• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The present invention discloses a planetary gear failure detection device comprising a vibration sensor and a processor configured to: receive a vibration signal from the vibration sensor; carrying out a low-pass filtering of said vibration signal obtaining a filtered signal; divide the filtered signal into a series of intermediate signals; operating each of said intermediate signals with a series of base signals obtaining a series of processed signals and calculating a series of scalar values calculated from each of said processed signals; and outputting to a user the calculated scalar values to determine a possible planetary gear failure. (Machine-translation by Google Translate, not legally binding)
    公开号:ES2754278A2
    申请号:ES201830943
    申请日:2018-09-28
    公开日:2020-04-16
    发明作者:Arraiza Maite Martincorena；La Cruz Blas Carlos Aristóteles De；Maestro Ignacio Raúl Matías；Martín Antonio Jesús López
    申请人:Universidad Publica de Navarra；
    IPC主号:
    专利说明:

    [0001]
    [0002]
    [0003] TECHNICAL SECTOR
    [0004]
    [0005] The invention falls within the field of gear failure sensors. Specifically, in the detection of failures in planetary gears from a signal associated with the vibration of the gear.
    [0006]
    [0007] BACKGROUND OF THE INVENTION
    [0008]
    [0009] Planetary Gears (EP) are widely used as mechanical elements in the automotive, aeronautical, wind power generation, etc. Generally these mechanisms are in charge of giving the necessary power to drive the motors that execute the functions associated with each application, so the detection of failures in predictive maintenance in these elements is crucial in terms of safety and cost.
    [0010]
    [0011] One of the most common planetary gear configurations has the following elements:
    [0012]
    [0013] • Crown or ring (or 'ring' as known in the art) with a number ZR of teeth: Rotation frequency f R = 0
    [0014] • Sol: input component with a number Zs of teeth: Rotation frequency fs ^ 0
    [0015] • Planets with a ZP number of teeth per planet: Rotation frequency f P ^ 0
    [0016] • Planet carrier (or Carrier as known in the art): Rotation frequency fc ^ 0
    [0017]
    [0018] The operation is as follows: The sun rotates around its axis at high speed and when engaging with the Planets, they rotate around their own axes and, in turn, mesh with the stationary corona, which causes them to move to turn around the axis of the sun (input axis) by rotating the planet carrier at a lower speed than that of entry. Thus, the output ratio is equal to Zs / Zp.
    [0019]
    [0020] The fault detection and predictive maintenance process in this type of gear is carried out, in a non-intrusive way, by a sensor located in the gear crown that collects the vibrations transmitted from the various gear points and, with this information and the relevant signal processing, it is decided whether there is a fault or not. The typical characteristics of the signals received by the sensor are: amplitude and / or phase modulations, a component of random signals and that the vibration system is of varying time (the faults appear over time).
    [0021]
    [0022] Spanish Patent 2 282 441 T3 discloses a method and system to detect signs of vibration in a reversible drive and perform a process based on calculating averages, kurtosis and obtaining the frequency spectrum.
    [0023]
    [0024] On the other hand, the Spanish patent application ES 2294959 A1 discloses a monitoring and process equipment for wind turbines and a predictive maintenance system for wind farms. The monitoring equipment has means for processing the signals captured from the accelerometers, preferably comprising means for signal conditioning, calculation of global values (RMS, crest factor), and alarm detection algorithm.
    [0025]
    [0026] DESCRIPTION OF THE INVENTION
    [0027]
    [0028] The techniques for detecting possible failures in planetary gears of the prior art have, mainly, two problems: i) a dependence on a single parameter for the detection (in particular, the magnitude of vibrations), therefore, If there is noise in the system that affects this parameter, detection is usually poor; and ii) they do not allow identifying or, at least, guiding the user on which gear element may be presenting the fault.
