• 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:
    Method and device for monitoring a bearing fitted to a rotating device The method comprises: obtaining (E10) a vibratory signal acquired by an accelerometric sensor; elimination (E20) of a deterministic component of the vibratory signal; obtaining (E30), for a determined defect, a theoretical frequency of this defect and a maximum deviation around this theoretical frequency; the calculation (E40-E50), as a function of a cyclic frequency, of an integrated cyclic coherence of the vibratory signal processed; estimating (E60) a real frequency of the fault from the integrated cyclic coherence, the theoretical frequency of the fault and the maximum deviation; the calculation (E70) of a fault diagnostic indicator by summing M integrated cyclic coherences of the vibratory signal evaluated in M cyclic frequencies equal respectively to M harmonics of the estimated actual frequency of the fault; - The comparison (E80-E90) of the fault diagnostic indicator with a predetermined threshold, and if exceeded, the detection (E100) of the fault on the bearing.
    公开号:FR3076348A1
    申请号:FR1763312
    申请日:2017-12-28
    公开日:2019-07-05
    发明作者:Dany Abboud；Mohamed El Badaoui
    申请人:Safran SA；
    IPC主号:
    专利说明:

    Invention background
    The invention relates to the general field of machines or rotating devices, and relates more particularly to the bearings fitted to such machines or devices, such as, for example, ball or roller bearings.
    It has a privileged but nonlimiting application in the aeronautics field in which many rotating machines are used.
    Bearings, and in particular ball or roller bearings, are rotating mechanical devices widely used in rotating machines and whose role is to drive the transmission shafts in rotation. Such a bearing is generally composed of two coaxial rings (an inner ring and an outer ring) between which are placed rolling elements (balls or rollers), held in a cage. The rotary movement of the shaft is ensured by the movement of the rolling elements.
    Because of their elementary role in mechanical transmission chains, the bearings are permanently stressed and therefore weakened. They can undergo degradations of the surface condition or the shape of the inner or outer rings, as well as rolling elements. Damage to a bearing can cause the machine in which it is used to stop unexpectedly.
    In order to avoid such a situation, it is known to monitor the condition of the bearings fitted to a rotating machine in order to be able to detect early the presence of faults on these bearings. One of the techniques most used to carry out this monitoring is based on a vibrational analysis of signals acquired by means of sensors such as accelerometers placed near the bearings.
    The principle on which the vibration analysis is based is as follows: when a rolling element (ball or roller) of the bearing comes into contact with a defective surface of the bearing (due for example to chipping or cracking at the surface of one of the bearing rings), periodic shocks occur which produce a vibration at a specific frequency, characteristic of the defect affecting the bearing, and which change the statistical structure of the vibratory signal acquired by the accelerometer. A theoretical value of the specific frequency of the fault can be easily obtained from the geometric characteristics of the bearing and its kinematics. The detection of the defect affecting the bearing is then done by identifying in the vibratory signal acquired by the accelerometer, a component at the specific frequency of this defect, for example by performing a spectral analysis or an analysis of the envelope of the vibratory signal.
    However, current monitoring systems implementing vibration analysis face several practical difficulties.
    Indeed, there is today a need to monitor increasingly complex equipment, in particular in the aeronautical field, which may include several rotating elements in addition to the bearings (eg compressors, gears, blowers , etc.). This monitoring, for obvious reasons of congestion, must also be carried out with a limited number of accelerometers. As a result, the component representative of the defect affecting the bearing can be masked by noises coming from other sources of parasitic vibrations (ie from other rotating elements of the equipment considered), so that the methods used in the systems of current monitors (e.g. envelope analysis) are not always able to detect this fault. This can lead to late detection of the fault or even an inability to detect the fault.
    To overcome this drawback, it is possible to use source separation techniques. However, these techniques have a high computational cost, and are often impossible to implement in real time.
    In addition, a particular difficulty which appears when monitoring bearings by vibration analysis is that the actual frequencies of faults which affect these bearings may differ from the theoretical values calculated for the purposes of vibration analysis from the geometric characteristics of the bearing and of its kinematics. This is due to the sliding of the bearings on the one hand, and on the other hand, to the fact that damage to the bearing is generally accompanied by a phenomenon of friction which tends to slow down the rotation of the damaged bearing and to modify its cinematic.
    Document EP 2 693 176 describes a method for detecting faults in a bearing by vibration analysis which attempts to solve the aforementioned problems. This process is based on a preprocessing by means of an autoregressive filter of a vibratory signal acquired by means of an accelerometer placed on the housing of the system comprising the bearings to be monitored, followed by an analysis of the envelope of the signal. vibration resulting from the pretreatment. However, this detection method presents a complexity in terms of relatively high implementation, in particular due to the pretreatment carried out. In addition, the effectiveness of the pretreatment carried out largely depends on the determination of several parameters (eg order of the autoregressive filter, etc.) which can prove complicated in practice.
    Subject and summary of the invention
    The aim of the present invention is in particular to remedy the abovementioned shortcomings of the prior art by proposing a method for monitoring a bearing fitted to a rotating device, comprising:
    - A step of obtaining a vibration signal acquired by an accelerometric sensor, said vibration signal containing a vibration signature of the bearing;
    - a step of processing the vibration signal comprising the elimination of a deterministic component of the vibration signal;
    A step of obtaining, for a determined fault likely to affect the bearing, a theoretical frequency characteristic of this fault and a maximum deviation determined around this theoretical frequency;
    A step of calculating, as a function of a cyclic frequency, a so-called integrated cyclic coherence of the vibrational signal processed averaged over a predetermined spectral frequency band;
    A step of estimating an actual frequency of the fault from the integrated cyclical coherence, the theoretical frequency characteristic of the fault and the maximum deviation determined around this theoretical frequency;
    A step of calculating a fault diagnostic indicator by summing an integer M of integrated cyclic coherences of the vibratory signal evaluated in M cyclic frequencies equal respectively to M harmonics of the estimated actual frequency of the fault;
    - a step of comparing the fault diagnostic indicator with a predetermined threshold for this fault; and - in the event of the threshold being exceeded by the diagnostic indicator, a step of detecting the fault on the bearing.
