000).

  Next, when the data processing device 380 reads the data of a plurality of candidate positions from the input unit, the data processing device 380 executes the vibration analysis method, whereby each of the speed increasing device bearing devices when damage occurs inside the rolling bearing is obtained. A vibration waveform generated at the candidate position is calculated (step S1100).

  Subsequently, the data processing device 380 calculates a vibration waveform that is excited in the speed increasing device bearing device by the excitation force generated in the meshing portion of the gear constituting the gear speed increasing mechanism (step S1240). For example, the data processing device 380 calculates the meshing transmission error using the rigidity of the pair of gears, and calculates the excitation force generated in the meshing portion of the gear based on the calculated meshing transmission error. Then, the data processing device 380 calculates a vibration waveform generated in the speed increasing gear bearing device by the calculated excitation force by performing a response analysis using the vibration characteristic model of the bearing device.

  Next, the data processing device 380 uses the vibration waveform generated at each candidate position of the gearbox bearing device calculated in step S1100 and the vibration waveform of the gearbox bearing device calculated in step S1240. Then, the SN ratio at each candidate position is calculated (step S1250). Thereby, a plurality of SN ratios are calculated in correspondence with a plurality of candidate positions.

  The data processing device 380 extracts the one with the largest value from the plurality of S / N ratios. Then, the candidate position corresponding to the extracted SN ratio is selected as the installation position of the vibration sensor (step S1260).

  As described above, a candidate position that minimizes the influence of noise among the plurality of candidate positions is selected as the installation position of the vibration sensor. Thereby, the vibration sensor can detect the vibration of the bearing device for the gearbox when the rolling bearing is damaged with high sensitivity.

  In addition, since the installation position of the vibration sensor can be selected using the analysis result of the vibration analysis method and the response analysis result using the excitation force generated in the gear, the installation of the vibration sensor is compared with the conventional qualitative selection method. The position can be selected easily and appropriately.

(Description of the method for selecting the installation position of the vibration sensor in the generator bearing device)
In the wind power generation facility 310, the torque of the generator 350 fluctuates due to fluctuations in the rotation speed of the rotating unit including the blade 330. This fluctuation in the rotational speed is not only due to the change in the wind speed, but even when the wind speed is constant, the influence of the tower shadow effect that the speed is temporarily reduced by the blade 330 and the tower 400 crossing each other, This occurs due to wind sharing or the like caused by the difference in wind power at the rotational positions of the huge blades 330. And the torque fluctuation of the generator 350 generates an excitation force (for example, torsional vibration) in the rotation direction with respect to the coupling part that connects the generator 350 and the speed increaser 340. When the excitation force generated in the coupling portion is transmitted to the generator bearing device, vibration is excited in the generator bearing device. In this way, in the generator bearing device, vibration (noise) is excited by the excitation force generated in the coupling portion, so that the detection sensitivity of the vibration sensor may be reduced due to the influence of noise.

  Therefore, when selecting the installation position of the vibration sensor, the magnitude of vibration generated at the installation position of the vibration sensor is calculated by the vibration analysis method using the same method as in the case of the main shaft bearing device described above. In addition to the vibration waveform of the bearing device, calculation is performed in consideration of the influence of vibration (noise) generated by the excitation force from the coupling unit on the detection sensitivity of the vibration sensor. Specifically, the SN ratio of the generator bearing device is calculated, and the installation position of the vibration sensor is selected based on the calculated SN ratio.

  FIG. 21 is a flowchart for explaining the processing procedure of the method for selecting the installation position of the vibration sensor with respect to the generator bearing device. The flowchart shown in FIG. 21 realizes step S1200 (selection of sensor position) in the flowchart shown in FIG. 5 by the processing of steps S1270 to S1290.

  Note that the flowchart shown in FIG. 21 can be realized by executing a program stored in advance in the data processing device 380 (FIG. 6). Alternatively, the vibration waveform data output from the data processing device 380 may be used to execute it outside the state monitoring device. In this embodiment, as an example, a configuration in which the data processing device 380 has the function of the sensor position selection unit 250 (FIG. 3) will be described.

  Referring to FIG. 21, first, data of a plurality of candidate positions that can be set as vibration sensor installation positions with respect to the generator bearing device are input from an input unit (not shown) of data processing device 380 ( Step S1000).

  Next, when the data processing device 380 reads data at a plurality of candidate positions from the input unit, the data processing device 380 executes the vibration analysis method, thereby causing each candidate for the generator bearing device when damage occurs inside the rolling bearing. A vibration waveform generated at the position is calculated (step S1100).

  Subsequently, the data processing device 380 calculates a vibration waveform that is excited in the generator bearing device by the excitation force generated in the coupling portion (step S1270). For example, the data processing device 380 calculates the excitation force that acts in the rotation direction of the coupling unit due to the torque fluctuation of the generator 350. Then, the data processing device 380 calculates a vibration waveform generated in the generator bearing device by the calculated excitation force by performing a response analysis using a vibration characteristic model of the bearing device.

  Next, the data processing device 380 uses the vibration waveform generated at each candidate position of the generator bearing device calculated in step S1100 and the vibration waveform of the generator bearing device calculated in step S1270 to The SN ratio at the candidate position is calculated (step S1280). Thereby, a plurality of SN ratios are calculated in correspondence with a plurality of candidate positions.

  The data processing device 380 extracts the one with the largest value from the plurality of S / N ratios. Then, a candidate position corresponding to the extracted SN ratio is selected as the installation position of the vibration sensor (step S1290).

