Estimation device for bearing life, method for estimating bearing device life, and method for extending bearing device life

Abstract

To provide an estimation device and the like for estimating bearing life that can estimate small damage to a bearing device.SOLUTION: An estimation device 100 includes an interface 106, a memory 104, and a calculation device. The memory 104 stores range information related to a plurality of first ranges of a first physical quantity. The calculation device identifies, for each of the plurality of first ranges, a time parameter related to a time during which the first physical quantity falls within the first range. Then, the calculation device estimates an occurrence time of surface damage of a bearing device based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to the surface stress of the bearing device for each of the plurality of first ranges.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a bearing life estimation device, a bearing device life estimation method, and a bearing device life extension method. [Background technology]

[0002] For example, JP 2021-12185 A (Patent Document 1) discloses a method for diagnosing the lifespan of a bearing device. The invention disclosed in Patent Document 1 makes it possible for a manager of the bearing device to recognize the lifespan of the bearing device. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-12185 Summary of the Invention [Problem to be solved by the invention]

[0004] Generally, when a bearing device is severely damaged, an operator must replace the bearing device. However, replacing a bearing device is very costly. In view of this problem, it is preferable to estimate the time when minor damage to the bearing device will occur. Such prediction can prevent major damage to the bearing device, and as a result, the maintenance costs of the bearing device can be reduced.

[0005] The present disclosure has been made to solve the above-mentioned problems, and has an object to provide an estimation device and an estimation method for estimating small-scale damage to a bearing device. [Means for solving the problem]

[0006] The estimation device of the present disclosure includes an interface, a memory, and a calculation device, and the memory stores range information related to a plurality of first ranges of a first physical quantity. The calculation device identifies, for each of the plurality of first ranges, a time parameter related to a time during which the first physical quantity falls within the first range. The calculation device then estimates the occurrence time of surface damage of the bearing device based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to the surface stress of the bearing device for each of the plurality of first ranges.

[0007] The estimation method of the present disclosure includes acquiring a first physical quantity from a first sensor that detects the first physical quantity of the bearing device. The estimation method includes identifying, for each of a plurality of first ranges of the first physical quantity, a time parameter related to a time during which the first physical quantity falls within the first range. The estimation method includes estimating an occurrence time of surface damage of the bearing device based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to a surface stress of the bearing device for each of the plurality of first ranges. Effect of the Invention

[0008] According to the present disclosure, small-scale damage to a bearing device can be estimated. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of a management system according to the present disclosure. [Diagram 2] FIG. 2 is a functional block diagram of the estimation device. [Diagram 3] FIG. 13 is a diagram for explaining a surface layer stress. [Figure 4] FIG. 13 is a diagram showing a method for estimating surface layer stress. [Diagram 5] FIG. 1 is a diagram showing an example of an SN curve. [Figure 6] FIG. 13 is a diagram showing an example of a history of surface layer stress. [Figure 7] FIG. 13 is a diagram showing an example of a precursor range and a non-precursor range. [Figure 8] 4 is a flowchart showing a flow of processing performed by the estimation device of the first embodiment. [Figure 9]1 is a flowchart of a method for determining a precursor range and a non-precursor range. [Figure 10] FIG. 2 is a schematic diagram showing the configuration of a two-cylinder testing machine. [Figure 11] FIG. 1 is a diagram showing the operating conditions of the equipment of a two-cylinder testing machine. [Figure 12] FIG. 13 is a diagram showing the characteristics of a driving side test piece D2. [Figure 13] FIG. 13 is a diagram showing the characteristics of the driven side test piece F2. [Figure 14] FIG. 13 is a diagram for explaining the "embodiment example." [Figure 15] FIG. 13 is a diagram showing experimental results. [Figure 16] 10 is a flowchart showing a process of an estimation device according to a second embodiment. [Figure 17] FIG. 4 is a diagram showing an example of a history of surface layer stress estimated by the estimation device of the first embodiment. [Figure 18] FIG. 4 is a diagram showing an example of a history of surface layer stress estimated by the estimation device of the first embodiment. [Figure 19] FIG. 4 is a diagram showing an example of a display screen. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, the embodiments of the present disclosure will be described in detail with reference to the drawings. In the embodiments described below, when numbers, amounts, etc. are mentioned, the scope of the present disclosure is not necessarily limited to the numbers, amounts, etc., unless otherwise specified. The same reference numbers are given to the same parts and corresponding parts, and duplicate descriptions may not be repeated. It is intended from the beginning that the configurations in the embodiments will be used in appropriate combination.

[0011] First Embodiment

[0012] 1 is a diagram showing an example of the configuration of a management system 10 according to the present disclosure. The management system 10 according to the present disclosure includes at least one wind power generation unit 60, an estimation device 100, and a creation device 12. The collection device 30, the estimation device 100, a user terminal 50 (described later), and the creation device 12 are capable of communicating with each other via a network NW. Note that in the present disclosure, the processing of the estimation device 100 includes both the processing being executed by the estimation device 100 itself and the processing being shared and executed by at least one information processing device (not shown).

[0013] The wind power generation unit 60 includes a wind power generation device 20, a speed sensor 41, a temperature sensor 42, a collection device 30, and a user terminal 50. The wind power generation device 20 is a device that receives wind power and generates power. The wind power generation device 20 includes a bearing device 25, a generator, and the like. The bearing device 25 includes a rotating shaft and bearings, and the like. The speed sensor 41 detects the rotation speed of the rotating shaft of the bearing device 25. The temperature sensor 42 detects the temperature (bearing temperature) of the bearing device 25. The rotation speed of the rotating shaft detected by the speed sensor 41 is transmitted to the collection device 30 as time series data. The bearing temperature of the bearing device 25 detected by the temperature sensor 42 is transmitted to the collection device 30 as time series data. Hereinafter, the time series data of the rotation speed and the time series data of the bearing temperature may be collectively referred to as "time series data".

[0014] The speed sensor 41 corresponds to a "first sensor" in this disclosure. The rotational speed of the rotating shaft detected by the speed sensor 41 corresponds to a "first physical quantity" in this disclosure. The temperature sensor 42 corresponds to a "second sensor" in this disclosure. The bearing temperature of the bearing device 25 detected by the temperature sensor 42 corresponds to a "second physical quantity" in this disclosure.

[0015] The collection device 30 transmits the time series data (the rotation speed from the speed sensor 41 and the bearing temperature from the temperature sensor 42) to the estimation device 100 via the network NW.

[0016] The user terminal 50 is a terminal device owned by the user A. The "user" typically refers to a person who owns the wind power generation device 20. Furthermore, the user terminal 50 is typically a portable terminal that the user A can carry with him or her. Furthermore, the user terminal 50 may be a dedicated computer terminal.

[0017] For example, at least one wind power generation device 20 included in the management system 10 is assigned a wind power generation device ID (identification). At least one user terminal 50 included in the management system 10 is assigned a user terminal ID. The wind power generation device ID and the user terminal ID are associated with each other. The estimation device 100 holds a table (not shown) showing this association. In a wind power generation device ID and a user terminal ID that correspond to each other, the wind power generation device 20 indicated by the wind power generation device ID corresponds to the user terminal 50 indicated by the user terminal ID.

[0018] The estimation device 100 includes a CPU 102, a memory 104, and an interface 106. The CPU 102 executes various processes. The CPU 102 corresponds to the "arithmetic device" of the present disclosure.

[0019] The memory 104 includes a read only memory (ROM) and a random access memory (RAM), etc. The ROM is a non-volatile memory that cannot be rewritten, and the RAM is a volatile memory.

[0020] The ROM stores a program describing the processing procedures of the CPU 102. The CPU 102 loads the program stored in the ROM into a RAM or the like and executes it. The interface 106 communicates with an external device (the collection device 30 or the user terminal 50) via the network NW.

[0021] The estimation device 100 estimates the occurrence time of surface damage of the bearing device 25 based on time series data. "Surface damage of the bearing device 25" typically refers to damage to the surface of a component constituting the bearing device 25. The components constituting the bearing device 25 include, for example, a rotating shaft and a bearing of the rotating shaft. For example, a first calculation formula 131 is used as a method for estimating the occurrence time of surface damage of the bearing device 25. A second calculation formula 132 will be described in the second embodiment. Note that, in the present disclosure, an example of the bearing device 25 to be estimated is described as being applied to a wind power generation device 20, but the bearing device 25 to be estimated may be applied to another device.

[0022] The first calculation formula 131 is a formula into which at least the rotation speed and the bearing temperature are input. The first calculation formula 131 is a formula in which the smaller the rotation speed, the larger the surface stress that is output, and the higher the bearing temperature, the larger the surface stress (described later). The details of the first calculation formula 131 will be described later.