    Correlatively, the invention also relates to a device for monitoring a bearing fitted to a rotating device, this monitoring device comprising:
    - A first obtaining module, configured to obtain a vibration signal acquired by an accelerometric sensor, said vibration signal containing a vibration signature of the bearing;
    - a vibration signal processing module, configured to eliminate a deterministic component of the vibration signal;
    A second obtaining module, configured to obtain, for a determined fault liable to affect the bearing, a theoretical frequency characteristic of this fault and a maximum deviation determined around this theoretical frequency;
    A first calculation module, configured to calculate, as a function of a cyclic frequency, a so-called integrated cyclic coherence of the vibrational signal processed averaged over a predetermined spectral frequency band;
    - an estimation module, configured to estimate an actual frequency of the fault from the integrated cyclic coherence, the theoretical frequency characteristic of the fault and the maximum deviation determined around this theoretical frequency;
    A second module for calculating a fault diagnostic indicator, configured to calculate said fault indicator by summing an integer M of integrated cyclic coherences of the vibratory signal evaluated in M cyclic frequencies equal respectively to M harmonics of the estimated actual frequency of the fault ;
    - a comparison module, configured to compare the fault diagnostic indicator with a predetermined threshold for this fault and to detect the fault on the bearing if the threshold is exceeded by the diagnostic indicator.
    The invention can be applied to various types of bearings, and in particular to a ball bearing or a roller bearing.
    The invention therefore proposes a technique for monitoring a bearing fitted to a rotating device based on a cyclostationary analysis of the vibratory signal acquired by an accelerometric sensor. The accelerometric sensor is placed in such a way that the vibration signal that it acquires contains a vibration signature of the bearing. It should be noted that for this purpose, the accelerometric sensor can be placed near the bearing or on the housing of the system which includes the bearing that one seeks to monitor, but it can also be placed at a point distant from the bearing. as soon as a sufficient signal-to-noise ratio is ensured (typically of the order of 5%).
    The cyclostationary analysis proposed by the invention makes it possible to extract from the vibratory signal an indicator for diagnosing a fault liable to affect the bearing (eg fault affecting an inner ring of the bearing, fault affecting an outer ring of the bearing, etc. .) which, when compared to a threshold determined for this fault, makes it possible to detect whether the bearing is affected or not by said fault. This diagnostic indicator is advantageously calculated after having applied a (preprocessing to the vibratory signal consisting in ridding it of its deterministic component; this deterministic component is typically generated by other rotating elements than the monitored bearing (which in turn contributes to the component vibration signal), and likely to interfere with the vibration signal acquired by the sensor.
    It is noted that the indicator derived in accordance with the invention is determined for a given fault likely to affect the bearing since it depends on an actual frequency of the fault estimated from the theoretical frequency characteristic of the fault and of a deviation maximum determined around this theoretical frequency, each fault resulting in a different theoretical frequency and a maximum possible deviation of the actual frequency of the fault from this theoretical frequency. Thus, by deriving several indicators relating to a plurality of faults liable to affect the bearing (eg an indicator associated with a defect on the inner ring of the bearing, an indicator associated with a defect on the outer ring of the bearing, an associated indicator to a fault on the bearing cage, an indicator associated with a fault on a rolling element of the bearing, etc.,), it is possible to easily identify the element or elements of the bearing which have a defect. This allows for targeted maintenance on the bearing.
    Advantageously, the cyclic coherence of the vibrational signal processed makes it possible to bring out a signature of the defect, even when it is "drowned" in significant background noise (linked for example to a plurality of defective rotating elements located nearby monitored bearing). In fact, the inventors had the judicious idea of exploiting the cyclostationary characteristic of the vibration signals generated by the faults liable to affect the bearings equipping the rotating machines. This cyclostationarity results in the presence in the vibration signals of a pattern of periodic pattern, exploited by the invention.
    This results in a robustness of the fault diagnosis indicators proposed by the invention. These indicators are also calculated taking into account an estimate of the “real” frequency of the fault, which may be different from the theoretical frequency characteristic of the fault which can be easily obtained from the geometric characteristics of the bearing and its kinematics. This avoids having indicators biased by an imprecise knowledge of the frequency of the fault. On the contrary, the diagnostic indicators proposed by the invention provide a reliable and precise estimate of the severity of the faults affecting, if necessary, the bearings monitored.
    The indicators proposed by the invention are easier to calculate so that the monitoring carried out thanks to the invention can be carried out in real time.
    The invention therefore allows a reliable and early detection of a fault affecting a bearing if necessary, and the location of this fault, with a low complexity of implementation.
    Furthermore, thanks to the invention, several bearings can be monitored in parallel by means of a single sensor (for example a single accelerometric sensor). The invention therefore makes it possible to minimize the number of sensors required on board a system including several bearings for monitoring these bearings, which results in a substantial gain in terms of size compared to certain techniques for monitoring the prior art.
    The invention therefore has a preferred but non-limiting application in the field of aeronautics. The rotating device, a bearing of which is monitored, can thus be embedded in an aircraft engine, the bearing monitored being for example a ball or roller bearing conventionally fitted to such an engine.
    Thus, the invention also relates to an aircraft engine comprising at least one bearing fitted to a rotating device of the aircraft engine, at least one accelerometric sensor capable of acquiring a vibratory signal containing a vibratory signature of the bearing, and a monitoring device of the bearing according to the invention.
    It is noted, however, that the invention can be applied in many other fields, such as for example wind turbines, car engines, gears, etc.
    In a particular embodiment of the invention, the vibratory signal was acquired by the accelerometric sensor during a steady state of the rotating device (in other words when its operating parameters, such as for example its average speed of rotation, its pressure, its temperature, its charge, etc., are constant or almost constant).
    However, this assumption is not limiting and the invention can also be applied when the vibratory signal was acquired during a variable speed of the rotating device, preferably when the variations of the speed remain of the order of 15-20%. . In this case, we consider in the invention "normalized" cyclic frequencies with respect to the average frequency of mechanical rotation of the bearing; such normalized frequencies are also more commonly referred to as "orders" and make it possible to overcome the variability of the rotation regime of the rotating device.
    In a particular embodiment, the processing step comprises a step of spectral whitening of the vibratory signal.
    Such a whitening of the spectrum of the vibratory signal is particularly simple and quick to implement. It consists in dividing the Fourier transform of the vibratory signal by its module while preserving its phase, and can therefore be carried out blindly, that is to say without knowing the frequencies characteristic of the deterministic component of the signal. Although the random component of the vibratory signal, which contains the signature of the fault affecting the bearing if necessary, may be weakened by this bleaching, this weakening has little impact on the detection of the fault due to the indicator diagnosis proposed by the invention which is calculated from the integrated cyclic coherence of the processed vibrational signal. The calculation of the integrated cyclic coherence of a signal in fact implicitly comprises a bleaching operation so that the bleaching operation carried out during the processing step has little or no effect on the diagnostic indicator. calculated.
