JP2019190943A - Rolling bearing fatigue state prediction system - Google Patents

Abstract

To provide a system for precisely predicting the fatigue state including the life of a rolling bearing even if the angle of deviation has generated in the rolling bearing part.SOLUTION: The present invention relates to a rolling bearing fatigue state prediction system for predicting the fatigue state of a rolling bearing supporting a rotational body, the system including: a load/angle of deviation calculation unit for calculating the angle of deviation and a magnitude of the divided load of the bearing of the rolling bearing; a load distribution calculation unit for calculating the load distribution of the rolling bearing on the basis of the angle of deviation and the bearing divided load; and a fatigue state prediction unit for predicting the fatigue state of the rolling bearing based on the load distribution.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a fatigue state prediction system for a rolling bearing that supports a rotating body.

  Predicting the life of a rolling bearing is important in selecting the bearing type and optimizing the replacement time of the bearing. However, since the rolling contact characteristics are complicated and the number of parts related to fatigue is large, even if the same type of bearing is used and used under the same conditions, its life varies greatly from bearing to bearing.

Therefore, a method has been proposed in which a Weibull distribution is applied to the bearing life distribution state and a value representative of the distribution is used, and this method is still used. Also, the life (basic rating life L ₁₀ ) in which 10% of all the bearings are damaged is the most frequently used with respect to the life of the bearing (see (Equation 1)).

  C in (Formula 1) is called a basic dynamic load rating, and is a parameter indicating the dynamic load capacity of the bearing. P is an equivalent load applied to the bearing. The index p is 3 for ball bearings and 10/3 for roller bearings.

It has been used for a long time the basic rating life L ₁₀ obtained by (Equation 1), in recent years, the life of the bearing, the fatigue limit load of the bearings, lubrication, driving environment, pollution particles during operation, when mounting It has become clear that cleanliness affects. Based on a number of test results, a correction factor a _ISO that takes them into account has been newly proposed. The correction factor a _ISO and a factor a ₁ for calculating an arbitrary damage probability n% are set as basic ratings. A modified rated life L _nm multiplied by the life L ₁₀ was proposed (see (Equation 2)).

  In recent years, there is an increasing demand for optimizing the replacement time of rolling bearings and maximizing the availability of products. As a method, a method has been proposed in which physical quantities related to rolling bearings in actual operation are measured, and the fatigue data of rolling bearings that change from moment to moment is evaluated using the measured data. A technique for measuring the load during actual operation is proposed in Patent Document 1.

  For example, in claim 1 of Patent Document 1, “an incident wave is generated from an ultrasonic probe attached to a bearing housing on which a bearing is supported toward a bearing outer ring of the bearing, and the bearing housing and the bearing outer ring are generated. A bearing support structure for measuring a bearing load by measuring a reflected wave from a boundary position between the bearing and the bearing housing, and a fitting hole into which the bearing outer ring is fitted is formed in the bearing housing. And a bearing support structure for measuring a bearing load, characterized in that a filler is interposed between the fitting hole and the bearing outer ring, and this ultrasonic probe is used. Thus, a method is described in which the degree of adhesion between the rolling element and the inner ring or outer ring of the rolling bearing is obtained, and the bearing load is estimated from the result (paragraph 0023 to paragraph 0027 of Patent Document 1).

Japanese Patent No. 3842555

  By the way, an actual rotating shaft may be deformed by its own weight or gear load, and a declination may occur in the rolling bearing portion. When there is a declination, the load distribution in the rolling bearing becomes non-uniform, resulting in a region where the contact load is locally increased. Since the bearing life is affected by the maximum contact load between the rolling element and the bearing inner and outer rings, the bearing contact angle is evaluated and the maximum contact load is estimated in order to accurately predict the bearing life. There is a need.

  However, since the technique described in Patent Document 1 does not take into account the deflection angle of the rolling bearing portion caused by load deformation of the rotating shaft, the prediction accuracy of the bearing life is low in principle.

  An object of the present invention is to provide a rolling bearing fatigue state prediction system for accurately predicting a fatigue state including the life of a rolling bearing even when a declination occurs in the rolling bearing portion.

  In order to solve the above problems, a rolling bearing fatigue state prediction system of the present invention predicts the fatigue state of a rolling bearing that supports a rotating body, and determines the magnitude and declination of the bearing shared load of the rolling bearing. A load / deflection calculation unit to be calculated, a load distribution calculation unit to calculate the load distribution of the rolling bearing based on the bearing shared load and the deflection angle, and a fatigue state of the rolling bearing to be predicted based on the load distribution And a fatigue state prediction unit.

  According to the present invention, even when a declination occurs in the rolling bearing, the maximum contact load at the declination can be obtained, so that the fatigue state of the rolling bearing can be accurately predicted. Further, since the operator can check the fatigue state, the replacement timing of the rolling bearing can be optimized, and as a result, the operating rate can be improved.