    [0029]
    [0030] Accordingly, the present invention discloses a device and method for fault detection in a planetary gear of the type comprising:
    [0031] • a sun gear;
    [0032] • a series of planet gears orbiting the sun gear; • a crown geared to the planet gears;
    [0033] • a planet carrier attached to the central zone of the planet gears; said gear comprising a vibration sensor coupled to the planetary gear and a processor associated with the vibration sensor, the processor being configured to:
    [0034] i) receiving a vibration signal from the vibration sensor;
    [0035] ii) perform a low-pass filtering of said vibration signal obtaining a filtered signal;
    [0036] iii) dividing the filtered signal into a series of intermediate signals with the number of intermediate signals being equal to the number of planet gears in the planet gear series;
    [0037] iv) operating each of said intermediate signals with a series of base signals obtaining a series of processed signals, and calculating a series of scalar values calculated from each of said processed signals;
    [0038] v) output to a user the calculated scalar values to determine a possible planetary gear failure.
    [0039]
    [0040] Preferably, the division of the filtered signal into intermediate signals is performed by dividing the period of the filtered signal by the number of planet gears.
    [0041]
    [0042] In a particular embodiment, the base signals are signals orthogonal to each other.
    [0043]
    [0044] Furthermore, obtaining scalar values can be done, for example, by integrating each processed signal for at least part of a period. Alternatively or additionally, obtaining scalar values can be performed by integrating each processed signal over a period. In another embodiment, obtaining scalar values is performed by calculating an average value, calculating a statistical value, etc.
    [0045]
    [0046] In a particularly preferred embodiment, the processor comprises or has access to a memory in which reference scalar values are stored.
    [0047]
    [0048] Furthermore, the processor may comprise instructions for the determination from the calculated scalar values of a possible failure by means of a comparison between the scalar values calculated with the reference scalar values.
    [0049]
    [0050] Preferably, the comparison between scalar values can be performed by calculating a Euclidean distance between each calculated scalar value and its corresponding reference scalar value; if said distance is greater than a threshold distance, a possible failure is determined.
    [0051]
    [0052] More preferably, the processor may be configured to correlate each scalar value to a possible type of planetary gear failure.
    [0053]
    [0054] In one example, the emission to the user of the scalar values is presented to the user by a graph such as a series of scalars arranged on a coordinate axis or an eye diagram.
    [0055]
    [0056] In a particularly preferred example, the vibration sensor is attached to the crown.
    [0057]
    [0058] BRIEF DESCRIPTION OF THE DRAWINGS
    [0059]
    [0060] Figure 1 shows a front and perspective view of a planetary gear according to the present invention.
    [0061] FIG. 2 shows a schematic view of an exemplary embodiment according to the present invention.
    [0062] Figure 3 shows an example of a filtered signal obtained for a planetary gear that does not have failures.
    [0063] Figure 4 shows another example of a filtered signal obtained for a planetary gear that has no faults.
    [0064] Figure 5 shows a graphical representation of the calculated scalar values and a comparison with the reference scalar values in an embodiment of the present invention.
    [0065] Figure 6 shows another example of graphic representation, according to a method of the present invention.
    [0066] PREFERRED EMBODIMENT OF THE INVENTION
    [0067] Figure 1 shows a planetary gear (1) of the type to be monitored according to the present invention. The gear (1) of figure 1 has a crown (2), a sun (3) and a series of planets (4, 5, 6) that orbit around said sun (3). Specifically, the gear (1) in Figure 1 has three planets whose bodies are joined by a planet carrier (7).
    [0068]
    [0069] The vibration signal of a planetary gear (1) like the one in figure 1 comprises a high-frequency component (fundamental frequency known as gear frequency) fm corresponding to the processes of gear between teeth of the different sprockets and an envelope of lower frequency due to the path of transmission of vibration at each gear point to the sensor.
    [0070]
    [0071] When faults occur in any tooth of the crown (2), the sun (3) or one of the planets (4, 5, 6), additional lateral bands are produced in the vibration signal spectrum around the frequency of gear, for example, at integer multiples distances of the characteristic frequencies of each fault (all of them less than the gear frequency). Therefore, the following components can be observed in the vibration signal of a planetary gear (1):
    [0072] 1. High-frequency component at the gear frequency (fm = Zrf c ). 2. Low frequency components corresponding to the vibration transmission window ( fA ( t) = Npf c), corona failures ( ffr = Npf c, failures
    [0073] in the sun ( ffs = N „ Z ^ s fc) and / or faults on the planets ( ffp = Z p - fc ).