    In a particular embodiment, the step of calculating the integrated cyclic coherence comprises:
    - the estimation, for a given cyclic frequency, of the cyclic correlation of the vibratory signal processed as a function of a spectral frequency;
    - the calculation, from the estimated cyclic correlation, of the cyclic coherence of the vibrational signal processed for said given cyclic frequency as a function of the spectral frequency; and - the average, over said predetermined spectral frequency band, of the amplitude squared of the cyclic coherence of the processed vibration signal calculated for said given cyclic frequency, the result of said average providing the integrated cyclic coherence for said given cyclic frequency.
    This way of calculating the integrated cyclic coherence makes it possible to preserve the information contained in the vibratory signal acquired by the sensor which relates to the fault affecting if necessary the bearing (and which is described by the cyclic frequency), while maximizing the signal-to-ratio over-noise thanks to the average carried out over a predetermined spectral frequency band.
    This frequency band is for example taken equal to [kl.ûf; k2.ûf], where ûf denotes the spectral resolution considered to estimate the cyclical coherence, and kl and k2 two real or whole numbers. Such a frequency range makes it possible, by appropriately adjusting the numbers kl and k2, on the one hand to minimize the contribution of stationary noise at low frequency (ie below kl.Af) while eliminating cyclic aliasing at high frequency (ie above k2.Af).
    It is noted that in a variant, it is possible in the calculated average, to raise the amplitude of the cyclical coherence to an order different from order 2, such as for example to order 1 or order 4. However, the inventors have found that the higher the order considered, the earlier the detection of the defect.
    In a particular embodiment, the cyclic correlation of the vibrational signal processed is estimated using a Welch estimator.
    This estimator has the advantage of having remarkable statistical properties and more specifically of providing an estimate having a low quadratic error.
    In addition, it is particularly simple and easy to implement, and requires a relatively low computation cost.
    Such an estimator is also very effective in a context where high rotational speeds are considered (eg when the rotating device is on board an aircraft such as an airplane or a helicopter).
    Of course, other estimators can be used to estimate the cyclic correlation of the vibrational signal processed, such as for example an estimator by cyclic periodogram, an estimator by smooth cyclic periodogram, an estimator by cyclic modulation spectrum, etc.
    In a particular embodiment, the step of estimating the real frequency of the fault comprises the calculation of the integrated cyclic coherence for a plurality of cyclic frequencies included in a defined interval between the theoretical frequency characteristic of the fault minus the defined maximum deviation for this theoretical frequency and the theoretical frequency characteristic of the fault plus the maximum deviation defined for this theoretical frequency, the actual frequency of the fault corresponding to the cyclic frequency among said plurality of cyclic frequencies for which the calculated integrated cyclic coherence is maximum.
    This embodiment makes it possible, by calculating the integrated cyclic coherence for a plurality of cyclic frequencies included in a defined interval as proposed, to compensate for the non-knowledge of the exact frequency of the fault affecting the bearing if necessary. This difficulty is resolved in this embodiment by considering that the actual frequency of the fault is that which maximizes the integrated cyclical coherence over the interval considered defined around the theoretical frequency of the fault. It should be noted that in the case of a healthy bearing (i.e. showing no fault), the value obtained for the actual frequency of the fault is not strictly linked to a fault; this is not annoying because, in this case, the indicator calculated as proposed in the invention will have a value which will not exceed the threshold considered during the comparison step (indeed, the probability that another rotating element coincides with the defined interval is zero or almost).
    In a particular embodiment, two consecutive cyclic frequencies of said plurality of cyclic frequencies, denoted respectively a and α + Δα, are chosen such that the ratio a / da is an integer.
    The application of this criterion to choose the cyclic frequencies at which the integrated cyclic coherence is calculated makes it possible to optimize the cost of calculation.
    In a particular embodiment, the integer M is between 6 and 10.
    The number of harmonics present in the vibration signal depends on the impulsivity of the vibration signal generated by this fault. In known manner, the impulsivity of a signal translates the presence of peaks of high amplitudes in this signal. However, this impulsivity itself depends, in a known manner, on the severity of the defect affecting the rotation. Consequently, considering a large value of M for determining the diagnostic indicator offers the possibility of taking into account more information on the condition of the bearing and the severity of the fault affecting this bearing. However, a large value of M can increase the estimation error tainting the calculated indicator and reduce its effectiveness in terms of early detection of the fault.
    On the contrary, considering a low value of M improves the early detection of the fault,
    The inventors have found in practice that a number M of between 6 and 10 presents a good compromise making it possible to ensure both early detection of a defect affecting the bearing while ensuring that the diagnostic indicator reflects sufficiently relevant the severity of this defect.
    In a particular embodiment, the monitoring method further comprises a step of notification of the fault comprising at least one item of information:
    - identification of the defective bearing;
    - an indication of a defective element on said bearing; and / or - an indication of the severity of the fault detected on the bearing.
    This allows to have an adapted maintenance on the monitored bearing.
    In a particular embodiment, the different steps of the monitoring method are determined by instructions from computer programs.
    Consequently, the invention also relates to a computer program on an information medium, this program being capable of being implemented in a monitoring device or more generally in a computer, this program comprising instructions adapted to the implementation of the steps of a monitoring method as described above.
    This program can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other desirable form.
    The invention also relates to an information medium readable by a computer, and comprising instructions of a computer program as mentioned above.
    The information medium can be any entity or device capable of storing the program. For example, the support may include a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or else a magnetic recording means, for example a floppy disk or a disc. hard.
    On the other hand, the information medium can be a transmissible medium such as an electrical or optical signal, which can be routed via an electrical or optical cable, by radio or by other means. The program according to the invention can in particular be downloaded from a network of the Internet type.
    Alternatively, the information medium can be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the process in question.
    Brief description of the drawings
    Other characteristics and advantages of the present invention will emerge from the description given below, with reference to the appended drawings which illustrate an embodiment thereof devoid of any limiting character. In the figures:
    - Figure 1 shows schematically a device for monitoring a bearing, according to the invention in a particular embodiment;
    - Figure 2 shows the hardware architecture of the bearing monitoring device of Figure 1; and
    FIG. 3 illustrates, in the form of a flowchart, the main steps of the monitoring method according to the invention in a particular embodiment in which it is implemented by the monitoring device of FIG. 1.
    Detailed description of an embodiment
    Figure 1 shows, in its environment, a monitoring device 1 of a bearing 2 fitted to a rotating mechanical device 3, according to the invention in a particular embodiment.
    In the example envisaged in FIG. 1, the bearing 2 is a ball bearing and the rotating device 3 equipped with this ball bearing is for example a compressor of a turbojet engine 4 according to the invention. The bearing 2 is assumed here to allow the rotation of a shaft of the compressor 3 of the turbojet engine 4 along a predefined axis of rotation.