1 is an overall schematic configuration diagram of a rolling bearing fatigue state prediction system of Example 1. FIG. The functional block diagram of the rolling bearing fatigue state prediction system of Example 1. FIG. The whole processing flow figure of the rolling bearing fatigue state prediction system of Example 1. FIG. Schematic which showed the example of the rotational power transmission mechanism. The schematic diagram when the rotating shaft connected with the motor deform | transformed in FIG. The figure which showed the contact load distribution state when a declination arises in a rolling bearing. FIG. 3 is a schematic diagram of a fatigue state calculation method in the first embodiment. The figure which shows an example of the screen display of the display apparatus shown in FIG. The figure which shows an example of the screen display of the display apparatus shown in FIG. Schematic which shows the deflection angle measuring method of the rolling bearing fatigue state prediction system of Example 2. FIG. The figure which showed the relationship of the maximum local load ratio at the time of a deflection angle and the presence or absence of a deflection angle of the rolling bearing fatigue state prediction system of Example 3. FIG. Schematic of rolling bearing fatigue state prediction system applied to wind turbine generator

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In each embodiment, a “rolling bearing” is a cylindrical inner ring arranged to cover the outer peripheral surface of a rotating body represented by a rotating shaft, A cylindrical outer ring that covers the outer peripheral surface of the cylindrical inner ring and that is spaced apart from the outer peripheral surface of the inner ring in the radial direction at a predetermined interval and arranged concentrically, and a bearing that is arranged to cover the outer peripheral surface of the outer ring A rolling “ball bearing” in which a plurality of spherical balls are arranged in the circumferential direction as rolling elements between the outer peripheral surface of the inner ring and the inner peripheral surface of the outer ring, or between the outer peripheral surface of the inner ring and the outer ring. A “rolling roller bearing” in which a plurality of cylindrical rollers are arranged in the circumferential direction as rolling elements between the inner peripheral surface and the inner peripheral surface is included.

  The “rolling bearing” has a plurality of arc-shaped deep grooves formed at predetermined intervals in the circumferential direction on the outer peripheral surface of the cylindrical inner ring, and is formed on the inner peripheral surface of the cylindrical outer ring and on the outer peripheral surface of the inner ring. A deep groove bearing in which an arc-shaped deep groove is formed at a position facing the formed deep groove.

  Furthermore, in each embodiment, as a rotating machine having a rolling bearing that rotatably supports the rotating shaft and changing the direction of the load applied to the rolling bearing, for example, at a wind power generator, an excavation site, or a construction site. Includes construction machinery used.

  Hereinafter, a rolling bearing fatigue state prediction system 1 according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 7.

  FIG. 1 is an overall schematic configuration diagram of a rolling bearing fatigue state prediction system 1 according to a first embodiment. As shown here, the rolling bearing fatigue state prediction system 1 is a system including at least an arithmetic device 4, an input device 5, and a display device 6. Among these, the arithmetic device 4 predicts the fatigue state of the rolling bearing 2 based on the measurement data (measurement signal), the input device 5 is a keyboard or a mouse operated by the operator, and the display device 6 is These are displays such as LCDs and organic ELs that display signals from the arithmetic unit 4.

  As shown in FIG. 1, the rolling bearing 2 to be monitored by the arithmetic device 4 is arranged so as to cover the outer peripheral surface of the rotating shaft 3, and has a cylindrical inner ring 2a fitted to the rotating shaft 3, and an inner ring 2a. Between the outer peripheral surface of the inner ring 2a and the outer peripheral surface of the inner ring 2a, the outer peripheral surface of the inner ring 2a and the outer peripheral surface of the inner ring 2a and the inner peripheral surface of the outer ring 2b. Are provided with a plurality of rolling elements 2c (spherical balls, cylindrical rollers) spaced apart from each other at a predetermined interval in the circumferential direction. In FIG. 1, a bearing housing that covers the outer peripheral surface of the outer ring 2b is omitted.

  FIG. 2 is a functional block diagram of the rolling bearing fatigue state prediction system 1. Note that the arithmetic device 4 is specifically a computer such as a personal computer, an arithmetic device such as a CPU (not shown), a main storage device such as a semiconductor memory, an auxiliary storage device such as a hard disk, and hardware such as a communication device. The arithmetic unit executes the program stored in the main storage device while referring to the database recorded in the auxiliary storage device, and implements each function to be described later. A description will be given while appropriately omitting known techniques.

