METHOD FOR AUTOMATIC DETECTION OF WEAR, GEAR DAMAGE AND / OR BEARING DAMAGE ON A GEARBOX

Description

[0001] Due to the increasing autonomization in the automotive and manufacturing industries, automatic condition monitoring of individual components and assemblies is becoming increasingly important.

[0002] An additional digital fallback level is becoming increasingly important, especially for safety-relevant components and assemblies. A process for timely detection before damage that endangers function is therefore of utmost importance.

[0003] Such a method is known, for example, from DE 10 2016 222 660 A1, in which the following process steps are carried out to detect damage and / or wear on translationally moving parts of an electromechanical drive: Measuring the electrical current supplied to the electrical drive, performing a time-frequency analysis, comparing the frequency spectrum of the measured electrical current determined from the time-frequency analysis with a predetermined frequency spectrum, and triggering an error signal if a deviation of a predetermined size is detected from the comparison of the two frequency spectra.

[0004] Furthermore, US 4,965,513 A describes a method for monitoring the operating conditions of an electrically driven valve by analyzing the motor current. For this purpose, various frequency analysis methods are applied to the motor current to create a noise signature of the motor current, which is intended to detect wear and abnormal operating conditions. Based on the noise signature, various characteristic operating conditions of the electrically driven valve should be detectable; in particular, the sum of all mechanical load changes that manifest themselves in the frequency spectrum and amplitudes should be detectable. If such noise signatures are created at different time intervals during operation, aging and wear or abnormal operating conditions should be detectable.

[0005] These prior art methods suffer from the fact that the processing of the motor current signals is insufficient to reliably predict wear and damage under varying environmental influences and / or operating conditions of the motor-gearbox unit, which consists of an electric motor and a gearbox connected to the same on the output side. The approach described in EP 4 012 426 A1 addresses these problems, but there is still room for improvement with regard to resource utilization and the early detection of damage.

[0006] Furthermore, US 8 180 143 B2 discloses a method and system for evaluating contact patterns of a transmission.

[0007] The objects of the invention are therefore to provide a method for the automatic detection of gearing damage on a transmission, which enables improved, resource-saving and early damage detection, which at the same time is largely independent of varying environmental influences and / or operating conditions of the engine-transmission unit.

[0008] In the method according to the invention and defined in claim 1 for detecting wear and / or damage to a motor-gearbox unit comprising a motor and a gearbox coupled to the same on the output side, a detection module determines an input and / or output torque either directly—for example, using a torque measuring shaft—or indirectly—for example, by measuring the motor current(s)—as a measured variable that can be clearly correlated with the input and / or output torque when the gearbox is moved by the motor. In this way, a signal curve correlating to the torque is generated for each such movement, which represents the directly or indirectly determined input and / or output torque as a function of a variable correlated with the position of the gearbox.

[0009] The term "gearbox position" refers to the changing orientation and / or spatial position of the gearbox components as the gearbox moves. The position of a gearbox can therefore be described, for example, by which teeth of the gears forming the gearbox are currently meshing with each other or, for example, by the angle of rotation of a gearbox relative to a reference point.

[0010] For the sake of simplicity, the term "gearbox damage" is used below as a collective term for wear and damage to gears or gear toothing damage as well as bearing damage.

[0011] Furthermore, for the sake of clarity, the previously described torque-correlating signal curve, which represents the directly or indirectly determined input and / or output torque as a function of a variable correlated with the position of the gearbox, is hereinafter referred to simply as "signal curve" or "signal curves".

[0012] Furthermore, according to the method according to the invention, at least some of these signal curves are analyzed automatically. During this automated analysis, a range of the variable correlated with the position of the transmission is identified in each case, in which a specified characteristic characteristic of transmission damage occurs.

[0013] If there is a gearbox failure, this regularly affects the torque behavior of the engine, which, for example, increases the input torque through appropriate control due to a temporarily inefficient torque transmission in order to ensure the required output torque.