    This example is however only given by way of illustration, and the invention can also be applied in other contexts. Thus for example, the bearing 2 can be a roller bearing, the rotating device 3 can be any mechanical rotating device of which an element is driven in rotation by means of the bearing 2, and other aircraft engines than a turbojet engine can be envisaged. The invention can also be applied in a context other than the aeronautical context.
    According to the invention, the monitoring device 1 is capable of operating a monitoring of the bearing 2, from a vibratory signal acquired by a sensor 5 equipping the turbojet engine 4 over at least one predefined period of time of duration T. The sensor 5 is an accelerometer placed here on the compressor housing 3 so as to pick up the vibrations emitted by the bearing 2 and more particularly by its elements (vibratory signature of the bearing within the meaning of the invention).
    In a manner known per se, a ball bearing is composed of various elements, and more particularly of two coaxial rings (a so-called internal or internal ring and a so-called external or external ring) between which are placed balls kept spaced apart by a cage. In this way, the balls can roll between the inner ring and the outer ring. The invention is intended to enable the detection of a defect affecting at least one of these elements.
    The placement of the accelerometer 5 to enable it to acquire a vibratory signal containing the vibratory signature of the bearing 2 poses no difficulty for those skilled in the art, and depends on the rotating mechanical device considered and on the context of use of the latter. . It is not described in detail here. In the case of a compressor shaft of a turbojet engine as envisaged in FIG. 1, the accelerometer 5 is for example placed on the compressor housing 3.
    As a variant, the accelerometer 5 can be placed in a place further from the bearing 2 proper since it makes it possible to acquire a signal containing a vibratory signature of the bearing 2 with a signal-to-noise ratio preferably greater than 5%.
    The temporal vibratory signal denoted X (t) acquired by the accelerometer 5 over the period T when the rotating mechanical device 3 is in rotation, is processed by an acquisition system integrated here in the accelerometer 5 comprising a conditioner, a filter analog anti-aliasing, a block sampler and an analog-to-digital converter. Such an acquisition system is known per se, and is not described in detail here. It delivers a digital signal sampled at a predefined sampling frequency Fs and derived from the vibratory signal X (t) acquired by the accelerometer. The sampling frequency Fs is chosen large enough to keep the kinematic and dynamic information of the bearing 2. In an aeronautical context such as that envisaged in FIG. 1, Fs is taken equal for example to 50kHz. It is assumed here that the period T over which the vibratory signal X is acquired is a multiple L of the sampling period, ie: T = L / Fs. L equivalently denotes the digital length of the vibratory signal (i.e. the number of samples Xb (n) derived by the acquisition system from the vibratory signal X (t) acquired by the accelerometer 5).
    The sampled vibratory signal (vibratory signal within the meaning of the invention), the samples of which are referenced in the following description by Xb (n), where n denotes an integer greater than 1, is then transmitted to the monitoring device 1 for analysis with a view to detecting the presence of a possible fault on the bearing 2. In the embodiment described here, this analysis is intended to be carried out in real time, and the monitoring device 1 is on board the aircraft powered by the turbojet 4 (for example in a computer of this turbojet 4).
    As a variant, the monitoring device 1 can be located in a remote device, for example in the example envisaged here, in a device on the ground capable of communicating via a telecommunications network with the accelerometer 5 or with a computer of the turbojet engine 4 able to recover the vibratory signal X (t) acquired by the accelerometer 5.
    In the embodiment described here, the monitoring device 1 has the hardware architecture of a computer, as illustrated in FIG. 2.
    It includes in particular a processor 6, a random access memory 7, a read-only memory 8, a non-volatile flash memory 9 as well as communication means 10 allowing in particular the monitoring device 1 to communicate with the accelerometer 5 to obtain the vibratory signal s (t) generated by bearing 2 and acquired by the latter. These communication means comprise for example a digital data bus or any other communication interface, in particular a communication interface on a telecommunications network, when the monitoring device 1 is not on board the aircraft propelled by the turbojet engine. 4.
    The ROM 8 of the monitoring device 1 constitutes a recording medium according to the invention, readable by the processor 6 and on which a computer program PROG according to the invention is recorded.
    The computer program PROG defines functional and software modules here, configured to implement a method for monitoring the bearing 2 according to the invention. These functional modules rely on and / or control the hardware elements 6-10 of the monitoring device 1 mentioned above. They notably include here, as illustrated in FIG. 1:
    - A first module for obtaining IA, configured to obtain the vibratory signal s (t) acquired by the accelerometer 5;
    - a processing module IB of the vibration signal s (t), configured to eliminate a deterministic component of the vibration signal;
    - A second obtaining module IC, configured to obtain for a determined defect noted d likely to affect the bearing 2, a theoretical frequency characteristic of this defect noted a maximum deviation noted δβ ά determined around this theoretical frequency. This maximum deviation characterizes the maximum possible deviation that is agreed between the theoretical frequency characteristic of the fault and the actual or exact frequency of this fault;
    - a first module 1D calculation configured to calculate, based on the cyclic frequency denoted here has been known an integrated cyclic consistency of the vibratory processed signal, averaged over a predetermined band of spectral frequencies chosen here so as to maximize the signal- over-noise;
    - an estimation module 1E, configured to estimate the real frequency of the defect noted β% * from the integrated cyclic coherence calculated by the first calculation module 1D, from the theoretical frequency // ^ characteristic of the defect and the deviation maximum δβ ά determined around this theoretical frequency;
    - a second module 1F for calculating a fault diagnostic indicator, configured to calculate this indicator by summing an integer M of integrated cyclic coherences of the processed vibrational signal evaluated in M cyclic frequencies equal respectively to M harmonics of the estimated actual frequency of the defect;
    - a comparison module IG, configured to compare the fault diagnostic indicator calculated by the second calculation module 1F with a predetermined threshold for this fault and to detect the fault on the bearing in the event of the threshold being exceeded by the indicator diagnostic.
    In the embodiment described here, the monitoring device 1 also has a notification module 1H, capable of notifying a user or a remote device of the existence of a fault on the bearing 2 if necessary. This 1H notification module can rely in particular on the communication means 10 of the monitoring device 1 or on input / output means of the latter, such as for example a screen or a microphone capable of signaling the detection of a fault on the bearing 2 to a user installed near the monitoring device 1.
    The functions of these different modules are described in more detail now with reference to the steps of the monitoring method according to the invention.
    FIG. 3 illustrates, in the form of a flowchart, the main steps of a monitoring method according to the invention in a particular embodiment in which it is implemented by the monitoring device 1 to monitor the bearing 2 fitted to the compressor 3 of the turbojet 4.
