Method and apparatus for determining a state of wear of a sliding bearing assembly

Abstract

The invention relates to a method (38) and an apparatus (10) for determining, by means of a sensor (14), a state of wear of a sliding bearing assembly (12) having a rotatable component (32), wherein by means of the sensor (14) a signal (44) is generated that is associated with an oscillation excitation which occurs on the sliding bearing assembly (12), and wherein the signal (44) has a plurality of measurement values (46) detected by the sensor (14), wherein the sensor (14) is a piezoelectric acceleration sensor (22) and wherein the piezoelectric acceleration sensor (22) has a oscillatable seismic mass (24), and wherein the signal (44) of the acceleration sensor (22) is evaluated.

Description

[0001] Title: Method and device for determining a wear condition of a plain bearing arrangement

[0002] Description

[0003] The invention relates to a method and a device for determining a wear condition of a plain bearing arrangement with a rotatable component by means of a sensor, wherein the sensor generates a signal which is associated with a vibration excitation occurring in the plain bearing arrangement, and wherein the signal comprises a plurality of measured values ​​detected by the sensor, wherein the sensor is a piezoelectric acceleration sensor and wherein the piezoelectric acceleration sensor has an oscillatable seismic mass, and wherein the signal of the acceleration sensor is evaluated.

[0004] Plain bearing arrangements are used in various industrial sectors. They are also used in large machines, such as wind turbines.

[0005] When reference is made in this application to a plain bearing arrangement with a rotatable component, this means that a component, for example a shaft or a journal, is mounted on a slide bearing with as little friction as possible for rotation or displacement relative to another component, for example a housing component, or is simply supported for sliding movement. For this purpose, a sliding layer, a sliding coating, or a lubricating film is formed between the components or on at least one of the components. A bushing-shaped plain bearing element with a sliding layer can be provided between the components or for supporting one of the components. For example, a plain bearing arrangement can comprise a housing, a bushing-shaped plain bearing element typically firmly inserted therein, and a shaft part mounted therein.However, a plain bearing arrangement can also comprise a machine part, in particular a gear, as the first component, which is rotatably mounted on a shaft or a bolt as the second component. A bush-shaped sliding element or a sliding coating can be provided on the shaft for this purpose.

[0006] The applicant's unpublished PCT / EP2023 / 072718 describes a method and device for monitoring a plain bearing assembly. Starting from a hydrodynamic friction state, a friction state characterized by mixed friction is determined.

[0007] From US 2013 / 0006551 A1, a diagnostic method for detecting wear in a journal bearing of a diesel engine is known, comprising acquiring waveform data representing an acceleration of a vibration occurring when a journal bearing is in operation, transforming the waveform data into a frequency domain acceleration spectrum by applying a Fourier transform to the acceleration waveform data, wherein a plurality of maximum information in the acceleration spectrum occurring in the rotation frequency interval of the shaft is quantified, wherein the quantifying step comprises performing a cepstrum calculation, obtaining time series data of a quefrence value obtained from the waveform data after the cepstrum calculation, and monitoring whether the quefrence value has exceeded a predetermined threshold.

[0008] The present invention is based on the object of providing a method for determining a wear condition and thus for ensuring the continued functionality of a plain bearing arrangement, which can be carried out in a simple, economical and process-reliable manner.

[0009] This object is achieved according to the invention by a method having the features of claim 1. Advantageous embodiments can be found in the subclaims.

[0010] According to the invention, the signal from the acceleration sensor is recorded and evaluated in chronologically successive, in particular temporally spaced, measuring cycles. During each measuring cycle, a rotational speed of the rotatable component comprised by the plain bearing arrangement is varied according to a predetermined or predeterminable rotational speed profile. The signal is evaluated by forming a characteristic value from the signal over each measuring cycle, which is used to determine the state of wear of the plain bearing arrangement. The functionality of the plain bearing arrangement is thus ensured by monitoring the state of wear of the plain bearing arrangement.

[0011] During operation, machine vibrations occur in the plain bearing arrangement, which spread throughout the plain bearing arrangement and also into adjacent components.

[0012] The vibrating seismic mass of the piezoelectric acceleration sensor is excited to vibrate by the vibrations generated during operation of the plain bearing assembly. The piezoelectric effect generates the sensor signal from the vibration of the seismic mass, which characterizes the vibration excitation associated with the plain bearing assembly.

[0013] The sensor signal comprises a multitude of measured values ​​recorded by the sensor. The recorded measured values ​​are acceleration values ​​of the vibrations and oscillations at the location where the sensor is located. These acceleration values ​​depend on the wear condition of the plain bearing assembly. When the plain bearing assembly is only slightly worn, lower acceleration measured values ​​are recorded than when the same plain bearing assembly is more advanced and more severely worn. Consequently, the generated signal changes depending on the wear condition of the plain bearing assembly. The measured values ​​are recorded in measuring cycles, whereby during each measuring cycle, the speed of the rotating component comprised of the plain bearing assembly is varied according to a predetermined or predeterminable speed profile.