  As shown in FIG. 2, the arithmetic device 4 of the present embodiment receives operation data and data from the input device 5 and outputs the processing results and the like to the display device 6. In order to realize, an input interface (hereinafter “input unit 4a”), a measurement value acquisition unit 4b, a load / deflection calculation unit 4c, a load distribution calculation unit 4d, a fatigue state prediction unit 4e, a storage unit 4f, a display control unit 4g, an output interface (hereinafter referred to as “output unit 4h”), and an internal bus 4i for interconnecting them. Each will be described in turn below.

  In the input unit 4a, measured values (running machine operation data) measured for predicting the fatigue state of the rolling bearing 2 and setting information (lubrication conditions, filter conditions, etc.) set by the operator via the input device 5 are provided. Is entered.

  The storage unit 4f acquires the setting information and design information input from the input device 5 via the internal bus 4i and stores them in a predetermined storage area, and also stores the output of a measured value acquisition unit 4b and the like described later in a predetermined storage. It is stored in the area, and specifically comprises a semiconductor memory, a hard disk or the like.

  The measurement value acquisition unit 4b extracts and acquires data corresponding to the measurement value from the data input to the input unit 4a. The measured values are motor power, rotation speed, and rotational speed that are always measured when the monitored rotating machine is a motor, and are always measured when the monitored rotating machine is a generator. Although the amount of power generation, the number of rotations, and the rotation speed, other types of data may be used as measurement values. The measured values acquired here are transferred to the load / deflection angle calculation unit 4c via the internal bus 4i and stored in a predetermined storage area of the storage unit 4f.

  The load / deflection calculation unit 4c is based on the measurement value acquired from the measurement value acquisition unit 4b via the internal bus 4i, and the bearing shared load F applied to each rolling bearing 2 and the rotation shaft 3 by the bearing shared load F. The deformation is calculated, and the deflection angle θ of each rolling bearing 2 is calculated based on the deformation calculation result. The bearing share load F and the deviation angle θ calculated here are transferred to the load distribution calculation unit 4d via the internal bus 4i and stored in a predetermined storage area of the storage unit 4f.

The load distribution calculation unit 4d calculates the bearing load distribution f in the rolling bearing 2 based on the bearing shared load F and the deflection angle θ acquired from the load / deflection angle calculation unit 4c via the internal bus 4i. Further, the maximum local load f _max ′, which is the maximum value of the obtained bearing load distribution f, is transferred to the fatigue state prediction unit 4e via the internal bus 4i and stored in a predetermined storage area of the storage unit 4f. .

The fatigue state prediction unit 4e includes the maximum local load f _max ′ acquired from the load distribution calculation unit 4d via the internal bus 4i, and the rotational speed acquired from the measurement value acquisition unit 4b or the storage unit 4f via the internal bus 4i. Based on the data, the fatigue state of the rolling bearing 2 is predicted. The prediction result of the fatigue state of the rolling bearing 2 obtained by the fatigue state prediction unit 4e is displayed on the display screen 7 of the display device 6 via the display control unit 4g and the output unit 4h.

  Here, the “fatigue state” is, for example, any one of the remaining life of the rolling bearing 2, the cumulative damage degree, the damage probability, or the risk that is an index obtained by multiplying the damage probability by the degree of influence at the time of damage, or Any combination of these. The “influence degree” used when calculating the risk means, for example, the cost required for parts replacement and the cost loss (damage amount) caused by stopping the rotating machine having the rolling bearing for the part replacement.

  Next, the processing flow of the rolling bearing fatigue state prediction system 1 of the present embodiment will be described with reference to FIGS.

  FIG. 3 is an overall process flow diagram of the rolling bearing fatigue state prediction system 1.

  First, in step S <b> 1, the load / deflection calculation unit 4 c calculates a bearing shared load F applied to each rolling bearing 2. Here, the calculation method of the bearing share load F by the load / deflection calculation unit 4c will be described with reference to FIG.

  In FIG. 4, the rotating shaft 3A of the motor 9 is supported by the rolling bearings 2A and 2B, and the rotational power of the motor 9 is transmitted to the rotating shaft 3B supported by the rolling bearings 2C and 2D via the gears 10A and 10B. It is the schematic which showed the rotational power transmission mechanism. In this example, the rotating shafts 3A and 3B are both supported by the pair of rolling bearings 2, but this is not necessarily the case, and there is no need for two rotating shafts 3. In FIG. 4, the motor 9 is used as a power source, but a configuration may be adopted in which rotational power is supplied from another power source.

  When such a rotational power transmission mechanism is to be monitored, the load / deflection calculation unit 4c first has motor power H (t) [W] and motor rotational speed N (t) [rev / min at a certain time t. ] Is acquired from the measured value acquisition unit 4b. Then, using the acquired motor power H and motor rotation speed N, torque T (t) [Nm] applied to the rotation shaft 3A is obtained from the following (Equation 3).

Next, by using this torque T (t) and the pitch radius R _A [m] of the gear 10A, which is design information obtained from the storage unit 4f, the tangential force F ₀ [N] applied to the gear 10A is obtained. It is obtained from the following (formula 4).