[0023] The display 120 is connected to the estimation device 100. The estimation device 100 displays an image showing the estimation result of the estimation device 100 on the display 120. In this way, an administrator who visually checks the estimation result displayed on the display 120 can recognize the estimation result of the estimation device 100.

[0024] Furthermore, the estimation device 100 may transmit the estimation result to a user terminal 50 corresponding to the wind power generation device 20 from which the estimation result was derived. This allows a user who visually checks the estimation result displayed on the user terminal 50 to recognize the estimation result of the estimation device 100. The estimation result is, for example, information including the occurrence time (year, month, and date) of the small-scale damage. Furthermore, the estimation device 100 may transmit a maintenance technique, which will be described later, to the user terminal 50 together with the occurrence time.

[0025] [Functional block diagram of the estimation device]

[0026] 2 is a functional block diagram of the estimation device 100. The estimation device 100 includes an acquisition unit 112, an estimation unit 114, and a storage unit 116. The acquisition unit 112 corresponds to the above-mentioned interface 106. The estimation unit 114 corresponds to the above-mentioned CPU 102. The storage unit 116 corresponds to the above-mentioned memory 104.

[0027] The acquisition unit 112 of the estimation device 100 acquires the time series data (time series data of bearing temperature and time series data of rotation speed) from the collection device 30. The time series data acquired by the acquisition unit 112 is output to the estimation unit 114.

[0028] The storage unit 116 stores range information 121. As described below, the range information 121 includes information on a plurality of ranges (first ranges) of the rotation speed (first physical quantity) from the speed sensor 41 and information on a plurality of ranges (second ranges) of the bearing temperature (second physical quantity) from the temperature sensor 42.

[0029] The estimation unit 114 estimates the surface layer stress of the bearing device 25. Then, the estimation unit 114 estimates the occurrence time of surface damage of the bearing device based on the surface layer stress. The surface layer stress will be described later with reference to FIG.

[0030] Here, the inventors have discovered that damage to the bearing device 25 of the wind turbine generator 20 occurs when small-scale surface damage first occurs at the rolling contact portion, and then the surface damage can lead to large-scale damage such as spalling. This small-scale surface damage includes at least one of micropitting and smearing.

[0031] Micropitting is typically an aggregate of minute peelings on the order of microns. Smearing is typically a phenomenon in which a plurality of minute seizures occur. In this embodiment, an example in which the small-scale damage is mainly micropitting will be described. Small-scale damage such as micropitting tends to occur less easily as the surface stress decreases.

[0032] The occurrence of micropitting itself tends not to cause much of a hindrance to the stable operation of the wind turbine generator 20. However, there are cases where micropitting develops into large-scale damage. If large-scale damage occurs to the bearing device 25, the workers of the wind turbine generator 20 or the like must replace the bearing device 25. Replacing the bearing device 25 is very costly.

[0033] In consideration of this problem, the estimation device 100 of this embodiment estimates the occurrence time of micropitting in the bearing device 25. For example, if the estimated occurrence time of micropitting is shorter than the service life of the wind power generation device 20, the user A or an operator can repair the small-scale damage by a maintenance process described below. By this maintenance process, the estimation device 100 can suppress the occurrence of large-scale damage in the bearing device 25. Therefore, the user A (see FIG. 1) and the like can limit the maintenance of the bearing device 25 to small-scale maintenance, and can reduce maintenance costs. In the present disclosure, the period from the present time to the estimated occurrence time of small-scale damage is also referred to as the "lifespan."

[0034] The inventors have discovered that an appropriate parameter for predicting the occurrence time of small-scale surface damage of the bearing device 25 is the surface layer stress of the bearing device 25. The estimation device 100 of this embodiment uses a surface layer stress parameter to estimate the occurrence time of small-scale surface damage of the bearing device 25. The surface layer stress parameter may be the surface layer stress itself, or may be a value correlated with the surface layer stress. In this embodiment, the surface layer stress parameter is the surface layer stress itself.

[0035] Fig. 3 is a diagram for explaining surface layer stress. Fig. 3 shows a part of bearing device 25. In the example of Fig. 3, bearing device 25 includes surface 25A, surface layer 25B, and inner portion 25C of bearing device 25. Surface layer 25B is located between surface 25A and inner portion 25C. Surface layer stress refers to stress within surface layer 25B. Surface 25A in bearing device 25 is typically a rolling contact portion between a first member and a second member. For example, surface 25A may be a rolling contact portion between a raceway surface of a raceway ring (first member) and a rolling surface of a rolling element (second member). Surface 25A may also be a sliding portion between a flange portion of the raceway ring (first member) and an end face of the rolling element (second member).

[0036] Fig. 4 is a diagram for explaining a method of estimating a surface layer stress by the estimating unit 114. In Fig. 4, a bearing temperature range and a rotation speed range are shown as range information 121.

[0037] In the example of FIG. 4, from the left, the range classification, range ID, bearing temperature range, rotational speed range, operation time ratio (%), surface stress, period, and fatigue level are specified.

[0038] The range classification includes a precursor range and a non-precursor range. The precursor range is a predetermined range related to the bearing temperature and rotation speed of the bearing device 25. When the bearing temperature and rotation speed of the bearing device 25 belong to the precursor range, the estimation device 100 judges that there is a possibility (precursor) that small-scale damage will occur. The non-precursor range is a range other than the precursor range. When the bearing temperature and rotation speed of the bearing device 25 belong to the non-precursor range, the estimation device 100 judges that there is no possibility that small-scale damage will occur. The precursor range and non-precursor range are, for example, defined by the creation device 12 (see FIG. 1) before the operation of the wind power generation device 20.

[0039] The range ID is an ID given to each range. The bearing temperature range is a range related to the bearing temperature of the bearing device 25. The bearing temperature range is made up of N ranges (N is an integer equal to or greater than 2). In the example of FIG. 4, the bearing temperature range is made up of four ranges. That is, N=4. In the example of FIG. 4, one range is, for example, equal to or greater than 55 degrees and less than 60 degrees. Each of the N ranges corresponds to a "second range" in the present disclosure.

[0040] The rotational speed range includes a plurality of ranges related to the rotational speed of the bearing device 25. In the example of Fig. 4, the rotational speed range is a range obtained by further dividing the second range into at least one range. The divided range corresponds to the "first range" of the present disclosure. That is, in the range information, each of the plurality of second ranges is composed of at least one first range.

[0041] In the example of Fig. 4, the second range in which the bearing temperature is equal to or greater than 55 degrees and less than 60 degrees is made up of four first ranges. One of the four first ranges is a range in which the rotation speed is equal to or greater than 3 rpm and less than 4 rpm. Also, in the example of Fig. 4, the second range in which the bearing temperature is less than 45 degrees is made up of one first range. The one first range is a range in which the rotation speed is equal to or greater than 0 rpm and less than 1 rpm.

[0042] In the example of FIG. 4, the precursor ranges are divided into ten ranges for the rotation speed and bearing temperature of the bearing device 25.

[0043] Furthermore, in the example of FIG. 4, the above non-precursor ranges are specified. That is, in the example of FIG. 4, the range information 121 is information indicating 11 ranges. In the example of FIG. 4, the above 10 ranges are each assigned a range ID of M1 to M10. Furthermore, one non-precursor range is assigned a range ID of M0. Hereinafter, the range IDs may be used as reference codes for the 11 ranges. For example, a range with a range ID of M1 is also referred to as a "range M1."

[0044] When the bearing temperature and rotation speed from the collection device 30 fall within the precursor range, the estimation unit 114 executes a sorting process to sort the time-series data into one of the 10 precursor ranges while monitoring the data in real time over a predetermined monitoring period. This monitoring period is the period from when the bearing temperature and rotation speed start to fall within the precursor range until they move to a non-precursor period. In the bearing device 25 of the wind power generation device 20, the rotation speed and bearing temperature change irregularly depending on the natural wind conditions, so the monitoring period changes from time to time.

[0045] Then, when the sorting process of the time series data during the monitoring period is completed, the estimation unit 114 identifies, for each of the eleven ranges (first ranges), a parameter (time parameter) relating to the time during which the bearing temperature and the rotation speed belong to that range. Here, the time parameter in this embodiment is the operation time ratio (%) for each first range. Fig. 4 shows a state in which the identification of the time parameter has been completed.

[0046] In the example of Fig. 4, the estimation unit 114 identifies the integrated value of the operation time in which the bearing temperature and the rotation speed belong to each of 11 ranges. Then, the estimation unit 114 divides the integrated value of the operation time for each of the 11 ranges by the total operation time of the wind power generation device 20. In this way, the estimation unit 114 can calculate the operation time ratio for each of the 11 ranges. Note that the time parameter may be the operation time itself.