[0014] In the simplest case, such an analysis can simply consist of determining a maximum value in a signal waveform, determining an area under a subsection of a signal waveform, or determining a point in the signal waveform where it most closely resembles a given curve shape. In more complex cases, a pre-selection or pre-processing of signal waveforms can also be performed prior to applying the respective criterion.

[0015] In each signal waveform analyzed, an area (which may, for example, correspond to an angular position of a specific gear in the transmission) is identified that meets the criterion used for damage detection (e.g., maximum in the signal waveform).

[0016] This identified range is automatically assigned to a class of a histogram (in other words, entered into this class so that the number of events or ranges assigned to this class increases, which can be done in particular by immediate summation; but also, for example, by maintaining a respective list of the signal curves assigned to a class). An automated check is carried out to determine whether there is an unexpected accumulation of entries in a class in the histogram (in particular by comparing summed event numbers or determining and comparing list lengths). If this is the case, there is indeed transmission damage. In this case, a warning message can be generated, for example, so that the damage can be remedied before a total failure of the engine-transmission unit occurs.

[0017] A first key finding underlying the invention is that it is possible to apply a criterion for damage detection to signal curves of engine-transmission units that do not exhibit transmission damage. This then generally leads to essentially randomly distributed results for the position at which this criterion locates the transmission damage in the respective signal curve. Even if, in an exceptional case for a specific application, no random distribution of these supposed transmission damages occurs, the expected result distribution for an undamaged transmission can be easily determined and used as a basis for damage monitoring, for example, by applying a weighted background correction.

[0018] However, when gear failure begins to occur, the change in the signal waveform caused by this, which increases with the severity of the damage, in the region or regions of the signal waveform where, for example, a damaged tooth is loaded for torque transmission, will cause the expected result distribution to change significantly.

[0019] A second important finding is that it is usually not necessary to determine the exact position or type of transmission damage, but rather it is only important to recognize that there is transmission damage at some position, since the resulting need for a transmission replacement is independent of which component of the transmission is damaged.

[0020] This allows a histogram-based approach in which the path / angle traversed during the recording of the signal curve is divided into ranges, and for each of these ranges, the number of times the damage detection criterion was met in this range is counted. For this purpose, a memory of a few kilobytes is sufficient, in which only very simple calculations need to be performed. The evaluation of whether real transmission damage is present, which is manifested by the deviation of the distribution in this histogram from an expected distribution, in particular a random distribution, can also be mathematically traced back to simple comparison operations. Therefore, the required computing power is just as low as the required storage space by using the method according to the invention.

[0021] In a preferred embodiment of the method, an automated preselection of the signal curves to be analyzed takes place.

[0022] In particular, it can be provided that during the preselection only signal curves that contain torque information on gear positions that completely cover a specified range are permitted for further evaluation.

[0023] Alternatively or additionally, the pre-selection can be carried out in such a way that only signal waveforms are considered where the deviation of the absolute maximum and the absolute minimum from the mean value of the signal waveform does not exceed a specified limit. The purpose of this measure becomes clear when considering extreme load changes. Short-term torque peaks, especially at a reversal of the direction of rotation and / or load, make it difficult to reliably determine characteristics that indicate damage.

[0024] In a preferred variant of the method, a DC component is determined and removed from the signal waveforms to be analyzed, which enables the analysis of the signal waveform. This determination can be performed, for example, by calculating a moving average of the signal waveform to smooth the signal waveform; the DC component can then be removed by subtracting this moving average from the corresponding measured value of the signal waveform.

[0025] If the smoothed signal waveform is also divided by the signal, correction of torque-dependent periodically occurring amplitudes can be made possible. Additional smoothing and / or low-pass filtering can also eliminate any interference signals from the signal waveform.

[0026] An alternative approach involves determining and removing the DC component by applying a bandpass filter to the signal. One advantage of this method is that it requires relatively little computing power, which facilitates its implementation under automotive standards.

[0027] Preferably, the upper and lower frequency limits of the bandpass filter are selected as a function of the input speed so that the meshing frequency of the gearbox components to be monitored represents the main part of the periodic fluctuations of the signal.

[0028] Normalizing the signal curve with the determined DC component can further improve the reliability of the analysis.