This tangential force F ₀ [N] is shared between the rolling bearing 2A and the rolling 2B. By using this tangential force F ₀ [N] and the distance l _A from the gear 10A to the rolling bearing 2A, or the design information acquired from the storage unit 4f, or l _B from the rolling bearing 2B, the rolling bearing 2A is used. Each of the 2B bearing shared loads F _A (t) and F _B (t) is obtained from the following (formula 5) and (formula 6).

Since the tangential force generated in the gear 10B is equal to the tangential force F ₀ (t) of the gear 10A because of the relationship between action and reaction, the distance l _C from the gear 10B to the rolling bearing 2C and the distance from the gear 10B to the rolling bearing 2D. When l _D is used, the bearing shared loads F _C (t) and F _D (t) of the rolling bearings 2C and 2D are obtained from the following (Expression 7) and (Expression 8), respectively.

  In step S <b> 2, the load / deflection calculation unit 4 c calculates the deflection angle θ of the rolling bearing 2. A method of calculating the deflection angle θ will be described with reference to FIG.

FIG. 5 is a schematic diagram when the rotating shaft 3A of FIG. 4 is subjected to load deformation. When a tangential force F ₀ is applied to the gear 10A, the deflection angles θ _A (t) and θ _B (t) generated in the rolling bearings 2A and 2B are the Young's moduli of the rotating shaft 3a obtained from the design information of the storage unit 4f. Using E _A [Pa] and the secondary moment of inertia I _A , the following (Equation 9) and (Equation 10) are obtained.

Similarly, the rolling bearing 2C, declination occurs _{2D θ C (t), θ} D (t) also, the Young's modulus _E B of the rotary shaft 3B [Pa] and with a moment of inertia of _{I B,} the following ( It can be obtained from Equation 11) and Equation 12).

In step S3, the load distribution calculation unit 4d performs contact calculation and calculates the maximum local load f _max ′ therein. A method of _{obtaining the} maximum local load f _max ′ will be described with reference to FIG.

FIG. 6 is a diagram showing the bearing load distribution f when the deflection angle θ is generated in the rolling bearing 2. In load distribution calculation unit 4d, it was determined at a load-argument calculator unit 4c, and the bearing shared load _{F A (t) ~F D (} t), and the deflection angle _{θ A (t) ~θ D (} t), stored Using the geometric dimensions of the rolling bearing 2 stored in the portion 4f, the contact load calculation f (t) of the rolling element 2c and the inner ring 2a or the outer ring 2b is calculated for each rolling bearing 2 by contact theory calculation. Based on this bearing load distribution f (t), the maximum local load f _max ′ (t) when there is a declination θ is calculated. Further, the maximum local load f _max (t) when there is no declination θ is calculated at the same time. Further, the maximum local load ratio β (t), which is the ratio thereof, is obtained by the following (formula 13).

In step S4, the fatigue state prediction unit 4e performs load frequency analysis. The load frequency analysis performed here is demonstrated using FIG. When the bearing shared loads F _{A to} F _D are acquired from the load / deflection calculation unit 4c and the rotation speed data (N (t)) is acquired from the measurement value acquisition unit 4b, the load frequency distribution is calculated based on these. Can do. Specifically, the load frequency distribution is obtained by the following (Equation 14) based on the bearing shared load data Fp (t) and the rotational speed data N (t).

Here, v _i is the total number of rotations (cumulative number of rotations) when a predetermined range of load P _i is applied, and Δt is a sampling interval (sampling period).

This will be specifically described with reference to FIG. FIG. 7A is a graph showing the change over time of the bearing shared load data Fp of a certain rolling bearing 2. FIG. 7B is a graph showing a change over time in the rotational speed data N of the rotating machine. Further, FIG. 7C shows a load frequency distribution in which the horizontal axis represents the total number of revolutions (cumulative number of revolutions) N and the vertical axis represents the load P, and the load ΔP applied to the rolling bearing 2. The total number of rotations (cumulative number of rotations) for each is recorded. For example, the total rotational speed to which the load P _i-1 is added is v _i-1 [rotation], the total rotational speed to which the load P _i is added is v _i [rotation], and the total rotational speed to which the load P _{i + 1} is added. Is shown to be v _{i + 1} [rotation].

  FIG. 7 (d) is a graph showing the time change of the maximum local load ratio β of a certain rolling bearing 2, and the damage degree D can be calculated based on FIG. 7 (d) and FIG. 7 (c). However, details of the calculation method will be described later.

  In step S5, the fatigue state prediction unit 4e acquires the lubrication condition and the filter condition from the storage unit 4f via the internal bus 4i.