    [0074]
    [0075] Being both the high-frequency components and the low-frequency components included in the information transmitted through pulses produced by each planet to a vibration sensor (not shown) that can be housed, for example, in the corona.
    [0076]
    [0077] Figure 2 shows a schematic view of an exemplary embodiment of a diagnostic device and method according to the present invention. Figure 2 shows how a vibration sensor (8) is arranged in the planetary gear (1), for example in the crown (2), as in an embodiment example.
    [0078]
    [0079] As shown in Figure 2, initially, a first stage (201) is performed that corresponds to a frequency shift, filtering and an analysis of possible errors in the system, these errors being, for example, manufacturing errors in the planetary gear (1), coupling errors, installation errors, etc.
    [0080]
    [0081] The filtering of the signal obtained by the vibration sensor (8) is performed to rule out the high-frequency component since, as explained with reference to Figure 1, said high-frequency component does not provide relevant information as to to possible failures of the planetary gear (1) or, at least, not relevant in terms of the failures that are intended to be analyzed by the device and method of the present invention.
    [0082]
    [0083] In an exemplary embodiment, the filtering of the signal obtained by the vibration sensor is carried out by multiplying the vibration signal v ( t), which is a sum of cosines at the gear frequency corresponding to the processes Crown-planet and Sun-planet gear, by cos ( 2nfm) and - sin ( 2nfm ) to obtain orthogonal signals and, subsequently, low-pass filtering is performed to eliminate high-frequency components and signal noise, giving rise to to the baseband signals v $ ( t) and v2 ( t), respectively. Thus, only the low-frequency component is used.
    [0084]
    [0085] Solving the trigonometric identity and considering only the non-integer part of the cosine lag, since in the cosine phase the integer multiples of a value generate the same result, the following expressions are obtained, prior to low-pass filtering:
    [0086] (to)
    [0087] Vlpb ( t) ( C lev Ae 2nfc {t ~ 24r) ( cos ( 2n2fmt - 9 d cos ( -W¿))
    [0088]
    [0089] (b)
    [0090] V2Vb ( t) = ( C iev Ae2nfc {t-24 r) ( - ( sin ( 2n2fmt - W ^ - sen ^ -W ^))
    [0091]
    [0092] Where DC * ev Ae 20fc ( tI <= c) is equivalent to the contribution, in low frequency, of each
    [0093] planet that corresponds to an amplitude modulation window applied to the vibration signal produced by the gear-to-tooth processes in the planetary gear (1); fm is the gear frequency, fc the rotation frequency of the planet carrier and Wi the angular position of each planet.
    [0094]
    [0095] After applying the low pass filter to equations (a) and (b) the components are removed frequency
    [0096]
    [0097]
    [0098] ^ 2 (t) = DC * ev Ae 20 ^ i (t -24/1) (d) (-sen (- 9 í))
    [0099]
    [0100] Once the frequency shift and filtering of the vibration signal obtained by the vibration sensor (8) have been carried out, a filtered signal is obtained as a result of the first stage (201), which will be used for the analysis and detection of faults. Said filtered signal will be v $ ( t) if in (201) a multiplication by cosine is performed and v2 ( t) if the multiplication is by sine, the later stages being common in both cases.
    [0101]
    [0102] As previously mentioned, said filtered signal can be analyzed to determine possible manufacturing and / or coupling errors, since the expected signal is ideally composed of a fundamental frequency corresponding to the gear frequency and lateral bands located in the frequencies Np * fc of the gear frequency where Np corresponds to the number of planets and fc to the rotation frequency of the planet carrier. When a tooth failure occurs, in addition to these side bands, additional frequency components appear around the gear frequency, the separation of which corresponds to the frequency of rotation of the sprocket on which the failed tooth is located.