[0029] Preferably, the histogram is stored in the form of a numerical array that stores, for each range of the quantity correlated with the transmission position, the number ni of signal curves whose analysis assigned the characteristic feature of transmission failure to this range, as well as the number of rollovers ri in this angular range. Here, i indicates the respective angular range. This represents an efficient storage method that requires little permanent storage space.

[0030] Since the number of measurement cycles increases, the numbers in the array can become very large and thus require more storage space, and since there are angular ranges that are rarely crossed, a correction or a kind of normalization of the damage events to the number of crossings is preferably performed. The following rule is used to determine the corrected values ​​n i-new and r i-new within the number array: 1 . n i − neu = n i − n ¯ ⋅ r i / r sup 2 . r i − neu = r i − c

[0031] Here, n represents the mean value of all entries ni. The maximum number of rollovers in the entire array is denoted by r sup. Any negative n i-new values ​​that may arise are set to a constant, positive value. Furthermore, possible decimal places are removed. In addition, to limit the decimal places of the ri after the cell-wise application of equation (1) 1. in step 2. of equation (1) a constant value c (e.g. smallest ri in the number array) is subtracted from the ri. This procedure ensures, on the one hand, the limitation of the decimal places of the entire number array and, on the other hand, an adequate correlation of the number of rollovers ri with their corresponding damage events ni .

[0032] To decide whether a transmission damage warning is issued or not, the number array can be evaluated using a limit value comparison. A limit value comparison according to the formula X = n max 1 j − 1 ∑ i n i − n max → Getriebe iO , wenn X < T rel Getriebe niO , wenn X ≥ T rel proven. In this formula, X represents the maximum number n max normalized to the mean of all entries ni without n max , and T rel represents the threshold value above which a transmission damage warning is issued. i represents the index of the angular range and is the control variable over which the summation is performed; j represents the number of classes into which the entire angular range is divided.

[0033] In a preferred embodiment of the invention, during the automated check to determine whether the histogram contains an unexpected cluster of entries in a class, indicating wear and / or transmission damage, a non-randomly distributed background is corrected. In this way, systematic effects that may arise depending on the specific use of the engine-transmission unit can be corrected.

[0034] The presented method enables the detection of transmission damage in various transmission types. Examples include spur gears, helical gears, and worm gears, which are also used for safety-relevant applications in the automotive sector.

[0035] Depending on the application, there may be situations in which certain transmission damage is only noticeable in one direction of rotation of the motor-transmission arrangement. Analyzing signal curves separately by direction of rotation makes reliable detection of such transmission damage possible.

[0036] The invention is explained in more detail below with reference to the figures. It shows: Fig. 1: an exemplary mechanical structure that can be monitored for the occurrence of gear damage by means of the method according to the invention, Fig. 2: a comparison of an exemplary signal curve of an intact gearbox and a damaged gearbox, Fig. 3: a representation of the development of the maximum of the measured signal curves of a gearbox as a function of the load cycle, Fig. 4: a real signal curve of a damaged gearbox of a motor-gearbox unit, Fig. 5: a flow chart of a possible analysis of signal curves, Fig. 6: the real signal curve from Figure 4 after executing the second step of the flowchart Figure 5 intermediate result obtained, Fig. 7: the intermediate result from Figure 6 after executing the third step of the flowchart Figure 5obtained intermediate result, Fig. 8: the histogram obtained after running through the flow chart from Figure 5 for signal curves that meet the damage criterion at randomly distributed angular positions (represents the behavior of an intact gearbox), and Fig. 9: the histogram obtained after running through the flow chart from Figure 5 for a damaged gearbox.

[0037] Figure 1 shows an exemplary mechanical structure of a motor-gearbox unit 10, which can be monitored for transmission damage using the method according to the invention. The motor-gearbox unit 10 necessarily has a motor 1, which in this case is implemented in particular as an electric motor. The specific design of the electric motor is not important; for example, it can be a servo motor, a brushless DC motor, or even a brushed electric motor.