[0014] The measurement cycles can be recorded and evaluated in direct succession, i.e., continuously or quasi-continuously, to determine the wear condition of the plain bearing assembly in a virtually continuous manner. The measurement cycles can also be spaced apart in time, thus reducing the amount of data and allowing normal operation of the plain bearing assembly between measurement cycles. The measurement cycles can be spaced evenly or irregularly.

[0015] As mentioned, within a measuring cycle the speed of the rotating component is varied according to a predetermined or specifiable speed profile. It is advantageous if the speed profile is the same across all measuring cycles in order to be able to easily compare the evaluated signals with one another and to draw conclusions about the wear condition of the plain bearing arrangement as precisely as possible. A specifiable speed profile can be provided by specifically controlling the rotating component. A predetermined speed profile can also be used that corresponds to a change in operating mode. An example of this is the startup and / or shutdown of a system. This can be a change in speed starting from a standstill or a low idle speed to a higher operating speed or conversely a reduction in speed starting from an operating speed to a lower idle speed or to a standstill.In this case, the temporal spacing of the successive measurement cycles can be determined by the current operation of a system.

[0016] In a further process step, the signal from each measuring cycle is evaluated. For this purpose, a characteristic value is created from the signal from the acceleration sensor. Creating a characteristic value makes it possible to put the large number of measured values ​​from the signal from a measuring cycle into a more comparable form. This means that several measuring cycles can be compared with one another in tabular or graphical form, for example to identify a trend. The wear condition of the plain bearing arrangement is determined based on the characteristic value. The development of the wear condition of the plain bearing arrangement can be monitored starting from a reference condition of a plain bearing arrangement, e.g. starting from the new condition, by directly comparing the characteristic values ​​of successively evaluated measuring cycles, since the characteristic value increases with increasing wear. A higher characteristic value therefore corresponds to greater wear on the plain bearing arrangement. An estimate of the wear condition of the plain bearing arrangement can, for example,based on empirical values. It is also conceivable that an actual wear value of the plain bearing assembly is determined for a generated characteristic value by interpolating or extrapolating the corresponding wear value with wear values ​​previously determined on a test bench or by measuring a reference bearing assembly. Thus, the characteristic value can also be used to determine an actual wear value and thus the wear condition, because there is a direct relationship between wear and the characteristic value.

[0017] The method according to the invention enables in a simple manner the continued functionality of the

[0018] Determining the wear condition of a plain bearing assembly with a rotating component allows for timely maintenance or replacement before damage occurs. Furthermore, determining the wear condition of plain bearings can optimize maintenance intervals.

[0019] In one embodiment, it has proven advantageous if the speed profile comprises an increase and / or a reduction in the speed of the rotatable component to a predetermined or predeterminable speed. The speed profile therefore has at least one section in which a change in the speed occurs. Furthermore, it is advantageous if the speed profile of the successive measuring cycles is always the same when determining the wear state of a plain bearing arrangement. The speed profile can, for example, comprise a sequence of one or more speed changes. A speed profile can therefore have an increase or a reduction in the speed, several consecutive increases or reductions in the speed, a combination of one or more increases and one or more reductions in the speed, and in some cases also a constant speed range.

[0020] It is further advantageous if the speed profile is limited by a predetermined or predeterminable measurement cycle time, i.e., the duration of the measurement cycle. Because the speed profile is limited by a measurement cycle time, a signal is always recorded over the same period of time. Thus, the characteristic values ​​determined from this, i.e., from the multitude of measured values, are more accurately comparable with each other and only change depending on the acceleration measured values, which change with increasing wear. The method according to the invention is thus more precise.

[0021] According to one embodiment of the invention, it has proven advantageous if the signal from measuring cycles spaced apart in time is recorded and evaluated, and the measuring cycles are spaced apart in time by at least one or more hours, in particular at least one or more days, in particular at least one or more weeks, in particular at most ten weeks. By recording the measuring cycles at spaced intervals in time, the amount of data is reduced. This reduces the hardware requirements for carrying out the method. In practice, it is possible to select a specific, recurring operating point which is approached at spaced times due to the use of a system. Therefore, no measuring operation deviating from normal operation needs to be carried out. Interrupting the operation of the system with the plain bearing arrangement to carry out the measuring operation is not necessary.

[0022] In an advantageous embodiment, the characteristic value of a respective measurement cycle is determined using a median, an a-trimmed mean, a Li norm, or an arithmetic mean. The a-trimmed mean is preferably used because the a-trimmed mean provides the most accurate indication of the wear condition, as demonstrated by test measurements on the test bench. The following describes how the characteristic value is calculated in each case.