    [0103]
    [0104] In order to analyze each of the gears individually, in a second stage (202) the vibration signal v1 (t) equ (c) and v2 (t) equ (d) is divided into a series of intermediate signals in the weather. The present invention proposes that there be as many intermediate signals as planets have the planetary gear (1). Therefore, for a gear such as the one shown in Figures 1 and 2, the filtered signal is divided into three intermediate signals. In a more general case, the period (Tc) of the filtered signal is divided into NP sections of duration Tc / NP corresponding to each planet. Obtaining a series of intermediate signals in which each intermediate signal corresponds to one of the planets of the planetary gear, ie, NP intermediate signals. Figure 2 shows the process for one of said intermediate signals, the subsequent stages being the same for the NP intermediate signals.
    [0105] In order to be able to carry out a temporal analysis in which information is available for each of the components of the planetary gear, the present invention contemplates that, in a third stage (203) each of the intermediate signals is operated with a base function in time (composed of N base functions, represented in the diagram as fb B- with j = 1,2, ..., N) and, in a fourth stage (204), is integrated throughout its duration ( TC / NP) so we obtain a scalar value P. At the end of this process we obtain scalar N x NP Pí; -, where i = 1,2, ..., NP the number of the planet and j = 1,2 ,. .., N the number of the base function. These scales P¿7- are the ones that will serve as indicators of failure and diagnosis, in turn, they can also be represented graphically with the bases as axes giving rise to an interpretation of points P: in N dimensions known as constellation. Obtaining the base functions can be performed in different ways that provide greater flexibility depending on the requirements and limitations of each case under study; Three methods are proposed in this invention and are discussed below. The value of N varies depending on the method used and, in the case of Method 1, as explained later, also on the number of planets.
    [0106] Examples of analyzes of such base signals can be found, for example, in S. Haykin, "Digital Communication Systems", 5th edition, John Wiley & Sons, Inc, chapters 7 and 8, the contents of which are incorporated by reference into the present disclosure. In short, the geometric representation of the signals is based on representing any set of M energy signals s: (t) as a linear combination of N orthonormal base functions where N <M, each base function is preferably normalized to have energy unit and, in addition, the base functions must be orthogonal to each other (null scalar product) throughout the interval corresponding to the planet period (Tc / Np).
    [0107]
    [0108] The present invention further contemplates having a series of reference parameters, for example, a series of previously obtained reference scalars and stored in a memory of the controller. These reference scalars can be obtained through a calibration process prior to using the sensor for fault detection or they can be values provided by the manufacturer and stored in said memory.
    [0109]
    [0110] Thus, with the diagram in Figure 2, the signal to be analyzed can be compared with the signal without failures graphically or by comparing the reference scalars P¿7 without failures with the that are obtained with the sensor signals, ie, the scalars measured or obtained through the measurement procedure previously explained. In this way, the components added by the fault will cause the constellation points to move (rotate, by phase change, or increase in amplitude) with respect to their ideal positions.
    [0111]
    [0112] Below are examples of base signals that can be used in a fault detection device and / or method according to the present invention. In a first example of embodiment, there is a set of M signals Pi (0, P2 (0> ..., Pm (0), which correspond to the modulation windows P1, P2 and P3 of each planet in v $ ( t) and v2 (t), a first base function example is calculated as follows:
    [0113]
    [0114]
    [0115] where = J07P i (t) dt is the energy of the signal p $ ( t) where T is the period of the signal, which, in this case, corresponds to the period of rotation of the planet carrier between the number of planets (T = 7Nwp ). The second base function is obtained from the first base function and the signal p2 ( t). First define p21:
    [0116]
    [0117]
    [0118]
    [0119]
    [0120] Next, an intermediate base signal g2 ( t) is defined :
    [0121] [2 (t) = P2 (t) - P2i (t) 0 i (t), (g)
    [0122]
    [0123] g2 (t) is orthogonal to 0 $ (t) on the interval 0 <t <T due to the definition of p and the fact that the base function 0 $ (t) has unit energy. Once these terms are defined, a second base function can be calculated using the following formula:
    [0124]
    [0125]
    [0126] Continuing with this procedure, the intermediate base function can be defined, in general, as:
    [0127] í - i (i) [í (t) = P (t) - ^ P í; 0; (t)
    [0128] B = i
    [0129] where the coefficients Pí; - are defined as follows:
    [0130]
    [0131] For i = 1, gi ( t) is reduced to pi (t). Having gi ( t), the set of base functions can be defined as:
    [0132] [i (t) (k)
    [0133] 0 j (t) =, j = 1,2, ..., N
    [0134] ] foTg 2 ( t) dt
    [0135]
    [0136] As a characteristic of this way of choosing the base, it is the fact that the number of base functions is directly proportional to the number of planets (the number of base functions being NP - 1), which can be a drawback since there will be no graphical representation in case the number of planets is greater than 4.