    It is assumed here that the accelerometer 5 is configured so as to acquire, over a period of time of duration T = L / Fs, a vibratory signal X (t) of acceleration generated by the bearing 2 while the compressor 3 is in rotation driven by the bearing 2. The accelerometer 5 is configured here to acquire the vibratory signal X (t) when the compressor 3 is operating in a stationary or quasi-stationary regime: in the example envisaged here, this means that the speed of rotation of the compressor and / or its load (ie its torque) are constant or quasi-constant (variation of less than 5% of the speed of rotation of the compressor and its load). It should be noted that the duration T should not be chosen too short so as to ensure the accuracy of the diagnostic indicators derived from the invention. For a sampling frequency
    Fs = 50kHz, a duration T = 2s represents a good compromise.
    The vibratory signal X (t) acquired by the accelerometer 5 is sampled by the acquisition system of the accelerometer 5, at the sampling frequency Fs. The sampled vibratory signal Xb (n), n = l, ..., L, is transmitted by the acquisition system of the accelerometer 5 to the monitoring device 1, and more particularly to its first module for obtaining IA ( step E10). The vibration signal sampled Xb (n) is designated more simply in the following description by vibration signal.
    The vibrating signal Xb (n), n = l, ..., L is supplied by the obtaining module IA to the processing module IB of the monitoring device 1. In a manner known per se, the vibrating signal Xb (n) , n = l, ..., L has a deterministic component and a random component; it is in this random component that the vibrations manifested, if any, are linked to the fault (s) affecting bearing 2.
    To better bring out the vibrations linked to the fault (s) affecting, if necessary, the bearing 2, the processing module IB processes the vibratory signal Xb (n), n = l, ..., L in order to eliminate its deterministic component (step E20).
    In the embodiment described here, the elimination of the deterministic component of the vibratory signal is carried out by the processing module IB via a spectral bleaching operation. To carry out this operation, the processing module IB first calculates the discrete Fourier transform of the vibratory signal Xb (n). We denote by DFT n _, m (Xb (n)} this discrete Fourier transform. It is defined in a known way by:
    DFT ^ k (X (rî)} =) X ( n ) e -j2nÇnMM <W
    2-t n = 0 where:
    P s Δ / “7 - = - J Δί. 147 W denotes the frequency resolution, W representing the size of the window on which the discrete Fourier transform is calculated, and
    At = -
    P s denotes the temporal resolution.
    Then the processing module IB divides the discrete Fourier transform obtained by the module of the discrete Fourier transform, and calculates the inverse discrete Fourier transform of the result obtained. In other words, the processing module IB calculates the following signal X (n), for n = l, ..., L (vibratory signal processed within the meaning of the invention):
    X (n) = IDFT ^ n
    ÇDFT n ^ m (Xb (ri)} ï (| DF7 '^ ni {XÔ (n) | J where / DF7 ^ n denotes the inverse discrete Fourier transform defined as follows y 1 <W ~ 1 IDFT ^ k (X (n)} = - YX (n) e + j2nWt) U <m
    W Zi n = o by considering the same notations used during the definition of the discrete Fourier transform.
    In this way, the phase of the vibratory signal Xb (n) is preserved.
    This laundering operation has the advantage of being simple to implement.
    As a variant, other techniques can be applied by the processing module IB to eliminate the deterministic component of the signal Xb (n). Such a technique is described for example in the document by N. Sawalhi et al. Entitled "Signal pre-whitening using cepstrum editing (liftering) to enhance fault detection in rolling element bearings", Proceedings of the 24th International Congress on Condition Monitoring and Diagnostics Engineering Management (COMADEM), 2011. It consists in canceling the entire complex cepstrum of the vibratory signal by keeping only the component of the signal relating to the zero frequency.
    The processed vibratory signal X (n), n = l, ..., L (and bleached) is transmitted to the first calculation module 1D of the monitoring device 1 so that it analyzes the cyclostationarity of the signal. Indeed, the presence of a defect on bearing 2 is manifested by a component in the cyclostationary signal which results in a periodic self-covariance function. This cyclostationarity property is due mainly on the one hand, to the recurrence of the fault on a periodic basis (linked to the rotation of the rotating device driven by the bearing), and on the other hand, to the presence of a fluctuation between the arrival time of impacts caused by the sliding of the bearing balls 2.
    Each fault d likely to affect a bearing is characterized by a specific frequency. No limitation is attached to the nature of the defect, which may relate to one or more elements of the bearing. Thus for example, the defect d may be a defect in the outer ring of the bearing 2, a defect in the inner ring of the bearing 2, a defect in the bearing cage 2 or a ball in the bearing 2. Each of these faults is characterized by a characteristic frequency of its own.
    This characteristic frequency of the fault d considered can be estimated (obtained) easily theoretically by the second obtaining module IC of the monitoring device 1 from knowledge of the geometric characteristics of the bearing 2 and its kinematics (step E30). The theoretical frequency characteristic of the fault is noted.
    d. The geometrical characteristics of the bearing 2 can be obtained easily from the technical sheet of the bearing 2. The kinematics of the bearing (ie its speed of rotation) can be obtained via a sensor placed in an appropriate manner within the compressor 3, in a known manner in itself. Thus, by way of illustration:
    - the theoretical theoretical frequency fd = bext of a defect d in the outer ring of the bearing is given by:
    fa = bext = ^ N B ^ l
    D B cosip
    D P J - the theoretical characteristic frequency f d = bint of a defect d of the inner ring of the bearing is given by:
    // = ^ = ^ (1 +
    D B cosxp
    Dp J - the characteristic theoretical frequency f d = c of a defect d in the bearing cage is given
    Z) b cosi / A
    Dp J - the characteristic theoretical frequency f d = b iii e of a defect d of a bearing ball is given by:
    or :
    - f r denotes the mechanical rotation frequency of the bearing 2;
    - N b denotes the number of balls of the bearing 2;
    - D b denotes the diameter of the balls of the bearing 2;
    - Dp denotes the average diameter of the bearing 2; and - ψ designates the contact angle of the bearing 2.
    The second obtaining module IC also obtains for each defect d considered, a maximum deviation δβ α determined around the theoretical frequency β ^ 1 characteristic of the defect d. In the example envisaged here, it is assumed for the sake of simplification that this maximum deviation is the same in percentage whatever the defect d envisaged, ie δβ ά = Αβ ^ 1 with A designating a real constant independent of the defect d. In practice, the inventors have found that it suffices to choose the maximum deviation equal to a few percent of the theoretical frequency characteristic of the defect; for example A = 3%.
    Then, in the embodiment described here, the first calculation module 1D of the monitoring device 1 calculates the cyclic coherence of the vibrational signal processed X (n), n = l, ..., L (step E40).