[0038] It is essential that the motor-gearbox unit 10 has a detection module 2, which is designed and operable to enable the recording of the input and / or output torque either directly or indirectly as a measured variable that can be clearly correlated therewith, in particular proportional thereto, such as the time-resolved motor current. This measured variable is therefore repeatedly measured during operation of the motor-gearbox unit 10, while the gearbox or the components that form it are in different positions, in particular engagement positions, so that during a movement of the motor-gearbox unit 10, a signal curve is generated that represents the directly or indirectly determined input and / or output torque as a function of a variable correlated with the position of the gearbox.

[0039] The drive 1 drives the gear 3 to be monitored, which here is an example of a helical gear / worm gear, so that under the influence of a load 4, the size of which can change, a movement is caused by the drive, for example a ram is moved.

[0040] According to the invention, the signal curves recorded by this detection module 2 serve as a parameter for the occurrence of gear damage, which enable the direct or indirect recording of the torque curve of a gear under load and enable the identification of irregularities in the signal curve, which allow early damage detection.

[0041] Figure 2illustrates the usability of signal curves recorded with such a detection module 2 for identifying transmission damage. Shown is a torque curve 21 of an intact transmission 3 as a dashed line, and the torque curve 22 of a damaged transmission 3, in which a tooth root exhibits a crack, as a solid line.

[0042] In this example, the data presented represents input torques measured using a torque transducer as a function of a value correlated to the angle of rotation, while the respective gear 3 is moved at a defined speed from a defined start position to a defined end position. In this example, the respective measured values ​​are plotted against a relative sampling index, which is thus a value proportional to the measurement time and can thus be reproducibly assigned to a given gear position, to generate the respective signal curve.

[0043] The gear position can generally be determined by a variety of variables correlated with it, for example as a function of a running time of the motor, by monitoring the position of the rotor of the motor, by monitoring the position of a plunger moved by the motor-gear unit 10 or similar measures.

[0044] The signal curve 22 of the damaged gear 3 differs from the signal curve 21 of the intact gear 3 due to the occurrence of maxima. These significantly pronounced modulations in the signal curve 22 are caused by the engagement of a damaged tooth, which leads to the drive motor increasing the input torque due to the inefficient torque transmission in order to ensure the required output torque.

[0045] Such anomalies can be used to define criteria for damage detection. Examples of such criteria include extreme values ​​of the data (maxima, minima, etc.), the area (sum, integral) under the maxima / minima of a specific range correlated with the rotation angle, or similar parameters. When damage occurs, one or more of these features increase significantly, marking the onset of damage.

[0046] To illustrate this fact, Figure 3 a representation of an exemplary, from the comparison according to Figure 2derivable feature, specifically in the form of the maximum of the individual signal curves, depending on the respective load cycle. A load cycle is the movement of the transmission 3 from the defined start position to the defined end position. For each of these load cycles, a torque curve was measured by the detection module 2. Each of the data points plotted in Figure 3 corresponds to the maximum amplitude of the torque curve of the corresponding load cycle.

[0047] One recognizes in Figure 3 that after approximately 95 percent of the maximum number of load cycles, the value of this feature, i.e. the maximum amplitude in the respective signal curve correlated with the torque, begins to increase, which in this example is an indicator of the onset of tooth fracture.

[0048] Under ideal conditions, such as those achievable in endurance tests with motor-gearbox units 10 in the laboratory, transmission damage can be reliably determined from directly or indirectly obtained torque measurement data. These conditions are ideal in two respects: First, the ambient conditions are controllable and therefore approximately constant, which facilitates the precise identification of the characteristics. When measuring the operation of a motor-gearbox unit 10 in situ in a real application, the situation is considerably more complex. The signal curves of two consecutive travel cycles can differ drastically from one another, for example, in terms of speed and the temporal progression of the output load.

[0049] Second, large amounts of computing power and storage space are usually available for automated data analysis under laboratory conditions, whereas these resources are available to a lesser extent during operation under real-world conditions (e.g., in a vehicle).