    [0137]
    [0138] A second procedure for the identification of base signals comprises trimming the window so that the overlap between windows is reduced and, thus, the trimmed and normalized window to 1 can be used as a base function.
    [0139]
    [0140] Despite the fact that the low computational complexity of this method is a great advantage, its main drawback is that, by cutting the base, it does not cover the entire signal portion to be analyzed and, therefore, information is lost. In addition, the overlap between windows is greater the greater the number of planets and, therefore, the cutout of the window should be greater to obtain minimal interference between the contributions of each planet.
    [0141]
    [0142] The loss of information in the areas of v $ ( t) and v2 ( t) that are not covered by the base, can be compensated by reconstructing these signals from the constellation and their subsequent comparison with the original ones without failures. That is, by obtaining the error signals. This results in increased signal processing.
    [0143]
    [0144] In a third procedure for identification of base functions, a calculation of the base can be performed from the modulation window produced by the transmission path of the reference planet and the tails of the adjacent windows that overlap with it. Using this method, regardless of the number of planets, there will always be three base functions (one will be to the modulation window and the other two tails to the right and left of it). This method avoids the loss of information from method 2 and the dependence on the number of planets in the first procedure.
    [0145]
    [0146] However, a person skilled in the art could identify another type of base functions without departing from the main concept of the present invention in which, in principle, any plurality of base functions that are orthogonal to each other and, preferably, normalized, could be used obtaining different detection levels depending on the specific applications of the planetary gear (1).
    [0147]
    [0148] Figures 3 and 4 show the filtered signals v $ ( t) and v2 ( t) respectively after the application of the frequency shift and low pass filtering corresponding to a planetary gear without failures (according to equations (c) and (d)) . In them, the contributions of each of the planets (P1, P2 and P3) of duration T c / N p , where T c is the period of rotation of the planet carrier and N p the number of planets, can be differentiated.
    [0149]
    [0150] Once the filtered signals have been obtained, the filtered signals are multiplied by the base functions, obtaining a processed signal. From this processed signal, a scalar value can be obtained for its graphic representation.
    [0151]
    [0152] In an exemplary embodiment, to obtain the scalar value of the processed signal, we proceed to integrate said processed signal in the period corresponding to each planet. That is, for the example of Figures 1 to 3 in which there are three planets, for the first planet (4) the integral will be from 0 to TC / N P, for the second planet (5) of TC / NP at 2TC / NP and for the third planet (6) from 2TC / NP to TC. In this way, a set of scalars are obtained that can be graphically represented in a signal-space diagram that we will call a constellation.
    [0153]
    [0154] In an exemplary embodiment, in a calibration step the aforementioned procedure is performed in a scenario where there are no failures. This scenario is defined as a reference scenario that corresponds to an ideal constellation. Once the ideal constellation has been defined, the processing of the vibration signals to be analyzed is executed to detect possible failures and, compared to the ideal constellation, if the points in the constellation to be analyzed exceed defined thresholds and are stored, for example, in a memory of a controller, it is considered to exist a planetary gear failure.
    [0155]
    [0156] In an exemplary embodiment, to define these thresholds, the displacements of the points with respect to their ideal position due to noise or to the tolerances to be taken into account due to manufacturing errors are considered.
    [0157]
    [0158] The device and method according to the present invention provide a user with a series of failure indicators suitable for different needs. The correct combination of the three indicators minimizes erroneous diagnoses. In an exemplary embodiment, the indicated faults can be:
    [0159]
    [0160] I. Constellation: It allows observing the displacements of the points with respect to their ideal positions for possible online monitoring of the gear condition (1) throughout its operation.