[0047] Next, the estimation unit 114 estimates the surface stress for each of the 10 ranges. In the example of Fig. 4, for example, the surface stress for the range M1 is F1. Also, for the non-precursor range, the surface stress is defined as "0".

[0048] When the total operating time of the wind power generation equipment exceeds a predetermined time, the estimation unit 114 estimates the time when small-scale damage will occur based on the operating time ratio in the 10 first ranges, the surface stress in the first ranges, and specified estimation information.

[0049] The estimated information in this embodiment is an SN curve. FIG. 5 is a diagram showing an example of an SN curve. The vertical axis indicates the surface stress F, and the horizontal axis indicates the life L. The life L is a value corresponding to the occurrence time of small-scale damage when a specific surface stress is repeatedly applied. The estimation unit 114 applies the idea of ​​Miner's rule and collates the operation time ratio in the above 10 first ranges and the surface stress in the first range with the SN curve to identify the occurrence time of small-scale damage. For example, as shown in FIG. 5, the life L when a certain surface stress F1 is repeatedly applied is uniquely determined from the SN curve as L=g(F1). Note that g() is a function indicating the shape of the Miner's rule. In Miner's rule, it is considered that the life is consumed by 1 / L (=1 / g(F1)) each time the surface stress F1 is applied. Therefore, when the surface stress F1 is repeatedly applied N times, the life is consumed by N / L. In this embodiment, the consumption of this life span is also referred to as "fatigue level" (see also FIG. 4). Based on this idea, the estimation unit 114 calculates the fatigue level P within the period corresponding to each of the 10 first ranges, and calculates the total operation time of the wind power generation device 20 at the time when the sum Pm (see FIG. 4) of the fatigue levels of each first range reaches "1" as the life span.

[0050] For example, if the number of loads in the period T1 of the range ID M1 in FIG. 4 is N1, the fatigue level in the period T1 is N1 / g(F1). Here, N1 is calculated from the rotation speed, the operation time ratio, and the bearing dimensions in the period T1. The estimation unit 114 similarly calculates the fatigue level for each period, and if the total fatigue level Pm is 1 or more, it determines that small-scale damage has already occurred in the bearing device 25. If the total fatigue level Pm is less than 1, it is assumed that the bearing device 25 will be operated with the same operation time ratio from now on. In other words, the estimation unit 114 calculates how many times the total operation time of the wind power generation device 20 from the start of operation to the present time will be required for small-scale damage to occur. In the example of FIG. 4, the estimation unit 114 estimates that the total fatigue level Pm is 0.2 (i.e., 20%), and that small-scale damage will occur when the total operation time of the wind power generation device 20 corresponds to 5 times the current total operation time.

[0051] The method for creating the SN curve is disclosed, for example, in "N. HASEGAWA, T. FUJITA, M. UCHIDATE, M. ABO, H. KINOSHITA: Estimation Method of Micropitting Life from SN Curve Established by Residual Stress Measurements and Numerical Contact Analysis, Tribol. Online, 14,3 (2019) 131".

[0052] In this procedure, the worker first performs rolling fatigue tests under various operating conditions. Then, for each test, the worker obtains the time when small-scale damage occurred and the stress history acting on the contact part (information on when, how much stress, and how many times the load was applied). The worker also interrupts each test at a predetermined number of loads. The worker then estimates the surface stress at the time of the load count from the results of contact analysis and residual stress measurement using the surface shape measured at the time of the interruption. For example, equation (3) described below is used to estimate this surface stress. This procedure determines the stress history, which is the history of fluctuations in the surface stress during the test. The worker then performs regression analysis using the data set of the acquired stress history of each test and the time when small-scale damage occurred, to obtain an SN curve.

[0053] Fig. 6 is a diagram showing the surface layer stresses F0 to F10 and the periods T0 to T10 shown in Fig. 4 as a history. In the example of Fig. 6, the vertical axis indicates the surface layer stress, and the horizontal axis indicates the time. In the example of Fig. 4, the ranges M0 to M3 are shown. The estimation unit 114 may generate a history as shown in Fig. 6, and identify the occurrence time of small-scale damage on the premise that the history is repeated.

[0054] Fig. 7 is a diagram showing an example of the precursor range and non-precursor range described in Fig. 4. In the example of Fig. 7, the vertical axis indicates the rotation speed, and the horizontal axis indicates the bearing temperature. The boundary between the precursor range and the non-precursor range is set as a threshold value. This boundary is determined by the creation device 12 as described below.

[0055] [About the first calculation formula]

[0056] Next, the specific content of the first calculation formula 131, which is an estimation formula for the surface layer stress, will be described. In the surface layer stress of this embodiment, a value equivalent to the average value of the Mises stress acting at a specific depth zs point directly below the real contact part of the bearing device 25 is used. This idea is based on the knowledge that initial cracks of small-scale damage (micropitting) occur due to plastic deformation of the real contact part. This knowledge is disclosed in documents such as "Hasegawa, Fujita, Uchidate, and Aho: Mechanism of peeling due to rolling contact and effect of black oxide treatment on peeling suppression, 1st report, Tribologist, 63, 8, (2018) 551".

[0057] The depth zs should be the depth at which the von Mises stress is maximum just below the real contact area. When actually calculating the surface stress, first calculate the triaxial stress component P acting at the depth zs just below the contact area of ​​multiple surface roughness protrusions. j and Q jk is calculated using the following equations (1) and (2). P j =P j、con +P j、res (1) Q jk =Q jk、con +Q jk、res (2)

[0058] Here, j = x, y, z, k = x, y, z, j ≠ k, and Q jk =Q kj The subscript "con" in formulas (1) and (2) indicates the contact stress obtained by contact stress analysis, and the subscript "res" indicates the residual stress obtained by residual stress measurement. Additionally, x, y, and z represent the circumferential direction, axial direction, and depth direction of the raceway surface, respectively. A method for calculating the triaxial stress components of contact stress is disclosed, for example, in JP 2021-012185 A. The triaxial stress components of residual stress are obtained in advance, for example, by an X-ray residual stress measuring device.

[0059] The von Mises stress R at a specific depth zs directly below each real contact is calculated by the triaxial stress components P calculated using equations (1) and (2). j and Q jk is calculated by substituting into the following equation (3).

number

[0060] The surface stress F is calculated by the average value of the von Mises stress R at a specific depth zs directly below each real contact point calculated by equation (3), as shown in the following equation (4).

number

[0061] This formula (4) corresponds to the first calculation formula 131. In the present embodiment, the residual stress P j、res、 and Q jk、res Regarding the stress, for example, a value measured by an X-ray residual stress measuring device for the raceway surface of the bearing device 25 at the shipping stage is used.

[0062] The contact stress P in equations (1) and (2) j、con and Q jk、con The following parameters are used in the contact stress analysis to obtain the abovementioned: the bearing temperature T when the bearing is in the precursor region, the rotation speed V, the load A, the bearing dimensions B (bearing dimension specifications), the surface shape C of the raceway surfaces and rolling surfaces, the density D of the lubricant in the bearing device 25, and the kinetic viscosity E of the lubricant.

[0063] The rotation speed V is detected by a speed sensor 41. The bearing temperature T is detected by a temperature sensor .

[0064] The load A is calculated from a relational expression that shows the relationship between the load A and the rotation speed V. If such a relational expression does not exist, the maximum load that can occur during the service life of the wind turbine generator 20 or the load that is most likely to occur during the service life can be used as the load A. If it is desired to predict the micropitting life to be on the safe side (the predicted life is shorter than the actual life), the maximum load that can occur during the service life of the wind turbine generator can be used.

[0065] A previously measured value is used for the bearing dimension B. A published value of the density of the lubricant is used for the density D of the lubricant. Also, a published value of the kinetic viscosity of the lubricant is used for the kinetic viscosity E of the lubricant. In this embodiment, the three-dimensional surface shape C is a shape measured with a laser microscope or the like for the bearing device 25 at the shipping stage.

[0066] Furthermore, as described above, to calculate the surface stress F, the residual stress H (the above residual stress P j、res、 and Q jk、res That is, the first calculation formula 131 can also be expressed by the following formula (5).

[0067] F=G(V, T, A, B, C, D, E, H) (5)

[0068] Function G is a function that takes as arguments the bearing temperature T, the rotational speed V, the load A, the bearing dimension B, the surface shape C, the lubricant density D, the lubricant kinetic viscosity E, and the residual stress H. As described above, the load A, the bearing dimension B, the surface shape C, the lubricant density D, the lubricant kinetic viscosity E, and the residual stress H are values ​​that have been measured or calculated in advance.

[0069] [flowchart]

[0070] Fig. 8 is a flowchart showing the flow of processing by the estimation device 100. First, in step S2, the estimation device 100 determines whether or not the bearing temperature and rotation speed acquired from the collection device 30 belong to the precursor region (see Fig. 7). In step S2, if the bearing temperature and rotation speed do not belong to the precursor region (NO in step S2), the processing in Fig. 8 ends.