[0023] The characteristic value can be calculated using the median, as described above. The median indicates the middle value of a series. The following formulas for calculating the median distinguish between an even number n (n = 2, 4, ...) of measured values ​​Xi of the signal and an odd number n (n = 1, 3, ...) of measured values ​​Xi of the signal:

[0024] 1. Formula 1 , . with n even

[0025] X = - (Xn + Xn ) 2 2 2 +1

[0026] 2. Formula x = x n +i with n odd

[0027] 2

[0028] The respective absolute values ​​of the n measured values ​​Xi of the signal are converted into a series sorted by value magnitude, and x is the median to be calculated (i.e., corresponds to the characteristic value to be calculated). For an even number n of measured values ​​Xi of the signal, the median is calculated by adding and then halving two middle measured values ​​of the sorted series of measured values ​​(1st formula). If, however, n is odd, the middle value of the sorted series of measured values ​​is determined using the 2nd formula.

[0029] The a-trimmed mean (x a ) as a characteristic value is calculated using the following formula: k = [a * nj

[0030] Here, too, the respective values ​​of the n measured values ​​Xi of the signal are converted into a series sorted by value, where n corresponds to the number of measured values ​​Xi and k is a lower bound of the product of the number n and a. When summing the sorted series of measured values ​​Xi, outliers within the signal are sorted out if they belong to the smallest a-percent of the measured values ​​Xi or to the largest a-percent of the measured values ​​Xi. For the calculation of the characteristic value, a can be assumed to be 0.1 (10%), for example. Accordingly, when forming the a-trimmed mean, the upper and lower 10% of the measured values ​​from the series of measured values ​​Xi sorted by value are not taken into account. Here, lower bound means that a decimal value of k is rounded down to the next smallest natural number, which is indicated in the formula for k by the Gaussian bracket. For example, 1.3 or 1.9 are rounded down to 1.

[0031] The Li standard has the following formula for calculating the characteristic value:

[0032] For the Li norm, n equals the number of measured values ​​Xi of the generated signal, and x is the characteristic value calculated using the Li norm. Again, the respective absolute values ​​of the measured values ​​are used. Subtracting 0 from Xi occurs because 0 is used as the reference vector.

[0033] The arithmetic mean is calculated using the following formula to determine the characteristic value:

[0034] Here, n is equal to the number of measured values ​​Xi of the generated signal and x is the characteristic value formed as an arithmetic mean from the respective amounts of the measured values.

[0035] It is also advantageous to determine the wear condition of a plain bearing assembly by determining the wear values ​​associated with at least two characteristic values ​​that are preferably significantly spaced apart from one another. After each characteristic value has been determined, a metrological investigation of the actual wear is carried out to determine the wear values ​​of the plain bearing assembly. It is advantageous to perform the measurement directly on a system; alternatively, the measurement could be performed on a test bench or by measuring a reference bearing assembly. In other words: After the characteristic value for a measurement cycle has been determined, the plain bearing assembly is removed from the system, and the wear value is determined by measurement.This means that at least one wear value can be assigned to each of these two characteristic values ​​and these wear values ​​can be used to interpolate or extrapolate further wear values ​​associated with the respective characteristic values.

[0036] According to an advantageous development of the invention, based on the at least two wear values ​​measured, further wear values ​​associated with the subsequently formed characteristic values ​​can be interpolated or extrapolated. The more wear values ​​measured, the more accurately the wear values ​​associated with the subsequently formed characteristic values ​​can be interpolated or extrapolated, since a larger data basis is available. When replacing a plain bearing assembly in a system with a new, identical plain bearing assembly, the already known characteristic values ​​and associated wear values ​​can still be used. In this case, too, further wear values ​​can be determined by interpolation or extrapolation.

[0037] In a further development, it has proven advantageous that a current characteristic value is only used if it is greater than the previous characteristic values, and otherwise replaced by the previous maximum characteristic value if it is smaller than this previous maximum characteristic value. During operation of the plain bearing arrangement, particles can be introduced into the plain bearing arrangement or become detached from one of the components as a result of wear. The particles have an abrasive or wearing effect on the plain bearing arrangement and also lead to measurably higher vibrations. The higher vibrations result in larger amplitudes in the generated signal. A characteristic value calculated using this measuring cycle can be significantly greater than the previously calculated characteristic values. If the particles are carried out of the plain bearing arrangement again over time, a subsequent characteristic value can be smaller than the previous characteristic values.To take into account the abrasive effect of the particles on the wear condition of the plain bearing assembly, a calculated characteristic value is only replaced by a subsequent one if this subsequent characteristic value exceeds the previous maximum characteristic value. Thus, the method depicts the wear condition of the plain bearing assembly even more precisely by taking into account particle ingress into the plain bearing assembly, even if they have been removed again.