    [0161] II. Euclidean error or distance between the points of the analyzed gear constellation (1) and their ideal positions: They provide a numerical representation of the gear state (1) and allow the definition of error thresholds from which a fault is considered to exist. must trigger an alarm.
    [0162] III. Error signals between the real vibration signal and its reconstruction from the constellation: For the modeled fault, they allow observing the variations produced in the signal due to the fault (amplitude and separation of the peaks). This information, complementary to the constellation and the error, can be used to determine the existence, location and severity of the failure.
    [0163]
    [0164] Figure 5 shows an example of a graphic representation of the vibration signals obtained from a planetary gear (1) without failures. Specifically, for a gear (1) without failures, the ideal position associated with the first planet P1 (51), the ideal position associated with the second planet P2 (52) and the ideal position associated with the third planet P3 (53) are determined.
    [0165]
    [0166] Furthermore, the positions obtained have been simulated when a gear tooth failure occurs in the gear (1). In figure 5 it can be seen that the contributions associated with each of the planets change, specifically, the position with corona failure associated with the first planet (54), the corona failure position associated with the second planet (55) and the corona failure position associated with the third planet (56).
    [0167]
    [0168] Similarly, the positions obtained have been simulated when a gear tooth (1) fails in a tooth on a planet. In figure 5 it can be seen that the contributions associated with each of the planets change, specifically, the position with failure on a planet associated with the first planet (57), the position with failure with a planet associated with the second planet (58 ) and the failed position of a planet associated with the third planet (59).
    [0169]
    [0170] Similarly, the positions obtained have been simulated when a sun tooth failure occurs in the gear (1). In figure 5 it can be seen that the contributions associated with each of the planets change, specifically, the position with failure of the sun associated with the first planet (60), the position with failure of the sun associated with the second planet (61) and the position with failure of the sun associated with the third planet (62).
    [0171]
    [0172] As can be deduced from figure 5, if the measured scalar values of the planetary gear (1) analyzed are separated from the reference scalar values (corresponding to an ideal constellation) by a distance greater than a threshold (which includes, for example, a margin of tolerances and noise) means that there is a failure in the planetary gear (1).
    [0173]
    [0174] Figure 5 shows that the measured scalar values that make up the constellation move from their ideal positions, ie, they move with respect to the reference scalar values, when there is a failed tooth. Furthermore, the magnitude of this displacement (error) follows the following order, from smallest to largest, depending on the sprocket on which the failed tooth is located: planet failure is less than the displacement in a crown failure that, at in turn, it is less than in a failure of the sun. This, knowing the characteristics of the planetary gear (1), allows locating the failed tooth between the different sprockets that make up the system. Thus, this method not only indicates the presence of a fault, but additionally, provides more information about it, specifically, where the fault comes from, greatly helping the diagnosis.
    [0175] Figure 6 shows another type of graphic representation of the information obtained by a method according to the present invention. Specifically, an eye diagram of the type disclosed is shown, for example in S. Haykin, "Digital Communication Systems", 5th edition, John Wiley & Sons, Inc, Chapter 8, p. 463. It is the superposition of planet periods (Tc / Np) (63) in a measurement. The ideal mask (64) corresponds to the superposition of periods of a vibration signal of an EP without failures. This is used as a reference since, when there is a failure in any tooth, peaks (6) are expected in the representation of overlapping periods that will exceed this mask and, in this way, failures can be detected when this occurs. In addition, the separation between peaks will be determined by the period of rotation of the sprocket where the failed tooth is located.
    权利要求:
    Claims (24)
    [1]
    1. Device for detecting faults in a planetary gear of the type comprising:
    • a sun gear;
    • a series of planet gears orbiting the sun gear;
    a crown geared to the planet gears;
    • a planet carrier attached to the central zone of the planet gears;
    characterized in that it comprises a vibration sensor coupled to the planetary gear and a processor associated with the vibration sensor, the processor being configured to:
    i) receiving a vibration signal from the vibration sensor;
    ii) perform a low-pass filtering of said vibration signal obtaining a filtered signal;
    iii) dividing the filtered signal into a series of intermediate signals with the number of intermediate signals being equal to the number of planet gears in the planet gear series;
    iv) operating each of said intermediate signals with a series of base signals obtaining a series of processed signals and calculating a series of scalar values calculated from each of said processed signals;
    v) output to a user the calculated scalar values to determine a possible planetary gear failure.