[0071] If the time series data values ​​(bearing temperature and rotation speed) belong to the precursor region in step S2 (YES in step S2), the estimation device 100 starts monitoring the bearing temperature and rotation speed in step S4. Then, in step S6, the estimation device 100 determines whether the monitoring period has elapsed. If the monitoring period has not elapsed (NO in step S6), the process proceeds to step S8. In step S8, the time series data is acquired and sorted into 10 precursor ranges (see FIG. 4).

[0072] In step S6, if the monitoring period has elapsed (YES in step S6), in step S10, the estimation device 100 calculates the operation time ratio for each of the 10 precursor ranges (see FIG. 4). Next, in step S12, the estimation device 100 judges whether or not the total operation time of the wind power generation device has exceeded a predetermined time. If the total operation time has exceeded the predetermined time (YES in step S12), in step S14, the estimation device 100 estimates the surface layer stress for each of the 10 precursor ranges. Next, in step S16, the estimation device 100 estimates the fatigue level for each range by collating the surface layer stress and the operation time ratio with the SN curve. If the total operation time has not exceeded the predetermined time (NO in step S12), the process returns to step S2. The estimation device 100 generates the data in FIG. 4 by executing the processes of steps S2 to S16.

[0073] Next, in step S18, the estimation device 100 calculates the total value of the fatigue levels, and estimates the occurrence time of small-scale damage based on the total value.

[0074] 9 is a flowchart of a method for determining a precursor area and a non-precursor area. This flowchart is executed by the creation device 12.

[0075] In step S102, the creation device 12 sets the bearing temperature number i to "1" and the rotation speed number j to "1". Next, in step S104, the creation device 12 sets a lower limit Tmin and an upper limit Tmax (see FIG. 7) of the bearing temperature. The lower limit Tmin and the upper limit Tmax are, for example, the minimum and maximum temperatures in the nacelle of the wind turbine generator 20, respectively. These minimum and maximum temperatures are estimated taking into consideration the installation location of the wind turbine generator 20, etc.

[0076] Next, in step S106, the creation device 12 sets the bearing temperature Ti=T1=Tmin, and then, in step S108, the creation device 12 sets the rotation speed Rj=R1=1.

[0077] Next, in step S110, the creation device 12 sets the bearing load P(i, j) under the conditions of the bearing temperature Ti and the rotation speed Rj. For example, if there is a specific formula that shows the relationship between the bearing load P(i, j) and the bearing temperature Ti and the rotation speed Rj, the creation device 12 calculates and sets the bearing load P(i, j) using that formula. Also, the bearing load P(i, j) may be the maximum load that can occur within the service life of the wind power generation device 20. In this case, the bearing load P(i, j) becomes a value that tends to be safe within the precursor range (threshold value).

[0078] Next, in step S112, the creation device 12 calculates the oil film thickness h(i,j) under the conditions of the bearing temperature Ti, the rotation speed Rj, and the bearing load P(i,j) using a predetermined calculation formula (not shown). Next, in step S114, the creation device 12 calculates the oil film parameter A(i,j) under the conditions of the bearing temperature Ti, the rotation speed Rj, and the bearing load P(i,j) using, for example, the following formula (6).

[0079] A(i, j) = h / {(R1 2 + R22 )} 0.5 (6)

[0080] The oil film thickness h on the right side of equation (6) represents the oil film thickness formed on the raceway surface and rolling surface of the bearing device 25. R1 represents the root-mean-square roughness of the raceway surface of the outer ring or inner ring of the bearing device 25. Furthermore, R2 represents the root-mean-square roughness of the rolling surfaces of the rolling elements of the bearing device 25.

[0081] Next, the creation device 12 determines whether the oil film parameter A(i, j) is greater than the oil film threshold value (here, "3"). If the oil film parameter A(i, j) is greater than the oil film threshold value (YES in step S116), the process proceeds to step S118. On the other hand, if the oil film parameter A(i, j) is equal to or less than the oil film threshold value (NO in step S116), the process proceeds to step S128.

[0082] In step S118, the creation device 12 calculates Rj as the rotational speed threshold value R when the bearing temperature is Ti. th The threshold is determined as (i). This threshold is shown in FIG.

[0083] Next, in step S118, the creation device 12 determines whether Ti has reached Tmax. If Ti has reached Tmax (YES in step S118), the processing in FIG. 9 ends. If Ti has not reached Tmax (NO in step S118), in step S122, the creation device 12 returns the rotation speed number j to its initial value (=1). Next, in step S124, the creation device 12 increments the bearing temperature number i by "1". Next, in step S126, the creation device 12 adds "1" to the bearing temperature Ti (temperature value) to set the new bearing temperature Ti. Then, the processing returns to step S108.

[0084] In addition, in step S128, the creation device 12 determines Ti and Rj as the precursor range. Next, in step S130, the creation device 12 increments the rotation speed number j by 1. Next, in step S132, the creation device 12 adds "1" to the rotation speed R (rotation speed value) to set the new rotation speed Rj. Then, the process returns to step S110.

[0085] The creation device 12 executes the process of FIG. 9 to set the precursor range and non-precursor range of FIG.

[0086] [Experimental Results]

[0087] Next, a maintenance method according to the present embodiment will be described in the case where the occurrence of small-scale damage is predicted. The maintenance method is a method for suppressing the occurrence of small-scale damage or postponing the occurrence of small-scale damage. The maintenance method according to the present embodiment is a method in which a concentrate is added to the lubricant of the bearing device 25. The concentrate includes a first component and a second component, graphite, and muscovite, which will be described later. The first component is a component belonging to three-layer silicate. The second component is at least one of bentonite, pyrogenic silicic acid, and talc. This maintenance method is disclosed, for example, in "US2014 / 0179569".

[0088] Next, experimental results when this maintenance method was carried out will be described. In this embodiment, a two-cylinder testing machine simulating the rolling contact portion between the raceway surface and the rolling surface of the bearing device 25 was used to evaluate the life extension effect of the above-mentioned maintenance method against small-scale damage. Figure 10 is a schematic diagram showing the configuration of this two-cylinder testing machine 2. Referring to Figure 10, the two-cylinder testing machine 2 has a driving-side rotating shaft D1 and a driven-side rotating shaft F1.

[0089] The drive-side rotating shaft D1 is a member extending in the left-right direction in FIG. 10, and a motor M is connected to the left end in FIG. 10. The motor M enables the drive-side rotating shaft D1 to rotate around a central axis C1 extending in the left-right direction in FIG. 10. A drive-side test piece D2 is attached to the right end of the drive-side rotating shaft D1 in FIG. 10. The drive-side test piece D2 is a member corresponding to either an outer ring (inner ring) or a rolling element included in the bearing device 25. The drive-side test piece D2 is fixed to the right end of the drive-side rotating shaft D1 so as to be rotatable around the central axis C1 as the drive-side rotating shaft D1 rotates.

[0090] On the other hand, the driven-side rotating shaft F1 is a member extending in the left-right direction of FIG. 10, and is rotatable about a central axis C2 extending in the left-right direction of FIG. 10. In FIG. 10, the driven-side rotating shaft F1 has a tip end on the left side and a terminal end on the right side, opposite to the driving-side rotating shaft D1. A driven-side test piece F2 is attached to the tip end on the left side of the driven-side rotating shaft F1 in FIG. 10. The driven-side test piece F2 is a member corresponding to either the outer ring (inner ring) or the rolling element (different from the driving-side test piece D2) included in the bearing device 25. The driven-side test piece F2 is fixed to the tip end on the left side of the driven-side rotating shaft F1 so as to be rotatable about the central axis C2 with the rotation of the driven-side rotating shaft F1.

[0091] The tip of the driving side rotating shaft D1 faces the right side in FIG. 10, and the tip of the driven side rotating shaft F1 faces the left side in FIG. 10. However, the central axis C1 of the driving side rotating shaft D1 does not coincide with the central axis C2 of the driven side rotating shaft F1. The central axis C1 of the driving side rotating shaft D1 and the central axis C2 of the driven side rotating shaft F1 are spaced apart in the vertical direction in FIG. 10. Therefore, the outer diameter surface of the driving side test piece D2 fixed to the tip of the driving side rotating shaft D1 and the outer diameter surface of the driven side test piece F2 fixed to the tip of the driven side rotating shaft F1 are arranged to contact each other at the outer diameter surface contact part DF. A grease retaining part 3 is provided below the driving side test piece D2 and the driven side test piece F2 arranged to contact each other in FIG. 10. The driving side test piece D2 and the driven side test piece F2 are held by the grease retaining part 3 and are in contact with the grease (lubricant) supplied from the grease retaining part 3.