[0038] In a further development, it has proven advantageous to specify a maximum permissible value for the characteristic value and to issue a warning signal if this value is exceeded. Continuous "manual" monitoring of the plain bearing assembly can be eliminated. If a critical wear level of the plain bearing assembly is exceeded, a warning signal is issued, and the plain bearing assembly can be serviced or completely replaced. This optimizes system downtime for repairs, failures, or maintenance. Damage to the plain bearing assembly or surrounding components is also reduced. It has proven advantageous for the piezoelectric acceleration sensor to have a resonant frequency, and for the resonant frequency to be in the range of 20 to 70 kHz in the sensor's installation situation.

[0039] The invention also relates to a device for determining a wear condition of a plain bearing arrangement with a rotatable component, the device comprising a sensor which is mechanically coupled or can be mechanically coupled to the plain bearing arrangement, wherein the sensor generates a signal which is associated with a vibration excitation occurring in the plain bearing arrangement, and wherein the signal comprises a plurality of measured values ​​recorded by the sensor, and wherein the device further comprises an electronic evaluation device with a processor and a memory with a computer program code stored thereon, wherein the sensor is a piezoelectric acceleration sensor and the piezoelectric acceleration sensor has an oscillatable seismic mass, wherein the seismic mass is excited to oscillate by operation of the plain bearing arrangement and the signal is thereby generated,and wherein the memory and the computer program code, in cooperation with the processor, are configured such that the signal from the acceleration sensor is recorded and evaluated in temporally successive, in particular temporally spaced, measuring cycles, and wherein, during a respective measuring cycle, a rotational speed of the rotatable component comprised by the plain bearing arrangement is varied according to a predetermined or predeterminable rotational speed profile, and wherein the signal is evaluated by forming a characteristic value from the signal over a respective measuring cycle, based on which the wear condition of the plain bearing arrangement is deduced. The device is thus designed to carry out the method according to the invention.

[0040] According to a further embodiment, it has proven advantageous if the speed profile comprises increasing and / or reducing the speed to a predetermined or predeterminable speed.

[0041] It has proven advantageous if the processor or a control device is configured to limit the speed profile to a predetermined or predeterminable measuring cycle time. The measuring cycle time corresponds to the duration of a measuring cycle and thus specifies the time over which measured values ​​are recorded. In a further embodiment, it has proven advantageous if the computer program code is designed such that the signal from temporally spaced measuring cycles can be recorded and evaluated, and the measuring cycles are spaced apart by at least one or more hours, in particular at least one or more days, in particular at least one or more weeks, in particular at most ten weeks.

[0042] According to a further embodiment of the invention, it has proven advantageous if the computer program code is designed to determine the characteristic value of a respective measuring cycle by means of a median, an a-trimmed mean, an Li norm or an arithmetic mean.

[0043] It is further advantageous if the computer program code is designed to interpolate or extrapolate further wear values ​​associated with the characteristic values ​​formed below, based on at least two characteristic values ​​and associated wear values ​​determined therefor

[0044] In a further development, it is provided that the computer program code is designed to use a current characteristic value only if it is greater than the previous characteristic values, and to replace it with the previous maximum characteristic value if it is smaller than this previous maximum characteristic value.

[0045] In a further embodiment, it is provided that the computer program code is designed to output a warning signal when a definable or defined maximum permissible value for the characteristic value is exceeded, wherein the device has a warning device.

[0046] In a further embodiment, it is provided that the piezoelectric

[0047] acceleration sensor has a resonance frequency, and that the resonance frequency is in a

[0048] The sensor's installation situation is in a range of 20 to 70 kHz. Further features and advantages of the invention are the subject of the following description and the drawings of exemplary embodiments. They show:

[0049] Fig. 1 is a schematic and partially sectioned view of a plain bearing assembly with a rotatable component and a device for determining a wear condition of the plain bearing assembly;

[0050] Fig. 2 is a schematic representation of steps of a method according to the invention for determining a wear condition of a plain bearing arrangement with a rotatable component;

[0051] Fig. 3 shows a diagram showing speed profiles;

[0052] Fig. 4 shows a model representation of the relationship between the multitude of

[0053] Measured values ​​and formed parameters;

[0054] Fig. 5 a model representation of two different characteristic value curves; and

[0055] Fig. 6 a model superimposed view of a characteristic value curve with outliers and a corrected characteristic value curve taking the outliers into account.

[0056] In the following, functionally equivalent elements and areas in different figures and in different embodiments bear the same reference symbols. They are normally only described in detail when first mentioned. Furthermore, for the sake of simplicity, not all reference symbols are shown in all figures.

[0057] Fig. 1 schematically shows a device 10 for determining the wear condition of a plain bearing assembly 12. The device 10 comprises a sensor 14, an electronic evaluation device 16 with a processor 18, and a memory 20 with a computer program stored thereon. The sensor 14 is designed as a piezoelectric acceleration sensor 22 with an oscillating seismic mass 24. The oscillating seismic mass 24 is arranged above a piezoelectric crystal 26 made of quartz in the figure. The acceleration sensor 22 is designed by means of the oscillating seismic mass 24 such that a resonant frequency of the acceleration sensor 22 in an installed situation is preferably in a range of 20 to 70 kHz.The evaluation device 16 is connected - wired or wirelessly - to the acceleration sensor 22 and can, for example, additionally be connected to an external display device 28 - e.g. mobile device, indicator light, screen - which is also spatially spaced from the acceleration sensor 22.