  As described above, a candidate position that minimizes the influence of noise among the plurality of candidate positions is selected as the installation position of the vibration sensor. Thereby, the vibration sensor can detect the vibration of the generator bearing device when the rolling bearing is damaged with high sensitivity.

  In addition, since the installation position of the vibration sensor can be selected using the analysis result of the vibration analysis method and the response analysis result using the excitation force generated in the gear, the installation of the vibration sensor is compared with the conventional qualitative selection method. The position can be selected easily and appropriately.

  In the above, the method for setting the installation position of the vibration sensor using the analysis result of the vibration analysis device is applied to the state monitoring device of the rolling bearing in the wind power generation facility, and the state monitoring of the rolling bearing in the railway vehicle. It is also possible to apply to an apparatus or the like.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description of the embodiments but by the claims, and is intended to include all modifications within the scope.

  DESCRIPTION OF SYMBOLS 10 Bearing apparatus, 12 Rotating shaft, 20 Rolling bearing, 22 Inner ring, 24 Rolling element, 26 Outer ring, 30 Housing, 40 Base, 100, 100A Vibration analysis apparatus, 110 Input part, 120 I / F part, 130 CPU, 140 RAM , 150 ROM, 160 output unit, 205 approach amount change amount calculation unit, 210, 210A dynamic analysis model setting unit, 220 displacement calculation unit, 230 vibration characteristic calculation unit, 240, 240A vibration waveform calculation unit, 250 sensor position selection unit 260, abnormal threshold setting unit, 270 base vibration input unit, 310 wind power generation equipment, 320 main shaft, 330 blade, 340 speed increaser, 350 generator, 360 bearing, 370 vibration sensor, 380, 380A data processing device, 390 Nacelle, 400 tower, 410, 450 HPF, 4 0,460 effective value calculation unit, 430 fix the vibrating calculator, 440 envelope processing unit, 470 modify the modulation degree calculation unit, 480 memory unit, 490 diagnostic unit, 500 speed function generator, 510 rotation sensor.

Claims (8)

A vibration sensor for measuring vibration of a rolling bearing and a bearing device including the housing thereof;
A determination unit that determines that the rolling bearing is abnormal when the magnitude of vibration measured using the vibration sensor exceeds a predetermined threshold;
The installation position of the vibration sensor is selected using a vibration waveform calculated by a vibration analysis method for analyzing vibration of the bearing device by a computer,
The vibration analysis method includes:
Inputting shape data of damage to the contact portion between the rolling element and the raceway of the rolling bearing;
Calculating a displacement between an inner ring and an outer ring of the rolling bearing, which occurs in the rolling bearing due to the damage;
A step of calculating a vibration characteristic model indicating a vibration characteristic of the bearing device by a mode analysis program for analyzing a vibration mode of the bearing device;
By giving the vibration characteristic model a history of excitation force generated in the rolling bearing obtained by multiplying the displacement calculated by the step of calculating the displacement by a spring constant between the inner ring and the outer ring. And a step of calculating a vibration waveform at an arbitrary position of the bearing device.   The step of calculating the displacement includes a step of calculating a history of the displacement generated in the rolling bearing due to the damage when a rotating shaft of the rolling bearing is rotated by a dynamic analysis program for performing a dynamic analysis of the rolling bearing. The rolling bearing state monitoring device according to claim 1, comprising: By using the vibration analysis method, a plurality of vibration waveforms are calculated corresponding to a plurality of candidate positions that can be set as the installation position of the vibration sensor,
The rolling bearing state monitoring device according to claim 1 or 2, wherein a candidate position having a maximum vibration acceleration amplitude among the plurality of candidate positions is selected as an installation position of the vibration sensor. The bearing device includes a bearing device provided on a main shaft in a wind power generation facility,
For each of the plurality of candidate positions, a ratio of the vibration acceleration amplitude calculated by the vibration analysis method to the vibration acceleration amplitude excited in the bearing device by the excitation force generated in the main shaft is calculated,
The rolling bearing state monitoring device according to claim 3, wherein a candidate position having the maximum ratio among the plurality of candidate positions is selected as an installation position of the vibration sensor. The bearing device includes a bearing device provided in a gearbox in a wind power generation facility,
For each of the plurality of candidate positions, the ratio of the vibration acceleration amplitude calculated by the vibration analysis method to the vibration acceleration amplitude excited in the bearing device by the excitation force generated in the gear of the gearbox is calculated. And
The rolling bearing state monitoring device according to claim 3, wherein a candidate position having the maximum ratio among the plurality of candidate positions is selected as an installation position of the vibration sensor. The bearing device includes a bearing device provided in a generator in a wind power generation facility,
The generator is connected to a speed increaser in the wind power generation facility by a coupling unit,
For each of the plurality of candidate positions, the ratio of the vibration acceleration amplitude calculated by the vibration analysis method to the vibration acceleration amplitude excited in the bearing device by the excitation force generated in the coupling portion is calculated,
The rolling bearing state monitoring device according to claim 3, wherein a candidate position having the maximum ratio among the plurality of candidate positions is selected as an installation position of the vibration sensor.   The rolling bearing state monitoring device according to any one of claims 1 to 6, further comprising a selection unit that selects an installation position of the vibration sensor using a vibration waveform calculated by the vibration analysis method.   The predetermined threshold value is determined using a vibration waveform assumed when an abnormality occurs in the rolling bearing at the selected installation position of the vibration sensor. The rolling bearing state monitoring device described.

Images