[0092] FIG. 11 is a diagram showing the operating conditions of the twin-cylinder testing machine 2. Referring to FIG. 11, the twin-cylinder testing machine 2 is equipped with a poly-α-olefin (kinetic viscosity at 40° C.: 17.4 mm 2 The lubricant was stored in the grease retainer 3 and kept in lubricating condition by contacting with the outer diameter surfaces of the driving test piece D2 and the driven test piece F2 during the test.

[0093] As driving conditions in the test, the rotation speed of the driving-side rotating shaft D1 around the central axis C1 was set to 1000 rpm. The load W applied to the driven-side test piece F2 was set to 230 kgf. The driving-side rotating shaft D1 was rotated around the central axis C1 by the motor M. As a result, the driven-side rotating shaft F1 was rotated around the central axis C2 due to the frictional force at the outer diameter surface contact portion DF. The rotation direction of the driven-side rotating shaft F1 was opposite to that of the driving-side rotating shaft D1.

[0094] After a certain number of loads had been applied to the test pieces, the test was interrupted. During the interruption of the test, the area ratio of micropitting (small-scale damage) on the outer diameter surface of the driven test piece F1 was measured. After that, the test was restarted, interrupted, and the area ratio measurement was repeated. The test was finally terminated when the total number of loads applied to the driven test piece F2 reached 5 million times.

[0095] FIG. 12 is a diagram showing the characteristics of the drive-side test piece D2. Referring to FIG. 12, the drive-side test piece D2 was cylindrical with an outer diameter of 40 mm, an inner diameter of 20 mm, and a width of 12 mm. The width means the dimension along the left-right direction along which the central axis C1 of FIG. 10 extends. The drive-side test piece D2 was given a minor curvature of a radius of 60 mm in the axial direction (left-right direction in FIG. 10) of the outer diameter surface. The arithmetic mean roughness in the axial direction of the outer diameter surface of the drive-side test piece D2 was about 0.40 μm. The drive-side test piece D2 was made of JIS-SUJ2. The drive-side test piece D2 was subjected to general quenching and tempering treatment to have a Rockwell hardness of about 61 HRC at the end surface. The end surfaces of the drive-side test piece D2 are the left end surface and the right end surface in FIG. 10.

[0096] FIG. 13 is a diagram showing the characteristics of the driven-side test piece F2. Referring to FIG. 13, the driven-side test piece F2 was cylindrical with an outer diameter of 40 mm, an inner diameter of 20 mm, and a width of 12 mm. The driven-side test piece F2 had no minor curvature in the axial direction of the outer diameter surface (the left-right direction in FIG. 10). The arithmetic mean roughness in the axial direction of the outer diameter surface of the driven-side test piece F2 was about 0.05 μm. The driven-side test piece F2 was made of JIS-SUJ2. The driven-side test piece F2 was subjected to general quenching and tempering treatment to have a Rockwell hardness of about 61 HRC at its end surface.

[0097] A total of two tests were conducted by combining the above-mentioned driving side test piece D2 and the driven side test piece F2. That is, two driving side test pieces D2 and two driven side test pieces F2 were prepared. In the first of the two tests, a test of the "Example" was conducted.

[0098] FIG. 14 is a diagram for explaining the "Example". As shown in FIG. 14, when the load number of the test piece reached 100,000 times, the above-mentioned concentrate was added to the lubricant. In the example, Duragear (registered trademark) manufactured by Rewitec of Germany was added to the lubricant to include the concentrate. The amount of the concentrate added was adjusted to be 2 wt % of the total weight of the lubricant after addition.

[0099] Meanwhile, in the second of the two tests, a "Comparative Example" was tested, in which no concentrate was added to the lubricant, as shown in FIG.

[0100] For each of the examples and comparative examples using the above test pieces, the rolling part (outer diameter surface) of the driven test piece F2 was observed with an optical microscope when the test was interrupted and completed, and the occurrence area rate of micropitting was investigated. Here, the occurrence area rate of micropitting refers to the ratio of the area of ​​individual micropeel of micropitting (aggregation of micropeel of micron order size) and the area of ​​crack initiation part as its precursor to the observation field area with the optical microscope.

[0101] Fig. 15 is a diagram showing the change in the occurrence area rate of micropitting with respect to the number of load cycles in the examples and the comparative examples. In the example of Fig. 15, the examples are shown by solid lines, and the comparative examples are shown by dashed lines. In the comparative example of Fig. 15, the occurrence area rate of micropitting increases after 100,000 load cycles. On the other hand, in the example of Fig. 15, the occurrence area rate of micropitting was 0 until the end of the test.

[0102] From the results of FIG. 15, it is possible to suppress the occurrence of micropitting on the surface of the rolling part of the driven test piece F2 by adding an additive component to the lubricant during operation as in the example of this embodiment. In this experiment, the lubricant contains the additive component (concentrate) at a stage where the area rate of micropitting occurrence is 0 (i.e., before micropitting occurs). However, users may contain the additive component at a stage where the area rate of micropitting occurrence is 0 or higher (i.e., after the timing when micropitting has already occurred).

[0103] [Summary of the first embodiment]

[0104] (1) In the first embodiment, as shown in Fig. 4, the range information 121 is information on a plurality of first ranges of a first physical quantity (rotation speed). Also, as shown in Fig. 4, the estimation device 100 identifies, for each of the plurality of first ranges, a time ratio (time parameter) related to a time during which the first physical quantity belongs to the first range. Then, the estimation device 100 estimates the occurrence time of surface damage of the bearing device 25 based on the time ratio for each of the plurality of first ranges and a surface layer stress parameter related to the surface layer stress of the bearing device for each of the plurality of first ranges.

[0105] According to this configuration, the estimation device 100 estimates the time of surface damage based on the surface layer stress parameters, and can estimate the time of occurrence of small-scale surface damage. In addition, for the surface layer stress parameters, the estimation device 100 estimates the time of occurrence of surface damage based on the time parameters for each of the multiple first ranges and the surface layer stress parameters for each of the multiple first ranges. Therefore, for example, the estimation accuracy can be improved compared to a configuration in which the time of occurrence of surface damage is estimated using only the surface layer stress parameters.

[0106] (2) Moreover, as shown in Fig. 4, the range information 121 includes information on a plurality of second ranges (ranges of bearing temperatures), each of which is constituted by at least one first range. The time parameter in the example of Fig. 4 is a parameter related to a time during which the second physical quantity belongs to each of the plurality of second ranges and the first physical quantity belongs to a plurality of first ranges. In this embodiment, the estimation device 100 estimates the occurrence time of surface damage of the bearing device 25 using not only the first physical quantity but also the second physical quantity, thereby improving the estimation accuracy.

[0107] (3) The estimation device 100 estimates the fatigue degree of the bearing device based on a value based on a time parameter in the first range (e.g., the number of loads), the surface layer stress indicated by the surface layer stress parameter in the first range, and estimation information, and estimates small-scale damage according to the fatigue degree. The estimation information is, for example, an SN curve that indicates the relationship between the surface layer stress and the micropitting life.

[0108] With this configuration, the magnitude of the load stress, which is determined by natural wind conditions, and the number of times that load is received can be taken into account, making it possible to accurately estimate the occurrence time of micropitting, which is small-scale surface damage to the bearing device.

[0109] (4) As shown in step S2 of FIG. 8, the estimation device 100 starts identifying a time parameter when a predetermined condition is established that the first physical quantity falls within a predetermined range (signal range) related to surface damage.

[0110] With this configuration, the estimation device 100 can reduce the processing load of identifying time parameters, as compared with a device that always identifies time parameters, for example.

[0111] (5) Furthermore, as shown in FIG. 9, the creation device 12 changes the first physical quantity by a first predetermined amount (in the example of step S132, "1"). At the same time, the creation device 12 changes the second physical quantity by a second predetermined amount (in the example of step S126, "1"). Then, the creation device 12 creates a predetermined range (precursor range) based on the oil film parameter (above formula (6)) of the bearing device 25 exceeding a threshold value due to the change in the first physical quantity and the change in the second physical quantity (see step S116). With this configuration, the predetermined range is created appropriately.

[0112] <Second embodiment>

[0113] Generally, the surface shape C of the raceway surface and rolling surface and the residual stress P are changed due to minor wear and plastic deformation of the bearing device 25 during operation of the wind power generation device 20. j、res、 and Q jk、res The change in the surface shape C is caused by the change in the contact stress P j、con , and Q jk、con In the present disclosure, these phenomena are also referred to as "run-in phenomena." In the first embodiment, the influence of run-in is not taken into account, so the contact stress P j、con , and Q jk、con and residual stress P j、res、 and Q jk、res was considered fixed.