[0050] To clarify this fact, Figure 4An example of a torque curve 31 of a motor-gearbox unit 10 with a damaged gearbox 3, which is subject to changing loads during operation, determined using a torque measuring shaft, as a solid line. Here, the torque is plotted against an angular position at the output of the motor-gearbox unit, which can be determined, for example, directly using a rotary encoder or indirectly from the temporal signal curve. The area in which the effects of the damaged tooth can be seen is highlighted with a circle; a moving average 32 of the measurement signal is shown as a dashed line. This example immediately makes it clear how complex the task of detecting such damage in a signal curve and reliably assigning it to an angular position is, even when it is certain that damage exists.

[0051] The challenge here is to achieve comparability between measurement signals recorded under different measurement conditions with the least possible computational effort.

[0052] If you apply the steps in Figure 5 However, even in this case, reliable monitoring for transmission damage is possible using manageable hardware resources.

[0053] First, a pre-selection of the measuring cycles to be used is made (point (1.) in Figure 5). Only cycles with defined angular ranges are permitted for evaluation. Furthermore, the deviation of the absolute maximum and minimum from the mean value of the signal curve should not exceed a limit. The first measure ensures that all desired angular ranges are included in the signal curve and that any undesirable areas of the signal curve are neglected; the second ensures that the fluctuations caused by the output torque do not become uncorrectably large.

[0054] In point (2.) in Figure 5 the DC component is removed. This can be done using two approaches: I) The difference between the highly smoothed signal curve (dashed line in Figure 4 ) and the original signal curve (solid line in Figure 4). Furthermore, a division by the highly smoothed signal curve can optionally be performed to correct torque-dependent, periodically occurring amplitudes and / or optionally further smoothing and / or low-pass filtering can be performed to eliminate any interference signals from the signal curve. II) A bandpass filter is applied to the signal. The upper and lower frequency limits can be selected as a function of the input speed so that the tooth meshing frequency of the transmission component under consideration represents the majority of the periodic fluctuations in the signal curve. An advantage of this method is its easier implementation in common programming languages.

[0055] Both approaches can be implemented with little computational effort and simple mathematical operations.

[0056] Figure 6The solid line shows the corrected signal 35 obtained by applying method I). For better visualization, the moving average 36 of this corrected signal is shown as a dashed line. Especially after averaging, the trained eye can now detect peaks in the corrected signal curve above approximately -300°, which indicate transmission damage.

[0057] In the next two steps, (3.) and (4.) in Figure 5 , the range of the quantity correlated with the position of the gearbox is now identified, to which a defined characteristic feature for a gearbox failure can be assigned for the respective signal curve. In this example, the occurrence of the maximum in the signal curve represents the characteristic feature for a gearbox failure. In step (4.) in Figure 5 the position α max at which the characteristic transmission damage feature occurs is determined.

[0058] In this case, the angle scale of the Figure 6 into an angle scale that indicates the corresponding position of the monitored transmission component (which is always between 0° and 360°). Furthermore, the corrected signal curve is divided into classes, with each class comprising a fixed angle range, in this example 10° of the angle scale. All values ​​of the corrected signal curve within a class are summed. The resulting signal 41 is schematically shown in the form of a bar chart in Figure 7The angle α max at the position of the bar with the highest value in the bar chart (41) is identified; it corresponds to the respective range identified with the quantity correlated with the position of the transmission, in which the defined characteristic feature for transmission damage can be found for the respective signal curve. This leads to a significant reduction in the amount of data requiring further processing, since the high-resolution corrected signal curve is converted into a structure with a few relevant data points, the respective bins with the respective associated sum.

[0059] If, due to the detection module 2, the detected signal curve is available as a function of the angle scale, which indicates the corresponding position of the monitored transmission component, the value obtained after steps (1.) and (2.) in Figure 5The corrected signal curve generated can be directly subjected to class classification and summation within a class.

[0060] For each of the preselection (step (1.) in Figure 5 ) passing individual signal waveforms, an α max is determined. This value is entered into a histogram (51, 52) in which the frequency with which α max occurs is shown (point (5.) in Figure 5 ).