[0058] The plain bearing arrangement 12, which is only indicated schematically, comprises, by way of example, a schematically indicated plain bearing housing 30 and a component 32 rotatably mounted in the plain bearing housing 30, wherein further sliding elements or sliding coatings that may act between the components are not shown. In the present case, the rotatably mounted component 32 is designed as a shaft (hereinafter referred to as shaft 32). The plain bearing housing 30 thus forms a component 33 that rotatably supports the shaft 32. A lubricating film 34 for hydrodynamically supporting the shaft 32 is typically provided between the plain bearing housing 30 and the shaft 32. This lubricating film 34 is formed using means known per se, which is therefore not described. The acceleration sensor 22 is arranged on an outer side 36 of the plain bearing housing 30 and can, for example, be screwed to the plain bearing housing 30.

[0059] The device 10 is designed to carry out a method according to the invention as described above. Figure 2 illustrates an embodiment of the method 38 with the individual method steps.

[0060] During operation of the plain bearing assembly 12, i.e., when the shaft 32 is set in rotation, or generally when any component of the plain bearing assembly 12 is set in sliding motion relative to the other component, machine vibrations arise in the plain bearing assembly 12, which propagate within the plain bearing assembly 12 and also into components adjacent to the plain bearing assembly 12. The oscillatable seismic mass 24 of the acceleration sensor 22 is excited to oscillate by the vibrations arising during operation of the plain bearing assembly 12.

[0061] In the present case, the memory 20 and the computer program code are configured in cooperation with the processor 18 to execute the method 38 shown in Fig. 2 for determining a wear condition of the plain bearing assembly 12. The method 38 begins with a starting block 40. The starting block 40 can, for example, coincide with the installation of a plain bearing assembly 12 or the initial commissioning of a system with the plain bearing assembly 12. In a first method step 42, a signal 44 is generated (Fig. 4) by exciting the oscillatable seismic mass 24 of the acceleration sensor 22 and the piezoelectric effect, which signal characterizes the vibration excitation associated with the plain bearing assembly 12.The signal 44 comprises a plurality of measured values ​​46 recorded by the acceleration sensor 22, wherein the signal 44 is recorded over at least one measuring cycle 48 or is generated only during a measuring cycle 48 - i.e., the acceleration sensor 22 is operated only during the measuring cycle 48 (see Fig. 4). During the measuring cycle 48, a rotational speed 49 of the shaft 32 is varied according to a predetermined or predeterminable rotational speed profile 50 (Fig. 3). The rotational speed profile 50 can comprise an increase and / or a reduction of the rotational speed 49 to a predetermined or predeterminable rotational speed 49. By varying the rotational speed profile 50, vibrations detectable by the acceleration sensor 22 are excited more strongly than with a rotational speed profile 50 with a constant rotational speed 49 (see Fig. 3).

[0062] In a second method step 52, the signal 44 of the acceleration sensor 22 of a measuring cycle 48 is evaluated by forming a characteristic value 54 from the signal 44 over the measuring cycle 48, based on which the current wear state of the plain bearing assembly 12 and, in this case, even the actual wear value is deduced. Typically, the signal 44 is evaluated directly after a measuring cycle 48. However, signals 44 from multiple measuring cycles 48 can also be generated and only evaluated later (i.e., not directly after a measuring cycle 48). The wear value of the plain bearing assembly 12 is deduced from the characteristic value 54 by interpolating or extrapolating the associated wear value with wear values ​​previously determined on a test bench or by a measurement on a comparison bearing assembly and corresponding to the respective characteristic values ​​54.It is also conceivable that the associated wear values ​​are determined for at least two characteristic values ​​54 by subsequently conducting a metrological investigation of the wear of the plain bearing arrangement 12 after each characteristic value 54 has been formed, and further wear values ​​associated with the subsequently formed characteristic values ​​54 are interpolated or extrapolated. For this purpose, the plain bearing arrangement 12 can be removed after at least two characteristic values ​​54 have been formed, and the wear can be measured, or characteristic value / wear value curves previously determined on a test bench or using an identical plain bearing arrangement in the same installation situation can be used as a basis in order to determine the wear state of the plain bearing arrangement based on a currently formed characteristic value.

[0063] In a third method step 56, a check is performed to determine whether the currently formed characteristic value 54a is greater than the preceding characteristic values ​​54b (characteristic values ​​are also identified overall by reference numeral 54). If the current characteristic value 54a is greater than the preceding characteristic values ​​54b, then the current characteristic value 54a is used in the fourth method step 58a. If the current characteristic value 54a is smaller than the preceding maximum characteristic value 54b, then in an alternative fourth method step 58b, the current characteristic value 54a is replaced by the preceding maximum characteristic value 54b, and in this case, the method 38 begins again with the first method step 42 (see also Fig. 6).