[0114] Generally, the surface layer stress tends to decrease over time due to the influence of break-in. Therefore, in the first embodiment in which the influence of break-in is not taken into account when estimating the surface layer stress, the surface layer stress is calculated to be larger than the actual one, and as a result, the total value Pm of the fatigue degree may be calculated to be larger than the actual one. In other words, the estimation device 100 of the first embodiment may estimate a lifespan that is shorter than the actual lifespan.

[0115] Therefore, according to the estimation device 100 of the second embodiment, the accuracy of estimating the occurrence time of small-scale damage is improved by taking into account the effect of run-in.

[0116] The estimation device 100 of the second embodiment uses a second calculation formula 132 (see FIG. 2). Hereinafter, the surface layer stress calculated when the first physical quantity and the second physical quantity are substituted into the first calculation formula 131 is also referred to as a "first surface layer stress." In addition, the surface layer stress calculated when the same value as the first physical quantity substituted into the first calculation formula 131 and the same value as the second physical quantity substituted into the second calculation formula 132 is also referred to as a "second surface layer stress." The second surface layer stress is smaller than the first surface layer stress.

[0117] Thus, in the second embodiment, from when the monitoring period starts (when the time-series parameter value falls within the precursor range) until the predetermined period has elapsed, the estimation device 100 calculates the surface stress for each of the 10 ranges using the first calculation formula 131. Furthermore, based on the surface stress calculated using the first calculation formula 131, the estimation device 100 calculates the first total value Pm1 as the total value of the fatigue level for each of the 10 ranges.

[0118] After the predetermined period of time has elapsed since the start of the monitoring period, the estimation device 100 calculates the surface stress for each of the 10 ranges using the second calculation formula 132. Furthermore, the estimation device 100 calculates a second total value Pm2 as a total value of the fatigue levels for each of the 10 ranges based on the surface stress calculated using the second calculation formula 132.

[0119] The estimation device 100 calculates the overall fatigue level sum Pm by adding the first sum Pm1 and the second sum Pm2.

[0120] The "predetermined period" in this embodiment is a period during which a load is applied a predetermined number of times Ns, which will be described later, to the bearing device 25. As a modified example, the predetermined period may be a predetermined amount of time.

[0121] Next, a method for creating the second calculation formula 132 will be described. First, in step A, an operator selects a representative precursor range from the multiple (10) precursor ranges in the table of Fig. 4. In addition, as the representative precursor range, for example, the precursor range with the maximum operating condition ratio is selected.

[0122] Next, the worker performs a rolling fatigue test simulating the precursor range selected in step A. For this test, for example, the above-mentioned two-cylinder testing machine 2 is used. Furthermore, the test specimen is prepared so that the material, surface shape, and residual stress of the rolling contact part are as similar as possible to the race and rolling element of the bearing device 25 of the wind power generation device 20 that is the subject of the life estimation. The conditions of the rolling fatigue test are set so that the surface speed of the rolling contact part, the contact pressure, the type of lubricant, the oil film parameters, the temperature of the lubricant, and the like are as similar as possible to the bearing device 25 of the wind power generation device 20 that is the subject of the life estimation under the precursor range.

[0123] In step C, the operator stops the rolling fatigue test in step B when the number of load cycles of the test piece reaches the above-mentioned predetermined number Ns, and measures the surface shape C1 and residual stress H1 of each test piece corresponding to the raceway and the rolling element. These surface shape C1 and residual stress H1 are values ​​that reflect the influence of the above-mentioned running-in. Based on the description of the first calculation formula 131, the second calculation formula 132 is expressed by the following formula (7).

[0124] F=G(V, T, A, B, C1, D, E, H1) (7)

[0125] 16 is a flowchart showing the process of the estimation device 100 of the second embodiment. If the determination in step S2 is YES, then in step S4, the estimation device 100 starts monitoring the time series data. Then, in step S24, it is determined whether the total time of the monitoring period has elapsed a predetermined period. The estimation device 100 executes the allocation process of the time series data until the predetermined period has elapsed (NO in step S24, step S26). Then, when the predetermined period has elapsed (YES in step S24), in step S28, the estimation device 100 calculates a first total value Pm1.

[0126] Next, the estimation device 100 executes the allocation process of the time series data until the monitoring period has elapsed (NO in step S30, step S32). Then, when the monitoring period has elapsed (YES in step S30), in step S34, the estimation device 100 determines whether the total operation time of the wind power generation device has elapsed a predetermined time. If the total operation time has elapsed a predetermined time (YES in step S34), in step S36, the estimation device 100 calculates a second total value Pm2. If the total operation time has not elapsed a predetermined time (NO in step S34), the process returns to step S2. Next, in step S38, the estimation device 100 calculates a total value Pm of the fatigue level by adding the first total value Pm1 and the second total value Pm2, and executes the process of step S14.

[0127] Fig. 17 is a diagram showing an example of the history of surface layer stress estimated by the estimation device 100 of the first embodiment. Fig. 18 is an example of the stress history when the estimation device 100 of the second embodiment is applied to the bearing device 25 when the stress history of Fig. 17 is obtained.

[0128] Timing C in Fig. 18 is an example of timing indicating that the total time of the monitoring period has passed a predetermined period. The estimation device 100 of the second embodiment estimates the surface layer stress using the first calculation formula 131 from the start of the monitoring period to timing C when the predetermined period has passed. Therefore, the surface layer stress calculated by timing C when the total time of the monitoring period has passed the predetermined period is the same between the estimation device 100 of the first embodiment and the estimation device 100 of the second embodiment.

[0129] After timing C, the estimation device 100 of the second embodiment calculates the surface layer stress using the second calculation formula 132. Therefore, as shown in Fig. 17 and Fig. 18, after timing C, the surface layer stress calculated by the estimation device 100 of the second embodiment becomes smaller than the surface layer stress calculated by the estimation device 100 of the first embodiment.

[0130] 18, the estimation device 100 of the second embodiment calculates a surface layer stress F11 that is smaller than the surface layer stress F1 in the period T1. Also, the estimation device 100 of the second embodiment calculates a surface layer stress F21 that is smaller than the surface layer stress F2 in the period T2.

[0131] As described above, in the second embodiment, the "influence of running in" is reflected. In other words, the surface layer stress tends to decrease after a predetermined period of time has elapsed since the establishment of a predetermined condition. Therefore, the estimation device 100 of the second embodiment calculates the surface layer stress using the second calculation formula 132 after timing C, when a predetermined period of time has elapsed since the start of the monitoring period. Therefore, the estimation device 100 of the second embodiment can estimate the occurrence time of surface damage in accordance with this tendency, thereby improving the estimation accuracy. [Third embodiment] In the third embodiment, the estimation device 100 executes a notification process for notifying a user A of the bearing device 25 of the occurrence time of surface damage (micropitting). Such notification process is, for example, a process of displaying an image indicating the occurrence time of the surface damage on the user terminal 50 of the user A. The occurrence time of the surface damage is the time estimated by the estimation device 100 described in the first or second embodiment above. FIG. 19 is a diagram for explaining a display image of the user terminal 50. In the example of FIG. 19, a first image 51 indicating the estimated time when surface damage will occur is displayed. This first image 51 is a text image stating "Micropitting is estimated to occur on year A, month B, day C." In this image, "year A, month B, day C" corresponds to the time of surface damage (the time estimated by the estimation device 100). With this configuration, the estimation device 100 can allow the user of the bearing device 25 to recognize the time of surface damage. Furthermore, the estimation device 100 of this embodiment notifies the user A of a maintenance method for the bearing device 25. The maintenance method for the bearing device 25 is, for example, a method for suppressing the occurrence of surface damage or postponing the occurrence of surface damage of the bearing device 25. This maintenance method includes at least one of a first method and a second method. 19, a second image 52 is displayed, which shows a maintenance method for the bearing device 25. The second image 52 includes a prompting image 59, a first method image 61, and a second method image 62. The promotion image 59 is an image for promoting maintenance to the user. The promotion image 59 is an image with text stating, "Please carry out one of the following maintenance procedures (a) or (b) on or after the date C of the month B of the year A. This will extend the life of the bearing device." The first technique image 61 is an image showing the first technique. The first technique is a technique of adding a concentrate to the lubricant of the bearing device 25. The first technique image 61 is a text image saying "(a) Add a concentrate to the lubricant of the bearing device." The second method image 62 is an image showing the second method. The second method is a method of replacing the (currently used) lubricant of the bearing device 25 with a lubricant having a higher kinetic viscosity of the base oil component than the currently used lubricant. The second method image 62 is a text image saying "(b) Replace the lubricant of the bearing device with a lubricant with a higher kinetic viscosity." In this way, the estimation device 100 notifies the user of the bearing device 25 of the maintenance method for the bearing device 25. Therefore, the user A can be made aware of the maintenance method for the bearing device 25. This maintenance technique is also a technique that can extend the life of the bearing device 25. Therefore, the user A can be encouraged to extend the life of the bearing device 25. Furthermore, as shown in the first method image 61 and the second method image 62, the estimation device 100 notifies the user A of the first method and the second method as maintenance methods for the bearing device 25. This allows the user to recognize a specific maintenance method for the bearing device 25. Then, by the user performing the maintenance indicated by the first method or the second method on the bearing device 25, the occurrence of surface damage to the bearing device 25 can be suppressed or the occurrence of surface damage can be postponed. The inventors have also discovered that the progress of micropitting tends to be appropriately suppressed by performing maintenance using the first and second methods after micropitting occurs. In other words, the first method is preferably a method of adding a concentrate to the lubricant of the bearing device 25 after micropitting occurs. In addition, the second method is preferably a method of replacing the lubricant of the bearing device 25 with a lubricant having a high kinetic viscosity after micropitting occurs. In other words, the life of the bearing device 25 can be extended by the first or second method. In other words, the first and second methods are life extension methods based on an estimation of the time when micropitting occurs. Therefore, the estimation device 100 displays "maintenance will be performed on or after A year, B month, C day," as shown in a prompting image 59 in Fig. 19. Therefore, the estimation device 100 can notify the user of a maintenance method that can appropriately suppress the occurrence of micropitting. By performing maintenance using the second method, the oil film parameter of the rolling contact area under the same operating conditions becomes larger than that before the lubricant change. This reduces the risk of micropitting due to contact with surface roughness.