[0061] In Figure 8 A histogram (51) is shown, which is based on randomly generated signal curves that are supposed to represent an intact transmission. These signal curves were calculated using the previously described algorithm according to Figure 5evaluated. As expected, it can be seen that the extracted α max values ​​are distributed approximately randomly, so that the histogram (51) forms a uniformly distributed base level. In some applications, this uniformly distributed base level may only be obtained after correcting for a systematic effect superimposing this distribution.

[0062] If a further thirty signal curves are added to the approximately thirty evaluated data sets, in which, in this example, a damaged worm wheel is present, the result is Figure 9 shown histogram 52, which now shows a striking maximum at the point where the damaged tooth is loaded, here at about 70°.

[0063] To meet the requirements of limited permanent storage space, the method according to the invention requires only a number array of a few kilobytes, which must be permanently stored. Table 1 schematically shows how such an array is constructed with a class size of 10 and class index i, and how the individual cells can be "filled." Table 1 Angle α (°) 0 10 20 30 40 ... 360 Number ni 2 4 9 3 25 ... 0 Rollovers ri 30 28 25 25 5 ... 2

[0064] Here, only the number ni of individual events at certain angular positions as well as the number of rollovers ri of this angular range are counted up and stored.

[0065] For example, a simple query of the angular position of the maximum in the array of numbers, combined with a limit value comparison, can allow an assessment of the condition of the transmission. Since the numbers in the array can become very large with increasing measurement cycle numbers, thus requiring more storage space, and since there are angular ranges that are rarely crossed, a correction or a type of normalization of the damage events to the number of crossings must be performed. The following rule is used to determine the corrected values ​​n i-new and r i-new within the array of numbers: 1 . n i − neu = n i − n ¯ ⋅ r i / r sup 2 . r i − neu = r i − c

[0066] Here, nrepresents the mean value of all entries ni. The maximum number of rollovers in the entire array is denoted by r sup. Any negative n i-new values ​​that may arise are set to a constant, positive value. Furthermore, possible decimal places are removed. In addition, to limit the decimal places of the ri after the cell-wise application of equation (1) 1. in step 2. of equation (1) a constant value c (e.g. smallest ri in the number array) is subtracted from the ri. This procedure ensures, on the one hand, the limitation of the decimal places of the entire number array and, on the other hand, an adequate correlation of the number of rollovers ri with their corresponding damage events ni .

[0067] After each application of the rule from equation (1) to all cells, a limit value comparison allows an assessment of the condition of the transmission. Equation (2) represents the limit value comparison symbolically: X = n max 1 j − 1 ∑ i n i − n max → Getriebe iO , wenn X < T rel Getriebe niO , wenn X ≥ T rel

[0068] Here, X is the maximum number n max divided by the mean of the individual entries ni without n max . Here, j represents the number of array classes. If X exceeds a previously defined relative limit T rel , the transmission is declared "out of order." Consequently, a maintenance recommendation can be issued ( Figure 4 , step (6.)).

[0069] The logic described represents a computing resource-saving option for condition monitoring of a motor-gearbox unit 10, in which system monitoring can be realized without additional hardware and with the lowest possible additional costs. List of reference symbols

[0070] 1 Motor 2 Detection module 3 Gearbox 4 Load 10 Motor-gearbox unit 21 Signal curve 22 Signal curve 31 Signal curve 32 Moving average 35 Corrected signal curve 36 Moving average 41 Histogram 51 Histogram 52 Histogram

Claims

1. Method for detecting wear, damage to the gearing and / or damage to the bearings, in an engine transmission unit (10), having an engine (1) and a transmission (3) which is coupled to the engine on its output side, in which - when the transmission (3) is put into motion by the engine (1), an input and / or output torque is either directly determined, or indirectly as a measurement variable which can be unambiguously correlated with the input and / or output torque, by means of a detection module (2), in order to generate a signal curve (21, 22, 31) for such motion which describes each directly or indirectly determined input and / or output torque in dependency on a variable which correlates with the position of the transmission (3), - at least some of these signal curves (21, 22, 31) are automatically evaluated, wherein a range of the variable which correlates with the position of the transmission (3) is identified, to which a value which is typical for a transmission defect can be assigned for the respective signal curve (21, 22, 23), characterized in that - the identified range is automatically assigned to a class of a histogram (51, 52) and - it is automatically verified whether there is an unexpected proliferation of entries in any class in the histogram (51, 52), indicating the presence of wear and / or damage to the transmission.