    In known manner, the cyclic coherence of a signal is a statistical measure which makes it possible to measure for each so-called cyclic frequency the degree of correlation between the signal and this same signal shifted in frequency. A coherence close to one for a cyclic cyclic frequency a indicates a strong correlation between the components of the signal considered at frequencies f and f- a.
    To calculate the cyclic coherence of the signal X (n) at the cyclic frequency a, n = l, .., L, the first calculation module 1D uses here a Welch estimator of the cyclic correlation of the signal, which is written from the following way, except for one factor:
    S 2 “ X W) => DFT44 k {w4n) X {n) e ^ anU } * DFT ^ k {w / n) X (n) e- ^ anÎit } or:
    - * designates the conjugation operator;
    - w (n) indicates a sliding window (ex. Hanning window, demisinus, etc.) and w s (n) = w s (n - sÆ) is the offset version of the sliding window with l <R <Nw, Nw designating an integer (eg a power of 2) representing the size of the Welch window, Nw-R designating the overlap between the windows;
    - S is the largest integer less than or equal to (L-Nw) / R + 1;
    - Af designates the frequency or spectral resolution, which is equal to i / (At.Nw). The size of the Welch Nw window will preferably be chosen so as to have a spectral resolution of the order of a few hundred hertz in the case of application envisaged here;
    - k is an integer, designating the spectral channel considered.
    We consider here as resolution of the cyclic frequency Aa = 1 / T to calculate the cyclic correlation.
    We note that in the Welch estimator defined above, the cyclic frequency a is a frequency expressed in hertz.
    Of course, other estimators can be used as a variant to estimate the cyclic correlation of the vibrational signal processed, such as for example an estimator by cyclic periodogram, an estimator by smooth cyclic periodogram, an estimator by cyclic modulation spectrum, etc.
    Then the 1D calculation module derives from the Welch estimator thus calculated the cyclic coherence of the signal X (n), n = l, ..., L as follows:
    __________S 2 a y (W) __________ (s ° x (kAf - afâS ^ kAf + a / 2)) V2
    It is noted that the cyclic correlation and the cyclic coherence thus calculated are functions of the spectral frequency kAf and are indexed by the cyclic frequency a (they are therefore also functions of this cyclic frequency), the latter being expressed in hertz. The spectral frequency brings out the dynamic characteristics of the system considered, while the cyclic frequency focuses on the modulations (mainly related to the fault affecting the bearing 2 if necessary). The inventors therefore wished to keep the cyclic information carried by the cyclic frequency, but averaged the information carried by the spectral frequency over a selected frequency band so as to maximize the signal-to-noise ratio.
    To this end, the first 1D calculation module calculates a so-called integrated cyclic coherence by evaluating the mean of the square of the amplitude of the cyclic coherence with respect to the cyclic frequency over a frequency band chosen so as to maximize the signal-to-signal ratio. over-noise (step E50). This frequency band is defined here by the interval [klA /; k2A /], where kl and k2 denote two integers. The integrated cyclic coherence calculated by the first 1D calculation module is thus given by the following expression:
    k2 = kl - kl + ΐΣ ^ * 0 ^ kl
    In the example envisaged here, for a sampling frequency Fs = 50 kHz, we choose klA / ~ lkHz and k2A / ~ 20kHz. These values make it possible to minimize the contribution of stationary noise at low frequency (below klA /) and to eliminate cyclic aliasing at high frequency (above k2Af).
    Of course, these values are given only by way of illustration and other values can be considered. In general, we will preferentially ensure that k2A / <Fs - a max where a max designates the maximum cyclic value sought and therefore depends on the theoretical characteristic frequency of the fault and its order of harmonics (the principle is taken into account here that the information useful for diagnosing a bearing defect is generally not found at the ends).
    It should also be noted that in the embodiment described here, the square of the amplitude of the cyclical coherence was considered in the average calculated to obtain the integrated cyclic correlation. As a variant, it is possible to envisage raising the cyclical coherence to another order than order 2, for example to order 1 or order 4.
    The integrated cyclic coherence calculated by the first calculation module 1D is then used by the monitoring device 1 to derive a diagnostic indicator for the fault d.
    More specifically, for this purpose, the monitoring device 1, via its estimation module 1E, firstly estimates the “real” (or even exact) frequency of the fault d (step E60). As mentioned previously and by definition, this real frequency characteristic of the fault is included in the interval [^ h - δβ ά ; β £ + δβ α . To estimate this real frequency, the estimation module 1E calculates the integrated cyclic coherence for a plurality of cyclic frequencies a included in this interval. In the embodiment described here, we assume a resolution denoted respectively Δα between two consecutive cyclic frequencies considered by the estimation module 1E chosen such that the ratio α / Δα is an integer. This criterion makes it possible to optimize the cost of calculating the cyclic coherence integrated into the different cyclic frequencies considered.
    Then the estimation module 1E estimates the actual frequency of the fault, noted, as being the value of the cyclic frequency corresponding to the maximum value of the integrated cyclic coherences calculated, that is:
    ΑΓ = argmax ^ _ ôl3dia ^ h + spd Î ^
    It is noted that in the case where the bearing 2 does not have the defect d, the value / £ ct has no meaning strictly speaking and does not have the frequency of a defect in the bearing. This is not a problem in itself; indeed, it is highly improbable that bearing 2 is affected by a defect distinct from defect d having a real frequency lying in the interval [β ^ 1 - δβα ·, β% 1 + δβα].
    Then the monitoring device 1, via its second calculation module 1F, calculates a fault diagnostic indicator d (step E70). This indicator, noted μ £, is the sum of the integrated cyclic coherences evaluated in M cyclic frequencies equal respectively to the M harmonics of the real frequency β ^ estimated of the fault, M denoting an integer, that is:
    M __ rfm = Σ / f τη = 1
    This indicator advantageously measures the cyclostationarity in the signal X, and consequently the contribution of the fault d associated with the frequency β ^.
    In the embodiment described here, M has a predetermined value chosen between 6 and 10.
    We note that M denotes the number of harmonics of the defect d considered. The number of harmonics present in the signal generated by the fault depends on the impulsivity of the signal, which in turn is linked to the severity of the fault. Thus, taking a large value of M provides more information on the state of the fault. However, taking a smaller value of M improves the early detection of the fault. Indeed, the inventors have found that in the case of a distributed fault, few harmonics are present in the integrated cyclic coherence, therefore the sum of a large number of harmonics leads to an increase in the error of estimation, and therefore reduces the efficiency of the indicator calculated in terms of earliness of detection.
    In an alternative embodiment, the estimation module 1F can estimate different diagnostic indicators by considering different values of the integer M.