In step S6, the fatigue state prediction unit 4e calculates a correction coefficient a _ISO based on the acquired lubrication condition and filter condition. As described above, the correction coefficient a _ISO is used when the corrected rated life L _nm is obtained by (Equation 2).

In step S7, the fatigue state prediction unit 4e uses the maximum local load ratio β obtained in step S3, the load frequency distribution obtained in step S4, and the correction coefficient a _ISO calculated in step S6 to determine the damage degree D. Calculated by (Equation 15).

  In Equation 15, the suffix i is a load step, the suffix I is the number of load steps, and the index p is 3 for ball bearings and 10/3 for roller bearings.

  In step S8, the fatigue state prediction unit 4e obtains the current damage probability based on the damage probability n% of the rolling bearing 2 when the cumulative damage degree D calculated in (Equation 15) is 1.

  In step S9, the fatigue state prediction unit 4e acquires the damage amount γ at the time of failure stored in the storage unit 4f via the internal bus 4i. Here, the damage amount γ at the time of occurrence of the failure is, for example, the cost required for replacing the parts of the rolling bearing 2 itself, the cost loss caused by stopping the rotating machine having the rolling bearing for the replacement of the parts, Corresponds to "degree".

  In step S10, the fatigue state prediction unit 4e multiplies the current damage probability obtained in step S8 by the “damage amount γ at the time of failure occurrence” acquired in step S9 to calculate the current risk.

  By executing the above processing, the fatigue state prediction processing by the rolling bearing fatigue state prediction system 1 is completed.

  During execution of the above processing, the storage unit 4f obtains the bearing shared load, declination, rotational speed (N (t)), load frequency distribution, cumulative damage degree D, damage probability, and risk obtained in each step. Data is associated with each time t and stored in a predetermined storage area.

  In FIG. 3, the load / deflection calculation unit 4c executes steps S1 and S2, the load distribution calculation unit 4d executes step S3, and the fatigue state prediction unit 4e executes other steps. Further, the fatigue state prediction unit 4e may perform the processing of all steps.

  Next, an example of a form in which the fatigue state prediction result of the rolling bearing 2 obtained by the process of FIG. 3 is displayed on the display screen 7 of the display device 6 will be described with reference to FIGS.

  As shown in FIG. 8, the display screen 7 of the display device 6 includes a first display area 7 a that displays a fatigue state prediction result of the rolling bearing 2, a second display area 7 b that displays a maintenance-related message, and various commands. It consists of a command input area 7c for inputting. In this example, an “execute” button 7d and a “maintenance” button 7e are displayed in the command input area 7c, but other buttons may be displayed. In the uppermost area of the display screen 7, buttons for closing, reducing / enlarging the entire window including the first display area 7a and the like are also displayed.

  When the operator operates the “execute” button 7d via the input device 5, the “execute” button 7d becomes active as shown in FIG. 8, and the arithmetic unit 4 starts the processing of FIG. The “bearing shared load”, “deflection angle”, “cumulative rotational speed”, “cumulative damage degree”, “damage probability”, and “risk” for each rolling bearing 2 are displayed in a table format in the first display area 7a. Is displayed. In the example of FIG. 8, as a fatigue state prediction result at time t, the magnitude of the bearing shared load is “F (t)”, the declination is “θ (t)”, the cumulative rotational speed is “N (t)”, The cumulative damage degree is displayed as “D (t)”, the damage probability is “〇%”, and the risk is “¥ ○”, and the damage probability for each rolling bearing 2 and the cost when it is damaged are specifically grasped. can do. Thus, the fatigue state (cumulative damage degree, damage probability, risk) of each rolling bearing 2 is predicted and displayed for each time.

  When the operator operates the “maintenance” button 7e via the input device 5, the “maintenance” button 7e becomes active as shown in FIG. 9 and, for example, “bearing C is damaged in the second display area 7b”. Please consider arranging replacement parts and performing maintenance. ”“ From the operation history, the risk of 〇 months later (years later or even days later) is ¥ 00 ”. An advice to the operator is displayed based on the display content of the first display area 7a, such as “It is necessary to change or replace the operation plan”. Note that these messages are stored in advance in the storage unit 4f, and those corresponding to the predicted fatigue state are extracted and stored in the second display area 7b of the display device 6 via the display control unit 4g and the output unit 4h. Is displayed.

  In addition, a plurality of threshold values are set in advance in the “cumulative damage degree” and “damage probability” obtained through the processing of FIG. 3 so that an appropriate message can be selected and displayed. On the other hand, different messages are easy. Accordingly, the display control unit 4g compares the “cumulative damage degree”, “damage probability”, and the like with a plurality of preset thresholds, and thus appropriately displays the “accumulated damage degree”, “damage probability”, and the like. After selecting a message from the storage unit 4f, it can be displayed in the second display area 7b. In addition, it is good also as a structure which replaces with the display control part 4g, and the fatigue state prediction part 4e performs the message selection mentioned above.