[0132] <Modification>

[0133] (1) In the above embodiment, the configuration has been described in which the maintenance method is a method performed by the user. However, the maintenance method may not be a method performed by the user, but may be a method performed by the estimation device 100. For example, when the estimation device 100 estimates the time when small-scale damage will occur, it executes the following suppression process, for example, before the time is reached. This suppression process is a process of increasing the rotation speed of the bearing device 25 so that the oil film parameter A (see the above formula (6)) of the rolling contact portion of the bearing device 25 becomes equal to or greater than a reference value (for example, 3). This suppression process can reduce the frequency of contact of the surface roughness at the rolling contact portion, thereby reducing the risk of micropitting. Specifically, the estimation device 100 outputs a control signal for increasing the shaft rotation speed of the bearing device 25. The wind power generation device 20 increases the rotation speed of the bearing device 25 by receiving this control signal. Also, a storage battery may be provided in the wind power generating device 20, and the power generated by the wind power generating device 20 may be stored in the storage battery, and the rotational speed of the bearing device 25 may be increased by using the stored power.

[0134] (2) The estimation device 100 in the above embodiment predicts the occurrence time of small-scale damage using two physical quantities (the first physical quantity (rotation speed) and the second physical quantity (bearing temperature)). However, the estimation device 100 may use only one physical quantity, or may use three or more physical quantities. When the estimation device 100 is configured to use only one physical quantity (rotation speed), for example, it may be configured to distribute the time series data values ​​into ranges M1 to M4.

[0135] (3) At least one function described in the estimation device 100 of this embodiment may be executed by another device (for example, the collection device 30). That is, as a modified example, the collection device 30 and the estimation device 100 may execute the above-mentioned processing of the estimation device 100. In this way, even if the processing is executed by two or more separate devices, the two or more devices are referred to as the estimation device.

[0136] (4) In the above embodiment, the configuration has been described in which the time series data collected by the collection device 30 is directly output to the estimation device 100. However, the time series data collected by the collection device 30 may be temporarily recorded in another data recording device. When such a configuration is adopted, the time series data recorded in the other data recording device is transmitted to the estimation device 100 at a predetermined interval. This interval may be a period of one day, one week, one month, or the like, depending on the upper limit of the amount of data that can be recorded in the other data recording device.

[0137] (5) In the above-described Fig. 4, the surface layer stress is estimated for each range ID. However, the surface layer stress for each range ID may be a fixed value.

[0138] [Note]

[0139] It will be appreciated by those skilled in the art that the exemplary embodiments described above are examples of the following aspects.

[0140] (Supplementary Note 1) An estimation device comprising: an interface that acquires a first physical quantity of a bearing device from a first sensor that detects the first physical quantity; a memory that stores range information relating to a plurality of first ranges of the first physical quantity; and a calculation device, wherein the calculation device identifies, for each of the plurality of first ranges, a time parameter relating to a time during which the first physical quantity falls within the first range, and estimates a time when surface damage of the bearing device will occur based on the time parameter for each of the plurality of first ranges and a surface stress parameter for each of the plurality of first ranges,

[0141] According to this configuration, the time of occurrence of surface damage is estimated based on the surface layer stress parameter, so that the time of occurrence of relatively small-scale surface damage can be estimated. In addition, the time of occurrence of surface damage is estimated based on the surface layer stress parameter, the time parameters for each of the multiple first ranges, and the surface layer stress parameter for each of the multiple first ranges. Therefore, for example, the estimation accuracy can be improved compared to a configuration in which the time of occurrence of surface damage is estimated using only the surface layer stress parameter.

[0142] (Supplementary Note 2) An estimation device as described in Supplementary Note 1, wherein the interface acquires a first physical quantity from a second sensor that detects a second physical quantity of the bearing device, the range information includes information on a plurality of second ranges each composed of at least one first range, and the calculation device identifies, as a time parameter, a parameter related to time at which the second physical quantity belongs to each of the plurality of second ranges and the first physical quantity belongs to a plurality of first ranges that constitute the second range.

[0143] According to such a configuration, the occurrence time of surface damage of the bearing device is estimated using not only the first physical quantity but also the second physical quantity, so that the estimation accuracy can be improved.

[0144] (Appendix 3) An estimation device as described in Appendix 1 or Appendix 2, wherein the calculation device calculates fatigue levels in multiple first ranges based on a value based on a time parameter in the first range, a surface stress indicated by a surface stress parameter in the first range, and specified estimation information, and estimates the time when the total value of the fatigue levels in the multiple first ranges reaches 1 as the time when surface damage occurs.

[0145] According to this configuration, it is possible to accurately estimate the occurrence time of micropitting, which is small-scale surface damage of the bearing device, taking into account the degree of fatigue.

[0146] (Supplementary Note 4) The estimation device according to any one of Supplementary Note 1 to Supplementary Note 3, wherein the calculation device starts identifying the time parameter when a predetermined condition is satisfied that the first physical quantity falls within a predetermined range related to surface damage.

[0147] According to such a configuration, for example, the processing load for identifying time parameters can be reduced compared to an apparatus that always identifies time parameters.

[0148] (Appendix 5) The memory stores a first calculation formula and a second calculation formula, the first calculation formula and the second calculation formula being formulas for calculating a surface layer stress parameter in a first range by substituting a first physical quantity, the surface layer stress parameter calculated when the first physical quantity is substituted into the second calculation formula is smaller than the surface layer stress parameter calculated when the same value as the first physical quantity substituted into the first calculation formula is substituted, the calculation device calculates a first total value in the first range using the first calculation formula from the time a specified condition is satisfied until a specified period of time has elapsed, and calculates a second total value in the first range using the second calculation formula after the specified period of time has elapsed since the specified condition is satisfied, and the estimation device described in Appendix 4 estimates the occurrence time of surface damage to be shorter in time from the present time as the sum of the first total value and the reduced second total value is larger.

[0149] Generally, the surface stress of a bearing device tends to decrease as the bearing device is used. In other words, the surface stress tends to decrease after a predetermined period of time has elapsed since a predetermined condition was met. Therefore, according to this configuration, the second total value is reduced. Therefore, the time when surface damage will occur can be estimated based on this tendency.

[0150] (Appendix 6) An estimation device as described in any one of Appendices 1 to 5, wherein the memory stores range information relating to a predetermined range including a plurality of first ranges, and the calculation device estimates the time of occurrence of surface damage based on a time parameter and a surface stress parameter without using a time parameter relating to a time belonging to a range outside the predetermined range.

[0151] Generally, time parameters that are outside the predetermined range and are not related to surface damage are not used, so that it is possible to suppress estimation of the occurrence time that is too short from the present time.

[0152] (Appendix 7) An estimation device as described in Appendix 2, wherein the memory stores range information relating to a predetermined range including a plurality of first ranges, and the calculation device estimates the time of occurrence of surface damage based on the time parameter and the surface stress parameter without using a time parameter relating to a time belonging to a range outside the predetermined range, and the predetermined range is created based on the oil film parameter of the bearing device exceeding a threshold value as a result of the first physical quantity being changed by a first predetermined amount and the second physical quantity being changed by a second predetermined amount.