[0064] A fifth method step 60 is executed if the current characteristic value 54a is greater than the previous characteristic values ​​54b. Here, the current characteristic value 54a is compared with a maximum permissible value for the characteristic value 54. If the current characteristic value 54a exceeds the maximum permissible value, a warning signal is output in a sixth method step 62, and the method 38 is terminated, which is symbolized in Fig. 2 by a stop block 64. The warning signal can be output, for example, via the display device 28, which then also functions as a warning device. Following the warning signal, the plain bearing arrangement 12 is preferably replaced or serviced as soon as possible.

[0065] If it was determined in the fifth method step 60 that the current characteristic value 54a is smaller than the maximum permissible value, the method 38 starts again with the first method step 42.

[0066] In an embodiment of the method according to the invention that is an alternative to Fig. 2 and not shown here, after the second method step 52, it can be checked whether the determined characteristic value exceeds a maximum permissible value for the characteristic value. If the characteristic value exceeds the maximum permissible value, a warning signal is issued and the method is terminated; otherwise, the method can start again from the beginning with the first method step. In Fig. 3, a diagram 66 shows three possible speed profiles 50 (also referred to as first speed profile 50a, second speed profile 50b, and third speed profile 50c) for the measuring cycle 48 by way of example. In the diagram 66 of Fig. 3, the speed 49 in rpm is plotted against time in seconds. All three speed profiles 50 shown here as examples are limited to a measuring cycle time 68 of 60 seconds. The measuring cycle time 68 represents the duration of the measuring cycle 48 and limits the speed profile 50.The first speed profile 50a (shown with a dotted line) begins with shaft 32 at a standstill. The speed 49 is increased linearly to 400 rpm within 40 seconds and then held constant at 400 rpm. The first speed profile 50a can represent the start-up of a stationary system.

[0067] The second speed profile 50b (shown with a dashed line) initially has a speed 49 of 600 rpm. After a measurement cycle time 68 of 25 seconds has elapsed, the speed 49 is reduced linearly until the shaft 32 comes to a standstill. This could correspond to a system shutdown.

[0068] The third speed profile 50c (shown with a solid line) is a combined speed profile 50 consisting of reducing the speed 49, increasing the speed 49, and keeping the speed 49 constant. At the beginning of the measuring cycle 48, the speed 49 is reduced linearly from 500 rpm to standstill within 5 seconds, and then immediately increased linearly to 500 rpm within 30 seconds. The speed 49 is then kept constant for the remaining measuring cycle time 68.

[0069] Referring to Fig. 4, the second method step 52 is explained in detail. The evaluation of a signal 44 of a measuring cycle 48 is shown in a model in Fig. 4. In Fig. 4, on the left side, in a diagram 70, the recorded signal 44 of six measuring cycles 48 is shown in a model (one measuring cycle 48 is highlighted by a black border 72 as an example), wherein the plurality of measured values ​​46 of a signal 44 are shown in the diagram 70 as acceleration in m / s 2are plotted against the cycle numbers. In a diagram 74 shown on the right in Fig. 4, the characteristic values ​​54 of the respective measuring cycles 48 formed from the signal 44 are also plotted against the cycle numbers and shown as a continuous characteristic value curve 76. The left-hand diagram 70 shows the absolute value of the measured values ​​46 of the respective measuring cycles 48. Thus, no negative acceleration values ​​are entered in the diagram 70. Across the measured values ​​46 of the respective measuring cycles 48 shown in the absolute value, an increase in the measured values ​​46 (acceleration values) can be seen. In other words: The vibrations that arise during operation of the plain bearing arrangement 12 increase between the measuring cycles 48 shown here, which is due to an increase in wear in the plain bearing arrangement 12. A characteristic value 54 is then formed from the respective amounts of the plurality of measured values ​​46 of the signal 44 of the respective measuring cycles 48.The characteristic values ​​54 plotted in the diagram 74 shown on the right increase with increasing cycle number. This reflects the increase in the measured values ​​46 over the measurement cycles 48 in the left-hand diagram 70 of Fig. 4. The characteristic value 54 can be calculated using various mathematical methods. Preferably, the characteristic value 54 is calculated as an a-trimmed mean.