2. Method in accordance with claim 1, characterized in that an automatic preliminary selection of the signal curves (21, 22, 31) intended for evaluation is carried out.

3. Method in accordance with claim 2, characterized in that, during the automatic preliminary selection, only signal curves (21, 22, 31) which contain torque data relating to positions of the transmission which cover the specified ranges is accepted for further evaluation.

4. Method in accordance with claim 2 or 3, characterized in that, during the preliminary selection, only those signal curves (21, 22, 31) in which the deviation of the global maximum and of the global minimum from the average value of the signal curve do not exceed a specified threshold value are accepted for further evaluation.

5. Method in accordance with any of claims 1 to 4, characterized in that a direct component is determined from the analyzed signal curves (21, 22, 31) and is removed.

6. Method in accordance with claim 5, characterized in that the direct component is determined by forming a moving average (32) of the signal curve (21, 22) for the smoothing of the signal curve and in that the direct component is removed by subtracting this moving average (32) from the corresponding related measured value of the unsmoothed signal curve (21, 22, 31).

7. Method in accordance with claim 6, characterized in that, in addition, a division by the smoothed signal curves is performed in order to carry out a correction of periodically occurring torque-dependent amplitudes and / or, optionally, a further smoothing and / or a low pass filtering is carried out in order to eliminate any possible interference signals from the signal curve.

8. Method in accordance with claim 6, characterized in that the direct component is determined and removed by applying a band pass filter to the signal curve (21, 22, 31).

9. Method in accordance with claim 8, characterized in that the upper and the lower frequency limits of the band pass filter are selected as a function of the input rotational speed so that the gear mesh frequencies of the monitored components of the transmission (3) represent the major portion of the periodic fluctuations of the signal.

10. Method in accordance with any of claims 5 to 9, characterized in that a normalization of the signal curve (21, 22, 31) is carried out with the determined direct component.

11. Method in accordance with any of claims 1 to 10, characterized in that the histogram (50, 51) is saved as an array of numbers which, for each range of the value correlating with the position of the transmission (3), the number ni of the signal curves (21, 22, 31), during the evaluation of which the value characteristic for a transmission defect is assigned to this range, as well as the number ri of overrollings of this angular range. are saved.

12. Method in accordance with claim 11, characterized in that the array of numbers is updated at least once with corrected values ni-new and ri-new, which are determined in accordance with the specification 1. n i − new = n i −<menclose notation="box"> n < / menclose> ⋅ r i / r sup 2. r i − new = r i − c wherein n represents the average value of all assignments ni, rsup designates the maximum number of overrollings in the entire array, any possibly occurring negative ni-new values are set to a constant positive value, any existing digits following the decimal point are removed, and c is a constant value, in particular, is the smallest ri in the array of numbers before the update.

13. Method in accordance with claim 11 or 12, characterized in that the array of numbers is evaluated by means of a threshold value comparison.

14. Method in accordance with claim 13, characterized in that the threshold value comparison is carried out in accordance with the equation X = n max 1 j − 1 ∑ i n i − n max → Getriebe iO , wenn X < T rel Getriebe niO , wenn X ≥ T rel wherein X is the present calculated value, Trel represents the threshold value at which a warning of a transmission defect is issued, nmax describes the maximum number of damage occurrences in the array of numbers, rmax indicates the matching number of overrollings, ni represents the i - th entry of the damage occurrences in the array of numbers, ri is the corresponding number of overrollings and j is the total number of angular ranges.

15. Method in accordance with any of claims 1 to 14, characterized in that, during the automatic verification of whether an unexpected proliferation of entries in any class is present in the histogram (51, 52), indicating the presence of wear and / or a transmission defect, a non-randomly arranged background is corrected.

16. Method in accordance with any of claims 1 to 15, characterized in that the signal curves (21, 22, 31) are analyzed separately according to their running direction.