    [2]
    Device according to claim 1, characterized in that the division of the filtered signal into intermediate signals is carried out by dividing the period of the filtered signal by the number of planet gears.
    [3]
    Device according to any of the preceding claims, characterized in that the base signals are signals orthogonal to each other.
    [4]
    Device according to any of the preceding claims, characterized in that the obtaining of scalar values is carried out by integrating each processed signal for at least part of a period.
    [5]
    5. Device according to claim 4, characterized in that the scalar values are obtained by integrating each processed signal during a period.
    [6]
    6. Device according to any of claims 1 to 3, characterized in that the scalar values are obtained by calculating an average value and / or calculating a statistical value.
    [7]
    Device according to any of the preceding claims, characterized in that it comprises a memory in which scalar reference values are stored.
    [8]
    Device according to claim 7, characterized in that it comprises the determination from the calculated scalar values of a possible failure by means of a comparison between the calculated scalar values with the reference scalar values.
    [9]
    9. Device according to claim 8, characterized in that the comparison between scalar values comprises calculating a Euclidean distance between each calculated scalar value and its corresponding reference scalar value, if said distance is greater than a threshold distance, a possible failure is determined.
    [10]
    10. Device according to any of the preceding claims, characterized in that it comprises correlating each scalar value to a possible type of failure in the planetary gear.
    [11]
    11. Device according to any of the preceding claims, characterized in that the emission to the user of the scalar values is presented to the user by means of a graph.
    [12]
    12. Device according to any of the preceding claims, characterized in that the vibration sensor is coupled to the crown.
    [13]
    13. Failure detection method in a planetary gear comprising the steps of:
    • receive a vibration signal from a vibration sensor associated with the planetary gear;
    • perform a low pass filtering of said vibration signal obtaining a signal filtered;
    • divide the filtered signal into a series of intermediate signals with the number of intermediate signals equal to the number of planet gears in the planet gear series;
    • operating each of said intermediate signals with a series of base signals obtaining a series of processed signals and calculating a series of scalar values calculated from each of said processed signals;
    • issue calculated scalar values to a user to determine possible planetary gear failure.
    [14]
    14. Method according to claim 13, characterized in that the division of the filtered signal into intermediate signals is carried out by dividing the period of the filtered signal by the number of planet gears.
    [15]
    15. Method according to any of claims 13 or 14, characterized in that the base signals are signals orthogonal to each other.
    [16]
    16. Method, according to any of claims 13 to 15, characterized in that the scalar values are obtained by integrating each processed signal for at least part of a period.
    [17]
    17. Method according to claim 16, characterized in that the obtaining of scalar values is carried out by integrating each processed signal during a period.
    [18]
    18. Method according to any of claims 13 to 15, characterized in that the obtaining of scalar values is carried out by calculating an average value, calculating a statistical value, etc.
    [19]
    19. Method according to any of claims 13 to 18, characterized in that it comprises a memory in which scalar reference values are stored.
    [20]
    20. Device, according to claim 19, characterized in that it comprises the determination from the calculated scalar values of a possible failure by means of a comparison between the calculated scalar values with the reference scalar values.
    [21]
    21. Method according to claim 20, characterized in that the comparison between Scalar values comprise calculating a Euclidean distance between each calculated scalar value and its corresponding reference scalar value, if said distance is greater than a threshold distance, a possible failure is determined.
    [22]
    22. Method according to any of claims 13 to 21, characterized in that it comprises correlating each scalar value to a possible type of failure in the planetary gear.
    [23]
    23. Method according to any of claims 13 to 22, characterized in that the emission to the user of the scalar values is presented to the user by means of a graph.