[0070] For further illustration, Fig. 5 shows a diagram 78 depicting two characteristic value curves 76 (also referred to as the first characteristic value curve 76a and the second characteristic value curve 76b). The characteristic values ​​54 are plotted against the cycle number. The first characteristic value curve 76a, depicted with a solid line, corresponds to characteristic values ​​54 formed from test bench measurements from a plurality of essentially continuously executed measuring cycles 48, with the characteristic value curve 76a appearing to rise steadily in this case due to the experiment. The increase in the characteristic values ​​54 between the measuring cycles 48 is attributable to a deterioration in the wear condition of the plain bearing arrangement 12. In practice, it is advantageous to perform the measuring cycles 48 at repetitive operating points, e.g., during startup and / or shutdown of a system with the plain bearing arrangement 12.This avoids having to carry out an extra measuring operation on the system in addition to normal operation. Successive measuring cycles 48 are thus spaced apart in time. The measuring cycles 48 can be spaced apart in time by at least one or more hours, in particular at least one or more days, in particular at least one or more weeks, in particular at most ten weeks. The spacing of the measuring cycles 48 is modeled by the second step-shaped, dotted characteristic value curve 76b in Fig. 5. It should be noted that each step increase results from a characteristic value 54 of a subsequent measuring cycle 48 which is significantly spaced apart in time. Here, too, the increase in the characteristic values ​​54 between the measuring cycles 48 is due to a deterioration in the wear condition of the plain bearing arrangement 12.Because the measuring cycles 48 are not carried out directly after one another, but are spaced apart in time, the increase in wear of the plain bearing arrangement 12 can be perceived as a jump or as a step increase in the second characteristic curve 76b.

[0071] The third and fourth method steps 56, 58a / b are explained in detail with reference to Fig. 6. In Fig. 6, a superimposed view 80 shows, by way of example, a third characteristic value curve 76c from formed characteristic values ​​54 of a continuous test bench measurement and a fourth corrected characteristic value curve 76d. The third characteristic value curve 76c is formed from measurement cycles 48 that are continuously recorded and evaluated one after the other, wherein some of the formed characteristic values ​​54 are outlier characteristic values ​​54c. The outlier characteristic values ​​54c are significantly larger than the neighboring characteristic values ​​54. These outlier characteristic values ​​54c can arise from particles being introduced between the rotatable shaft 32 and the plain bearing housing 30 or from becoming detached from one of the components, e.g., the shaft 32 or the plain bearing housing 30, as a result of wear during operation.The particles in the plain bearing assembly 12 have a wear-related effect on the plain bearing assembly 12 and lead to measurably higher vibrations, which can be recorded during a measurement cycle 48 and which result in the runaway characteristic values ​​54c. If the particles are removed from the plain bearing assembly 12 over time, one of the subsequent characteristic values ​​54 may again be smaller than the preceding runaway characteristic values ​​54c. In the third method step 56, the runaway characteristic values ​​54c are taken into account. If, as described above, the currently generated characteristic value 54a is greater than the preceding characteristic values ​​54b, then the current characteristic value 54a is used in the fourth method step 58a.However, if the current characteristic value 54a is smaller than the preceding maximum characteristic value 54b, then in the alternative fourth method step 58b the current characteristic value 54a is replaced by the preceding maximum characteristic value 54b, which may be an outlier characteristic value 54c. The fourth corrected characteristic value curve 76d depicts the third and fourth method steps 56, 58a / b. In the superimposed view 80 of Fig. 6, it can be clearly seen that the fourth corrected characteristic value curve 76d increases, in particular abruptly, at the outlier characteristic values ​​54c and is then held constant until another, even larger characteristic value 54 is determined, which then forms the new maximum characteristic value 54. This results in a step-like corrected characteristic value curve 76d that follows the maximum characteristic values ​​54.

Claims

Patent claims 1. A method (38) for determining a wear condition of a plain bearing arrangement (12) with a rotatable component (32) by means of a sensor (14), wherein the sensor (14) generates a signal (44) associated with a vibration excitation occurring in the plain bearing arrangement (12), and wherein the signal (44) comprises a plurality of measured values (46) detected by the sensor (14), wherein the sensor (14) is a piezoelectric acceleration sensor (22) and wherein the piezoelectric acceleration sensor (22) has an oscillatable seismic mass (24), and wherein the signal (44) of the acceleration sensor (22) is evaluated, characterized in that the signal (44) of the acceleration sensor (22) is detected and evaluated in temporally successive, in particular temporally spaced-apart, measuring cycles (48),and that during a respective measuring cycle (48), a rotational speed (49) of the rotatable component (32) encompassed by the plain bearing arrangement (12) is varied in accordance with a predetermined or predeterminable rotational speed profile (50), and that the signal (44) is evaluated by forming a characteristic value (54) from the signal (44) over a respective measuring cycle (48), on the basis of which the wear condition of the plain bearing arrangement (12) is deduced.

2. Method (38) according to claim 1, characterized in that the speed profile (50) comprises increasing and / or reducing the speed (49) of the rotatable component (32) to a predetermined or predeterminable speed (49).

3. Method (38) according to claim 1 or 2, characterized in that the speed profile (50) is limited by a predetermined or predeterminable measuring cycle time (68).

4. Method (38) according to one of the preceding claims, characterized in that the signal (44) is detected by measuring cycles (48) spaced apart in time and is evaluated, and a temporal spacing of the measuring cycles (48) is at least one or more hours, in particular at least one or more days, in particular at least one or more weeks, in particular at most ten weeks.