    The diagnostic indicator μ ^ Μ) thus calculated is supplied to the comparison module IG of the monitoring device 1. This then compares the indicator μ * (Μ) with a predetermined alert threshold for the fault d, noted THR (d) (step E80). This threshold is chosen so as to allow detection of the fault d. It can be determined beforehand empirically by observing healthy bearings and bearings presenting the defect d, or by a statistical calculation under the assumption of stationarity in the healthy case. The THR (d) threshold value should not be chosen too high so as to ensure early detection while not being too low either to avoid multiplying false alarms.
    If the comparison module IG determines that the diagnostic indicator exceeds the threshold THR (d) (ie is greater than this) (answer yes in test step E90), then it detects that the bearing 2 is affected by fault d (step El00).
    In the embodiment described here, this detection is then notified by the 1H notification module via an alarm message, comprising all or part of the following information:
    - identification of the defective bearing (in this case, bearing 2);
    - indication of the defective element on the bearing (depends on the fault d detected); and - indication of the severity of the fault detected on bearing 2 (given by the value of the indicator μ $ (Μ)).
    Then monitoring of bearing 2 is resumed (repeating steps E10 to E100).
    It is noted that what has just been described for the defect d can be carried out for various faults liable to affect the bearing 2. In this way, the monitoring device 1 is able to carry out a differential diagnosis and to identify which defect affects bearing 2, that is to say to locate the origin or the origins of degradations undergone by bearing 2 (internal ring, external ring, balls, cage).
    In the embodiment described here, the vibratory signal X (t) acquired by the accelerometer 5 was acquired during a steady state of the compressor 3.
    In another embodiment of the invention, it is possible to consider a vibratory signal acquired when the speed of rotation of the compressor 3 experiences a certain variability, typically up to 1520%.
    In such a context, it is preferable to consider a normalized cyclic frequency denoted by with respect to the average rotation frequency f r in hertz of the bearing 2 over the acquisition period T considered, namely:
    at
    Such a standardized frequency is also more commonly called "order".
    In this embodiment, during step E40, the first calculation module 1D then uses the Welch estimator of the cyclic correlation of the phase corrected signal, which is written as follows, to within a factor:
    where θ (η) designates the angular measurement of a reference shaft (compressor 3 rotation shaft here) which allows to evaluate the rotation frequency f r of the bearing.
    The first 1D calculation module then derives from this Welch estimator the cyclic coherence with phase corrected of the signal X (n), n = l, ..., L as follows:
    y 2 a x (/ cA /) = _________S X W) _________ (S ° X <W ~ à.f r / 2) sl x (kàf + à. / R / 2)) 1/2
    The other steps of the method remain unchanged as soon as we consider standardized frequencies instead of frequencies expressed in hertz. Thus, the theoretical frequencies of the defects are the same as those previously introduced except for the factor f r (ie one divides these theoretical frequencies by f r ).
    Furthermore, in the embodiments described here, the signal from a single accelerometer has been considered for the sake of simplification. The invention can however be applied to several accelerometers. Likewise, it allows multiple bearings to be monitored simultaneously.
    The invention thus provides a robust monitoring technique for bearings integrated in a rotating device. The inventors have been able to observe, during various experiments, that this technique allows an early detection of the defects affecting the bearings, earlier than the technique of the prior art proposed in the document EP 2 693 176, and this in various contexts ( accelerometer placed on the defective bearing or distant from it, with or without electromagnetic interference).
    It should be noted that the invention has been described with reference to a vibratory signal acquired by means of an accelerometer. The monitoring method proposed by the invention can however also be applied to an acoustic signal, acquired for example by means of a microphone or any other acoustic sensor, and containing an acoustic signature of the bearing which it is desired to monitor.
    权利要求:
    Claims (15)
    [1" id="c-fr-0001]
    1. Method for monitoring a bearing (2) fitted to a rotating device (3), comprising:
    A step of obtaining (E10) a vibratory signal acquired by an accelerometric sensor, said vibratory signal comprising a vibratory signature of the bearing;
    - a processing step (E20) of the vibration signal comprising the elimination of a deterministic component of the vibration signal;
    A step for obtaining (E30), for a determined fault liable to affect the bearing, a theoretical frequency characteristic of this fault and a maximum deviation determined around this theoretical frequency;
    A calculation step (E40-E50), as a function of a cyclic frequency, of a so-called integrated cyclic coherence of the processed vibrational signal averaged over a predetermined spectral frequency band;
    A step of estimating (E60) a so-called real frequency of the fault from the integrated cyclic coherence, the theoretical frequency characteristic of the fault and the maximum deviation determined around this theoretical frequency;
    A calculation step (E70) of a fault diagnostic indicator by summing an integer M of integrated cyclic coherences of the vibratory signal evaluated in M cyclic frequencies equal respectively to M harmonics of the estimated real frequency of the fault;
    - a step of comparison (E80-E90) of the fault diagnostic indicator with a predetermined threshold for this fault; and - in the event of the threshold being exceeded by the diagnostic indicator, a step of detecting (E100) the fault on the bearing
    [2" id="c-fr-0002]
    2. Monitoring method according to claim 1 wherein the processing step comprises a step of spectral whitening of the vibratory signal.
    [3" id="c-fr-0003]
    3. Monitoring method according to claim 1 or 2 in which the step of calculating the integrated cyclic coherence comprises:
    - the estimation (E40), for a given cyclic frequency, of the cyclic correlation of the vibratory signal processed as a function of a spectral frequency;
    - the calculation (E40), from the estimated cyclic correlation, of the cyclic coherence of the vibrational signal processed for said given cyclic frequency as a function of the spectral frequency; and - the average (E50), over said predetermined spectral frequency band, of the amplitude squared of the cyclic coherence of the vibrational signal processed calculated for said given cyclic frequency, the result of said average providing the integrated cyclic coherence for said frequency cyclical given.
    [4" id="c-fr-0004]
    4. The monitoring method as claimed in claim 3, in which the cyclic correlation of the vibrational signal processed is estimated using a Welch estimator.
    [5" id="c-fr-0005]
    5. Monitoring method according to any one of claims 1 to 4 wherein the step of estimating (E60) of the actual frequency of the fault comprises the calculation of the integrated cyclic coherence for a plurality of cyclic frequencies included in an interval defined between the theoretical frequency characteristic of the fault minus the maximum deviation defined for this theoretical frequency and the theoretical frequency characteristic of the fault plus the maximum deviation defined for this theoretical frequency, the actual frequency of the fault corresponding to the cyclic frequency among said plurality of cyclic frequencies for which the calculated integrated cyclic coherence is maximum.
    [6" id="c-fr-0006]
    6. Monitoring method according to claim 5, in which two consecutive cyclic frequencies of said plurality of cyclic frequencies, denoted respectively a and α + Δα, are chosen such that the ratio a / da is an integer.