  Thus, according to the present embodiment, since the fatigue state (cumulative damage degree, damage probability, risk) for each rolling bearing 2 at each time is displayed in the first display area 7a, the operator can display the screen. The fatigue state of the rolling bearing 2 can be easily confirmed above, and the replacement time of the rolling bearing 2 can be optimized. Moreover, since the message regarding the maintenance corresponding to the fatigue state prediction result of the rolling bearing 2 is displayed in the second display area 7b, even a less skilled operator can recognize an appropriate response, If necessary, maintenance work can be started immediately.

  8 and 9 of the present embodiment, the first display area 7a and the second display area 7b are provided on the display screen 7 of the display device 6. However, the present invention is not limited to this. For example, the display screen 7 is provided with one display area, and the fatigue state prediction result displayed in the first display area 7a in FIG. 8 and the maintenance-related message displayed in the second display area 7b in FIG. 9 are switched and displayed. Also good.

  Furthermore, it is good also as a structure which displays on the display screen 7 only the fatigue state prediction result displayed on the 1st display area 7a shown in FIG. Even in this case, since the fatigue state (cumulative damage degree, damage probability, risk) of each rolling bearing 2 at each time is displayed, the operator can easily check the fatigue state of the rolling bearing on the display screen. Therefore, the replacement time of the rolling bearing can be optimized.

  According to the present embodiment described above, the fatigue state of the rolling bearing can be accurately predicted even when the load or the deflection angle applied to the rolling bearing 2 changes. In addition, according to the present embodiment, the operator can easily confirm the fatigue state of the rolling bearing 2 on the display screen, so that the replacement time of the rolling bearing 2 can be optimized. As a result, the operating rate is improved and the maintenance cost is reduced. Reduction can be realized.

  Next, a method for measuring the deflection angle of the load / deflection angle calculation unit 4c in the rolling bearing fatigue state prediction system according to the second embodiment of the present invention will be described with reference to a shaft-containing sectional view of the rolling bearing 2 in FIG. In addition, duplication description is abbreviate | omitted in common with Example 1. FIG.

  In the first embodiment, the deflection angle θ is calculated using operation data of the rotating electrical machine such as the motor power, the power generation amount, the rotation speed, and the rotation speed, and the fatigue state of the rolling bearing 2 is predicted based on the deflection angle θ. However, in this embodiment, a plurality of non-contact type displacement sensors 21 are provided in the bearing housing 20 of the rolling bearing 2, and the deflection angle θ is calculated using the displacement data output from the displacement sensor 21. Based on this, the fatigue state of the rolling bearing 2 was predicted.

FIG. 10 exemplifies the arrangement of the displacement sensor 21 in the present embodiment. As shown in the right diagram of FIG. 10, the bearing housing 20 a on the front side of the rolling bearing 2 has four non-circular shafts surrounding the rotating shaft 3. Contact type displacement sensors 21 a _{1 to} 21 a ₄ are installed, and four non-contact type displacement sensors 21 b _{1 to} 21 b ₄ surrounding the rotating shaft 3 are installed in the bearing housing 20 b on the back side of the rolling bearing 2. Has been. Among these displacement sensors 21, two pairs of (21a ₁ , 21a ₃ ) and (21b ₁ , 21b ₃ ) are arranged symmetrically so as to sandwich the rotating shaft 3 on the x-axis, (21a ₂ , Two pairs of 21a ₄ ) and (21b ₂ , 21b ₄ ) are arranged symmetrically so as to sandwich the rotating shaft 3 on the y-axis. As a result, as can be seen from the cross-sectional view in the left diagram of FIG. 10, four pairs of (21a ₁ , 21b ₁ ), (21a ₂ , 21b ₂ ), (21a ₃ , 21b ₃ ), (21a ₄ , 21b ₄ ) Are arranged so as to sandwich the rolling bearing 2 in the z-axis direction. As the non-contact type displacement sensor 21, for example, an eddy current type displacement sensor or a laser displacement meter can be used, but other types of non-contact type sensors may be used.

Next, a method for calculating the deflection angle θ using these displacement sensors 21 will be described. When the measured values of the displacement sensors 20a _{1 to} 20a ₄ installed in the bearing housing 20a are M _{a1 to} M _a4 , the change C _a of the center position coordinate of the rotating shaft 3 on the same plane of the bearing housing 20a is expressed by 16).