5. Method (38) according to one of the preceding claims, characterized in that the characteristic value (54) of a respective measuring cycle (48) is determined by means of a median, an a-trimmed mean, an Li norm or an arithmetic mean.

6. Method (38) according to one of the preceding claims, characterized in that, in order to determine a wear state of a plain bearing arrangement (12), associated wear values are determined for at least two characteristic values (54) that are preferably clearly spaced apart from one another, in that, after the characteristic value (54) has been formed, a metrological investigation of the actual wear is subsequently carried out to determine the wear values of the plain bearing arrangement (12).

7. Method (38) according to claim 6, characterized in that, on the basis of the at least two wear values measured, further wear values associated with the characteristic values (54) formed subsequently are interpolated or extrapolated.

8. Method (38) according to one of the preceding claims, characterized in that a current characteristic value (54a) is used only if it is greater than the preceding characteristic values (54b), and otherwise is replaced by the preceding maximum characteristic value (54b) if it is smaller than this preceding maximum characteristic value (54b).

9. Method (38) according to one of the preceding claims, characterized in that a maximum permissible value is defined for the characteristic value (54) and a warning signal is issued when this value is exceeded.

10. Method (38) according to one of the preceding claims, characterized in that the piezoelectric acceleration sensor (22) has a resonance frequency, and that the resonance frequency in an installation situation of the sensor (14) is in a range of 20 to 70 kHz.

11. A device (10) for determining a wear condition of a plain bearing arrangement (12) with a rotatable component (32) and for carrying out the method (38) according to one or more of the preceding claims, the device (10) comprising a sensor (14) that is mechanically coupled or mechanically coupleable to the plain bearing arrangement (12), wherein the sensor (14) generates a signal (44) associated with a vibration excitation occurring in the plain bearing arrangement (12), and wherein the signal (44) comprises a plurality of measured values (46) detected by the sensor (14), and wherein the device (10) further comprises an electronic evaluation device (16) with a processor (18) and a memory (20) with a computer program code stored thereon, wherein the sensor (14) is a piezoelectric acceleration sensor (22) and the piezoelectric acceleration sensor (22) has an oscillatable seismic mass (24),wherein the seismic mass (24) is excited to vibrate by operation of the plain bearing arrangement (12), thereby generating the signal (44), and wherein the memory (20) and the computer program code, in cooperation with the processor (18), are configured such that the signal (44) of the acceleration sensor (22) is detected and evaluated in temporally successive, in particular temporally spaced-apart, measuring cycles (48), and wherein, during a respective measuring cycle (48), a rotational speed (49) of the rotatable component (32) encompassed by the plain bearing arrangement (12) is varied according to a predetermined or predeterminable rotational speed profile (50), and wherein the signal (44) is evaluated by forming a characteristic value (54) from the signal (44) over a respective measuring cycle (48), on the basis of which characteristic value the wear state of the plain bearing arrangement (12) is inferred.

12. Device (10) according to claim 11, characterized in that the speed profile (50) comprises increasing and / or reducing the speed (49) to a predetermined or predeterminable speed (49).

13. Device (10) according to claim 11 or 12, characterized in that the processor (18) or a control device is configured to limit the speed profile (50) to a predetermined or predeterminable measuring cycle time (68).

14. Device (10) according to one of the preceding claims 11 to 13, characterized in that the computer program code is designed such that the signal (44) can be detected and evaluated by measuring cycles (48) spaced apart in time from one another, and a temporal spacing of the measuring cycles (48) is at least one or more hours, in particular at least one or more days, in particular at least one or more weeks, in particular at most ten weeks.

15. Device (10) according to one of the preceding claims 11 to 14, characterized in that the computer program code is designed to determine the characteristic value (54) of a respective measuring cycle (48) by means of a median, an a-trimmed mean, an Li norm or an arithmetic mean.

16. Device (10) according to one of the preceding claims 11 to 15, characterized in that the computer program code is designed to interpolate or extrapolate, on the basis of at least two characteristic values (54) and associated wear values determined therefor, further wear values associated with the characteristic values (54) formed subsequently.

17. Device (10) according to one of the preceding claims 11 to 16, characterized in that the computer program code is designed to use a current characteristic value (54a) only if it is greater than the preceding characteristic values (54b), and to replace it with the preceding maximum characteristic value (54b) if it is smaller than this preceding maximum characteristic value (54b).

18. Device (10) according to one of the preceding claims 11 to 17, characterized in that the computer program code is designed, when exceeding a definable or fixed maximum permissible value for the characteristic value (54) to emit a warning signal, wherein the device (10) has a warning device (28).

19. Device (10) according to one of the preceding claims 11 to 18, characterized in that the piezoelectric acceleration sensor (22) has a resonance frequency, and that the resonance frequency in an installation situation of the Sensor (14) is in a range of 20 to 70 kHz.

Images