    [7" id="c-fr-0007]
    7. Monitoring method according to any one of claims 1 to 6 in which the integer M is between 6 and
    10.
    [8" id="c-fr-0008]
    8. Monitoring method according to any one of claims 1 to 7 further comprising a step of notification of said defect comprising at least one item of information:
    - identification of the defective bearing;
    - an indication of a defective element on said bearing; and / or - an indication of the severity of the fault detected on the bearing.
    [9" id="c-fr-0009]
    9. Monitoring method according to any one of claims 1 to 8 wherein the vibratory signal was acquired by the accelerometric sensor during a steady state of the rotating device.
    [10" id="c-fr-0010]
    10. Monitoring method according to any one of claims 1 to 9 wherein the cyclic frequency is normalized with respect to the rotation frequency of the bearing.
    [11" id="c-fr-0011]
    11. Computer program (PROG) comprising instructions for the execution of the steps of the monitoring method according to any one of claims 1 to 10 when said program is executed by a computer.
    [12" id="c-fr-0012]
    12. Recording medium readable by a computer on which a computer program is recorded according to claim 11.
    [13" id="c-fr-0013]
    13. Monitoring device (1) for a bearing (2) fitted to a rotating device (3), said monitoring device comprising:
    - a first obtaining module (IA), configured to obtain a vibration signal acquired by an accelerometric sensor, said vibration signal comprising a vibration signature of the bearing;
    - a vibration signal processing module (IB), configured to eliminate a deterministic component of the vibration signal;
    - a second obtaining module (IC), configured to obtain, for a determined fault liable to affect the bearing, a theoretical frequency characteristic of this fault and a maximum deviation determined around this theoretical frequency;
    A first calculation module (1D), configured to calculate, as a function of a cyclic frequency, a so-called integrated cyclic coherence of the processed vibrational signal averaged over a predetermined spectral frequency band;
    - an estimation module (1E), configured to estimate an actual frequency of the fault on the basis of the integrated cyclic coherence, the theoretical frequency characteristic of the fault and the maximum deviation determined around this theoretical frequency;
    - a second calculation module (1F) of a fault diagnostic indicator, configured to calculate said indicator by summing an integer M of integrated cyclic coherences of the vibratory signal evaluated in M cyclic frequencies equal respectively to M harmonics of the real frequency estimated defect;
    - a comparison module (IG), configured to compare the fault diagnostic indicator with a predetermined threshold for this fault and to detect the fault on the bearing if the threshold is exceeded by the diagnostic indicator.
    [14" id="c-fr-0014]
    14. Monitoring device according to claim 13 wherein the bearing (2) is a ball or roller bearing and the rotating device is on board an aircraft.
    [15" id="c-fr-0015]
    15. aircraft engine (4) comprising at least one bearing (2) fitted to a rotating device (3) of the aircraft engine, at least one accelerometric sensor (5) capable of acquiring a vibratory signal comprising a vibratory signature of said bearing , and a bearing monitoring device (1) according to claim 13 or 14.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3732457B1|2021-09-29|Method and device for monitoring a bearing equipping a rotary device
    EP2693176B1|2019-11-06|Method for detecting defects of a bearing by vibrational analysis
    EP2344728B1|2016-02-10|Method for monitoring a turbojet engine
    CA2328398C|2007-06-05|Detection of damage to engine parts
    CA2782043C|2018-03-20|Method and device for monitoring torsional vibrations of a rotary shaft of a turbine engine
    EP3250974B1|2021-09-08|Method, system and computer program for learning phase of an acoustic or vibratory analysis of a machine
    EP3658881B1|2021-07-07|Method and device for searching for a defect capable of affecting a rotating mechanical power transmission device
    EP3136111A1|2017-03-01|Method for measuring the rotational speed of a vehicle wheel
    EP0902263A1|1999-03-17|Method and apparatus for determining the vibrations of the rotor of a rotary machine
    EP3014779B1|2020-08-12|Monitoring the quality of a communication channel supported by a sliding contact
    FR3096133A1|2020-11-20|Extraction of a vibro-acoustic component generated by a mechanical source in variable regime
    FR3087265A1|2020-04-17|METHOD AND DEVICE FOR MONITORING A ROTATING MECHANICAL PART
    FR3079977A1|2019-10-11|METHOD AND DEVICE FOR MONITORING A ROTATING MACHINE
    EP3791235A1|2021-03-17|Method for monitoring the operation of a machine that generates vibrations and device for implementing such a method
    FR3058477A1|2018-05-11|METHOD FOR DETECTING THE FAILURE OF A MOBILE ELEMENT OF AN AIR GENERATOR
    同族专利:
    公开号 | 公开日
    WO2019129956A1|2019-07-04|
    EP3732457B1|2021-09-29|
    US20210063276A1|2021-03-04|
    EP3732457A1|2020-11-04|
    FR3076348B1|2020-01-17|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    FR2994261B1|2012-07-31|2014-07-18|Eurocopter France|METHOD OF DETECTING DEFECTS OF A BEARING BY VIBRATION ANALYSIS|CN110261116A|2019-07-08|2019-09-20|华南理工大学|A kind of Bearing Fault Detection Method and device|
    CN111189640A|2020-01-09|2020-05-22|珠海格力电器股份有限公司|Bearing fault monitoring method, monitoring device adopting same and washing machine|
    CN113029569A|2021-03-11|2021-06-25|北京交通大学|Train bearing autonomous fault identification method based on cyclic strength index|
    法律状态:
    2018-11-26| PLFP| Fee payment|Year of fee payment: 2 |
    2019-07-05| PLSC| Publication of the preliminary search report|Effective date: 20190705 |
    2019-11-20| PLFP| Fee payment|Year of fee payment: 3 |
    2020-11-20| PLFP| Fee payment|Year of fee payment: 4 |
    2021-11-18| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1763312A|FR3076348B1|2017-12-28|2017-12-28|METHOD AND DEVICE FOR MONITORING A BEARING EQUIPPED WITH A ROTATING DEVICE|
    FR1763312|2017-12-28|FR1763312A| FR3076348B1|2017-12-28|2017-12-28|METHOD AND DEVICE FOR MONITORING A BEARING EQUIPPED WITH A ROTATING DEVICE|
    US16/958,419| US20210063276A1|2017-12-28|2018-12-19|Method and device for monitoring a bearing equipping a rotary device|
    PCT/FR2018/053421| WO2019129956A1|2017-12-28|2018-12-19|Method and device for monitoring a bearing equipping a rotary device|
    EP18842779.3A| EP3732457B1|2017-12-28|2018-12-19|Method and device for monitoring a bearing equipping a rotary device|
    [返回顶部]