    [24]
    24. Method according to any of claims 13 to 23, characterized in that the vibration sensor is coupled to the crown.
    类似技术:
    公开号 | 公开日 | 专利标题
    Liang et al.2016|A windowing and mapping strategy for gear tooth fault detection of a planetary gearbox
    Jablonski et al.2013|Validation of vibration measurements for heavy duty machinery diagnostics
    Hong et al.2014|A time domain approach to diagnose gearbox fault based on measured vibration signals
    ES2702475T3|2019-03-01|Verification of wind turbine gondola yaw position sensor and yaw control system
    Chen et al.2013|Detecting of transient vibration signatures using an improved fast spatial–spectral ensemble kurtosis kurtogram and its applications to mechanical signature analysis of short duration data from rotating machinery
    US9482647B2|2016-11-01|Gear fault detection
    EP3260836B1|2021-08-04|Vibration monitoring and diagnosing system for wind power generator
    DE102010034749A1|2012-02-23|Device for monitoring a rotating machine part
    ES2435843T3|2013-12-23|Procedure for detecting gear damage
    Zhang et al.2019|Fault diagnosis of sun gear based on continuous vibration separation and minimum entropy deconvolution
    Ha et al.2017|Toothwise fault identification for a planetary gearbox based on a health data map
    CN104019831B|2017-01-04|Gyroscope method for diagnosing faults based on EMD and entropy weight
    Schmidt et al.2019|A discrepancy analysis methodology for rolling element bearing diagnostics under variable speed conditions
    D'Elia et al.2017|On the identification of the angular position of gears for the diagnostics of planetary gearboxes
    ES2774006T3|2020-07-16|Method and device for providing a condition assessment of a mechanical structure that includes a rotating machine component
    KR102068446B1|2020-01-20|Wind Power Electric Generatiom System Condition Monitoring System
    ES2754278B2|2022-01-25|DIAGNOSIS OF FAULTS IN PLANETARY GEARS
    KR20150137868A|2015-12-09|Apparatus and method for vibration signal extraction of planetary gear
    US10444254B2|2019-10-15|Virtual tachometers based on time series filtering
    CN104417565B|2018-07-06|The system and method that sensor merges automobile speedestimate
    Zhang et al.2020|Motion periods of sun gear dynamic fault meshing positions in planetary gear systems
    Hong et al.2016|Vibration based diagnosis for planetary gearboxes using an analytical model
    CN108106838B|2020-02-04|Whole-period time domain regression and alarm method for mechanical fault precision diagnosis
    US11205025B2|2021-12-21|Gearbox monitoring
    Wang et al.2000|A model-based gear diagnostic technique
    同族专利:
    公开号 | 公开日
    ES2754278R1|2021-07-01|
    ES2754278B2|2022-01-25|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US7912659B2|2004-06-28|2011-03-22|General Electric Company|System and method for monitoring the condition of a drive train|
    FR2957668B1|2010-03-19|2012-10-19|Eurocopter France|MECHANICAL ASSEMBLY HAVING STRUCTURAL ANOMALY MONITORING MONITOR, TRANSMISSION BOX PROVIDED WITH SUCH A MECHANICAL ASSEMBLY, AND METHOD FOR MONITORING STRUCTURAL ANOMALY|
    CN103398843B|2013-07-01|2016-03-02|西安交通大学|Based on the epicyclic gearbox sun gear Fault Classification of many classification Method Using Relevance Vector Machines|
    CN106769000B|2016-11-10|2019-03-26|哈尔滨工业大学|A kind of transmission path of the wind turbine gearbox fault vibration signal based on power flow finite element method determines method|
    CN107608936B|2017-09-22|2020-09-15|桂林电子科技大学|Method for extracting compound fault characteristics of planetary gear box|
    法律状态:
    2020-04-16| BA2A| Patent application published|Ref document number: 2754278 Country of ref document: ES Kind code of ref document: A2 Effective date: 20200416 |
    2022-01-25| FG2A| Definitive protection|Ref document number: 2754278 Country of ref document: ES Kind code of ref document: B2 Effective date: 20220125 |
    优先权:
    申请号 | 申请日 | 专利标题
    ES201830943A|ES2754278B2|2018-09-28|2018-09-28|DIAGNOSIS OF FAULTS IN PLANETARY GEARS|ES201830943A| ES2754278B2|2018-09-28|2018-09-28|DIAGNOSIS OF FAULTS IN PLANETARY GEARS|
    [返回顶部]