Similarly, if the measured values of the displacement sensors 20b _{1 to} 20b ₄ installed in the bearing housing 20b are M _{b1 to} M _b4 , the change C _b of the center position coordinate of the rotating shaft 3 on the same plane of the bearing housing 20b is as follows. (Equation 17)

An inclination angle (declination angle θ) in the axial direction of the rotating shaft 3 in the vicinity of the rolling bearing 2 due to the changes C _a and C _b of the center position coordinates of the rotating shaft 3 calculated from (Expression 16) and (Expression 17). Can be obtained. Then, by using the declination θ obtained here, the cumulative damage degree, the damage probability, the damage amount, the risk, and the like are estimated as in the first embodiment by executing the processing after step S3 in FIG. Can do.

  According to the present embodiment described above, since the fatigue state of the rolling bearing is estimated using the directly measured declination θ, the accuracy can be further improved as compared with the configuration of the first embodiment.

  Next, another example of the calculation method in the load distribution calculation unit 4d in the rolling bearing fatigue state prediction system according to the third embodiment of the present invention will be described using the graph of FIG. In addition, duplication description is abbreviate | omitted in common with the above-mentioned Example.

  In the first embodiment, the load distribution calculation unit 4d calculates the bearing load distribution f shown in FIG. 6 by performing contact calculation in Step 3 of FIG. 3, and based on this, the deflection angle θ is generated. The maximum local load fmax ′ and the maximum local load ratio β were obtained. However, in order to execute the contact calculation in real time, it is necessary to use a relatively fast and expensive CPU for the rolling bearing fatigue state prediction system 1. Therefore, in this embodiment, the maximum local load ratio β can be easily obtained by referring to a table (table) recorded in advance in the storage unit 4f.

  FIG. 11 is a diagram showing the relationship between the deflection angle θ and the maximum local load ratio β. This curve is obtained in advance by performing contact calculation, and is stored in the storage unit 4f as tabular data.

  Also in the rolling bearing fatigue state prediction system 1 of the present embodiment, the processing according to the flow of FIG. 3 of the first embodiment is performed, but the processing content of step S3 is different. That is, in this embodiment, when the bearing share load F and the deflection angle θ are input to the load distribution calculation unit 4d, the load distribution calculation unit 4d reads the deflection angle θ and the maximum local load ratio shown in FIG. A table showing the relationship of β is called, the maximum load local load ratio β corresponding to the inputted θ is specified, and this is output. Since the process after step S4 is the same as that of Example 1, the overlapping description is abbreviate | omitted.

According to the present embodiment described above, in the rolling bearing fatigue state prediction system 1, it is not necessary to calculate the contact load distribution every time the load F ₀ or the deflection angle θ changes, so that the calculation load can be reduced. And the fatigue state calculation speed can be increased.

  Next, a specific example in which the rolling bearing fatigue state prediction system 1 according to any one of the first to third embodiments is used for monitoring a rotating machine will be described with reference to the schematic configuration diagram of the wind turbine generator shown in FIG. In addition, duplication description is abbreviate | omitted in common with the above-mentioned Example.

  The wind power generator 30 shown in FIG. 12 includes a plurality of blades 31 that rotate by receiving wind, a hub 32 that supports the plurality of blades 31, a nacelle 33, and a tower 34 that rotatably supports the nacelle 33. The nacelle 33 is connected to a main shaft 35 that is connected to the hub 32 and rotates together with the hub 32, a speed increasing device 36 that increases the rotational speed of the main shaft 35, and a generator shaft 37 that is accelerated by the speed increasing device 36. The generator 38 is provided. When a wind load is applied to the plurality of blades 31, the plurality of blades 31 rotate, and the rotational energy is converted into power generation energy.

  A plurality of rotating shafts 3, gears 10, and rolling bearings 2 as shown in FIG. Therefore, as described in the first embodiment, the load / deflection calculation unit 4c receives the power generation amount, the rotation speed data, and the design information (gear of the gearbox 36) stored in the storage unit 4f from the measurement value acquisition unit 4b. 10, the dimensions of the rolling bearing 2 and the rotating shaft 3, etc.), the bearing shared load F and the deflection angle θ of the rolling bearing 2 used in the speed increaser 36 can be calculated. And the fatigue state of the rolling bearing 2 in the speed-up gear 36 can be estimated by completing the process of FIG.

  Note that the arithmetic device 4 or the like constituting the rolling bearing fatigue state prediction system 1 does not need to be installed in the vicinity of the wind power generator 30, and is installed, for example, at a central power supply command station located remotely from the wind power generator 30. It is desirable. As a result, even if the bearing shared load F and the deflection angle θ applied to the rolling bearing 2 of the speed increaser 36 change, the operator can predict the fatigue state of the rolling bearing 2 while staying at the central power supply command center. Since the result can be easily confirmed, the replacement time of the speed increaser 36 can be appropriately determined, and as a result, the operating rate of the wind power generator 30 can be improved and maintenance costs can be reduced. In the present embodiment in which the rotating machine to be monitored is the wind power generator 30, the “influence” used by the fatigue state prediction unit 4 e is to stop the wind power generator 30 for the gearbox replacement. The power charge (cost) during the outage period is also included.