[0153] According to this configuration, the predetermined range is appropriately created.

[0154] (Supplementary Note 8) The estimation device according to any one of Supplementary Notes 1 to 7, wherein the calculation device executes a process for notifying a user of the bearing device of the occurrence time.

[0155] With this configuration, it is possible to let the user of the bearing device know when surface damage has occurred.

[0156] (Supplementary Note 9) The estimation device according to Supplementary Note 8, wherein the notification process includes a process for notifying a user of the bearing device of a maintenance method for the bearing device.

[0157] With this configuration, it is possible to let the user of the bearing device know when surface damage has occurred.

[0158] (Appendix 10) The estimation device according to appendix 9, wherein the maintenance method includes at least one of a first method of adding a concentrate to a lubricant of the bearing device, and a second method of replacing the lubricant of the bearing device with a lubricant having a higher kinetic viscosity than the lubricant of the bearing device.

[0159] According to such a configuration, a specific maintenance method for the bearing device 25 can be made known to the user.

[0160] (Appendix 11) The estimation device described in Appendix 10, wherein the first method is a method of adding a concentrate to the lubricant of the bearing device after the occurrence of the problem, and the second method is a method of replacing the lubricant of the bearing device with a lubricant having a higher kinetic viscosity than the original lubricant after the occurrence of the problem.

[0161] According to this configuration, it is possible to notify the user of a maintenance method that can appropriately suppress the occurrence of surface damage.

[0162] (Appendix 12) A method for estimating a lifespan of a bearing device, comprising: acquiring a first physical quantity of the bearing device from a first sensor that detects the first physical quantity; identifying, for each of a plurality of first ranges of the first physical quantity, a time parameter related to a time during which the first physical quantity falls within the first range; and estimating a time when surface damage of the bearing device will occur based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to the surface stress of the bearing device for each of the plurality of first ranges.

[0163] According to this estimation method, the estimation accuracy can be improved as described above.

[0164] (Appendix 13) A method for extending the life of a bearing device, comprising: acquiring a first physical quantity from a first sensor that detects the first physical quantity of the bearing device; identifying, for each of a plurality of first ranges of the first physical quantity, a time parameter related to a time during which the first physical quantity falls within the first range; and estimating a time when surface damage will occur in the bearing device based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to the surface stress of the bearing device for each of the plurality of first ranges, wherein the life extension method based on the estimation of the time of occurrence includes at least one of a first method of adding a concentrate to a lubricant of the bearing device, and a second method of replacing the lubricant of the bearing device with a lubricant having a higher kinetic viscosity than the original lubricant.

[0165] Such a life extension method can appropriately extend the life of the bearing device.

[0166] (Appendix 14) A life extension method as described in Appendix 13, wherein the first method is to add a concentrate to the lubricant of the bearing device after the occurrence of the problem, and the second method is to replace the lubricant of the bearing device with a lubricant having a higher kinetic viscosity than the original lubricant after the occurrence of the problem.

[0167] Such a life extension method can appropriately extend the life of the bearing device.

[0168] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is defined by the claims, not by the description of the embodiments described above, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0169] 2 Two-cylinder testing machine, 3 Grease retention unit, 10 Management system, 12 Creation device, 20 Wind power generation device, 25 Bearing device, 25A Surface, 25B Surface layer, 25C Interior, 30 Collection device, 41 Speed ​​sensor, 42 Temperature sensor, 50 User terminal, 60 Wind power generation unit, 100 Estimation device, 102 CPU, 104 Memory, 106 Interface, 112 Acquisition unit, 114 Estimation unit, 116 Memory unit, 120 Display, 121 Range information, 131 First calculation formula, 132 Second calculation formula.

Claims

1. an interface that acquires a first physical quantity from a first sensor that detects a first physical quantity of the bearing device; a memory that stores range information relating to a plurality of first ranges of the first physical quantity; A computing device, The computing device includes: identifying a time parameter relating to a time during which the first physical quantity falls within each of the first ranges; an estimation device that estimates a time when surface damage of the bearing device will occur based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to a surface stress of the bearing device for each of the plurality of first ranges.

2. the interface acquires a second physical quantity from a second sensor that detects a second physical quantity of the bearing device; the range information includes information regarding a plurality of second ranges, each of which is constituted by at least one of the first ranges; The computing device includes: The estimation device according to claim 1 , wherein the second physical quantity belongs to each of the plurality of second ranges, and a parameter related to time in which the first physical quantity belongs to a plurality of first ranges constituting the second range is identified as the time parameter.

3. The computing device includes: Calculating fatigue levels in the first ranges based on a value based on the time parameter in the first range, a surface layer stress indicated by a surface layer stress parameter in the first range, and predetermined estimated information; The estimation device according to claim 1 , wherein a time point at which a total value of fatigue levels in the plurality of first ranges reaches 1 is estimated as the time point at which the surface damage occurs.

4. The estimation device according to any one of claims 1 to 3, wherein the calculation device starts identifying the time parameter when a predetermined condition is satisfied that the first physical quantity falls within a predetermined range related to the surface damage.

5. the memory stores a first calculation formula and a second calculation formula; the first calculation formula and the second calculation formula are formulas into which the first physical quantity is substituted to calculate a surface layer stress parameter in the first range, a surface layer stress parameter calculated when the first physical quantity is substituted into the second calculation formula is smaller than a surface layer stress parameter calculated when the same value as the substituted first physical quantity is substituted into the first calculation formula; The computing device includes: calculating a first total value in a first range by the first calculation formula until a predetermined period of time has elapsed since the predetermined condition is satisfied; calculating a second total value in the first range by the second calculation formula after a predetermined period of time has elapsed since the predetermined condition was satisfied; The estimation device according to claim 4 , wherein the estimation device estimates the occurrence time of the surface damage to be shorter in time from the present time as the total value of the first sum value and the lowered second sum value is larger.

6. the memory stores range information regarding a predetermined range including the plurality of first ranges; The estimation device according to any one of claims 1 to 3, wherein the calculation device estimates the time of occurrence of the surface damage based on the time parameter and the surface stress parameter without using a time parameter relating to a time that falls within a range outside the specified range.

7. the memory stores range information regarding a predetermined range including the plurality of first ranges; The computing device estimates the occurrence time of the surface damage based on the time parameter and the surface layer stress parameter without using a time parameter related to a time outside the predetermined range; 3. The estimation device according to claim 2, wherein the predetermined range is created based on an oil film parameter of the bearing device exceeding a threshold value when the first physical quantity is changed by a first predetermined amount and the second physical quantity is changed by a second predetermined amount.

8. The estimation device according to any one of claims 1 to 3, wherein the estimation device executes a notification process for notifying a user of the bearing device of the occurrence time.

9. The estimation device according to claim 8 , wherein the notification process includes a process for notifying a user of the bearing device of a maintenance technique for the bearing device.

10. The maintenance method includes: A first method of dosing a concentrate in the lubricant of the bearing assembly; The estimation device according to claim 9 , further comprising at least one of a second technique of replacing a lubricant of the bearing device with a lubricant having a higher kinetic viscosity than the lubricant of the bearing device.

11. The first method is a method of adding a concentrate to the lubricant of the bearing device after the occurrence time, The estimation device according to claim 10 , wherein the second method is a method of replacing the lubricant in the bearing device with a lubricant having a higher kinetic viscosity than the lubricant in the bearing device after the occurrence time.

12. 1. A method for estimating a life of a bearing device, comprising: acquiring a first physical quantity from a first sensor that detects a first physical quantity of the bearing device; identifying, for each of a plurality of first ranges of the first physical quantity, a time parameter relating to a time during which the first physical quantity falls within the first range; and estimating a time when surface damage of the bearing device will occur based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to a surface stress of the bearing device for each of the plurality of first ranges.

13. A method for extending the life of a bearing assembly, comprising: acquiring a first physical quantity from a first sensor that detects a first physical quantity of the bearing device; identifying, for each of a plurality of first ranges of the first physical quantity, a time parameter relating to a time during which the first physical quantity falls within the first range; and estimating a time when surface damage of the bearing device will occur based on the time parameter for each of the plurality of first ranges and a surface stress parameter related to a surface stress of the bearing device for each of the plurality of first ranges, The method of extending life based on the estimation of the occurrence time is as follows: A first method of dosing a concentrate in the lubricant of the bearing assembly; and a second method of replacing the lubricant in the bearing device with a lubricant having a higher kinetic viscosity than the lubricant in the bearing device.

14. The first method is a method of adding a concentrate to the lubricant of the bearing device after the occurrence time, 14. The method for extending the life of a bearing device according to claim 13, wherein the second technique is a technique of replacing the lubricant of the bearing device with a lubricant having a higher kinetic viscosity than the lubricant of the bearing device after the occurrence of the defect.

Images