  In the present embodiment, the wind power generator 30 is exemplified as a rotating machine in which the bearing share load and the deflection angle applied to the rolling bearing change. However, the rotating machine to be monitored is not limited to this, and the construction machine The present invention can be similarly applied to a rotating machine such as. In the case of construction machinery, the above-mentioned “influence” includes penalties due to extension of the construction period due to suspension of construction machinery.

  As described above, according to the present embodiment, in a rotating machine in which the bearing shared load F and the deflection angle θ applied to the rolling bearing change, the fatigue state of the rolling bearing can be predicted with high accuracy, and the operator can Since the fatigue state of the rolling bearing can be easily confirmed, the replacement time of the rolling bearing can be optimized, and as a result, the operating rate of the rotating machine can be improved.

1 Rolling bearing fatigue condition prediction system,
2, 2A, 2B, 2C, 2D rolling bearings,
2a inner ring,
2b outer ring,
2c rolling elements,
3, 3A, 3B rotation axis,
4 arithmetic unit,
4a input section,
4b Measurement value acquisition unit,
4c Load / deflection calculation section,
4d load distribution calculation unit,
4e fatigue state prediction unit,
4f storage unit,
4g display control unit,
4h output section,
4i internal bus,
5 input devices,
6 Display device,
7 Display screen,
7a first display area,
7b Second display area,
7c Command input area,
7d “Execute” button,
7e “Maintenance” button,
9 Motor,
10, 10A, 10B gear,
20, 20a, 20b bearing housing,
21, 21a, 21b displacement sensor,
30 Wind power generator,
31 blades,
32 hubs,
33 Nasser,
34 Tower,
35 spindle,
36 gearbox,
37 generator shaft,
38 generator,
F ₀ tangential force,
F Bearing load sharing,
f Bearing load distribution,
f _max maximum local load without declination,
f _max 'maximum local load with declination,
θ declination,
β Maximum local load ratio

Claims (10)

A rolling bearing fatigue state prediction system for predicting a fatigue state of a rolling bearing that supports a rotating body,
A load / deflection calculation unit for calculating the magnitude and declination of the bearing shared load of the rolling bearing;
A load distribution calculation unit for calculating a load distribution of the rolling bearing based on the bearing shared load and the deflection angle;
A fatigue state prediction unit that predicts a fatigue state of the rolling bearing based on the load distribution;
A rolling bearing fatigue state prediction system characterized by comprising: In the rolling bearing fatigue state prediction system according to claim 1,
The rolling bearing fatigue state prediction system, wherein the load / deflection calculation unit calculates the magnitude of the bearing shared load based on a power generation amount or motor power of a rotary machine including the rotating body. In the rolling bearing fatigue state prediction system according to claim 1,
The rolling bearing fatigue state prediction system, wherein the load / deflection angle calculation unit calculates the deflection angle by calculating deformation of the rotating body due to the bearing shared load. In the rolling bearing fatigue state prediction system according to claim 1,
The rolling bearing fatigue state prediction system, wherein the load / deflection calculation unit calculates the deflection angle based on an output of a displacement sensor installed in the vicinity of the rolling bearing. In the rolling bearing fatigue state prediction system according to claim 1,
The rolling bearing fatigue state prediction system, wherein the load distribution calculating unit calculates the load distribution based on a geometric shape of the rolling bearing and a contact theory. In the rolling bearing fatigue state prediction system according to claim 1,
The rolling bearing fatigue state prediction system, wherein the load distribution calculation unit calculates the load distribution based on a table showing a relationship between a magnitude of the declination and a maximum local load of the load distribution. In the rolling bearing fatigue state prediction system according to claim 1,
The fatigue state prediction unit analyzes the cumulative rotational speed of the rotating body for each maximum local load of the load distribution, and determines the fatigue state based on the analysis result, the bearing rated life, and the linear cumulative damage law. A rolling bearing fatigue state prediction system characterized by calculating. In the rolling bearing fatigue state prediction system according to claim 1,
A rolling bearing fatigue state prediction system, wherein the fatigue state predicted by the fatigue state prediction unit is displayed on a display device. In the rolling bearing fatigue state prediction system according to claim 8,
The fatigue state displayed on the display device is any one of a remaining life of the rolling bearing, a cumulative damage degree, a damage probability, or a risk that is an index obtained by multiplying the damage probability by a loss generated at the time of damage, or A rolling bearing fatigue state prediction system characterized by any combination of these. In the rolling bearing fatigue state prediction system according to claim 8 or 9,
A rolling bearing fatigue state prediction system, characterized in that the maintenance content corresponding to the fatigue state predicted by the fatigue state prediction unit is displayed on the display device.

Images