WO2024082103A1

WO2024082103A1 - Method and apparatus for detecting gearbox fault

Info

Publication number: WO2024082103A1
Application number: PCT/CN2022/125740
Authority: WO
Inventors: 邱志; 霍华; 潘智慧; 刘泳
Original assignee: 舍弗勒技术股份两合公司
Priority date: 2022-10-17
Filing date: 2022-10-17
Publication date: 2024-04-25

Abstract

A method for detecting a gearbox fault, comprising: acquiring detection data of a gearbox (S110); dividing the detection data into a plurality of groups of sub-band data (S120); acquiring a feature value of each group of sub-band data in the plurality of groups of sub-band data (S130); and according to all of the acquired feature values, determining whether the gearbox has a fault (S140). The accuracy of fault detection of the gearbox can be improved by means of dividing the detection data into a plurality of groups of sub-band data, and determining the fault state of the gearbox by means of each group of sub-band data, and the impact of noise on fault detection of the gearbox is reduced.

Description

Method and device for detecting gearbox faults

Technical Field

The present application relates to the field of data processing, and in particular, to a method and a device for detecting a fault of a gearbox.

Background technique

As the power transmission system of vehicles, gearboxes play a vital role in the safe operation of vehicles, etc. Therefore, it is necessary to detect the fault status of gearboxes.

Usually, when detecting the fault state of a gearbox, the vibration signal of the gear is analyzed by a spectrum analysis method such as cepstrum analysis. For example, the meshing frequency and sideband energy changes of the gearbox gears are detected by the spectrum analysis method to determine whether the gearbox has a fault. However, this detection method is easily affected by the noise from the road surface and the vibration of the vehicle body, resulting in the meshing frequency and sideband energy being submerged in the noise and unable to be accurately detected, which in turn leads to the inability to accurately detect the fault of the gearbox. For example, when a gear fault occurs during the operation of the vehicle, it is impossible to perform timely and accurate detection.

Therefore, an effective gear fault detection method is very necessary. A method that can accurately detect gearbox faults is needed.

Summary of the invention

One aspect of the present disclosure provides a method for detecting a gearbox fault, the method comprising: acquiring detection data of the gearbox; dividing the detection data into multiple groups of sub-band data; acquiring characteristic values of each group of sub-band data in the multiple groups of sub-band data; and determining whether the gearbox is faulty based on all the acquired characteristic values.

Another aspect of the present disclosure provides a device for detecting a gearbox fault, the device comprising: a data acquisition unit configured to acquire detection data of the gearbox; a data division unit configured to divide the detection data into multiple groups of sub-band data; a characteristic value acquisition unit configured to acquire characteristic values of each group of sub-band data in the multiple groups of sub-band data; and a fault determination unit configured to determine whether the gearbox has a fault based on all acquired characteristic values.

Another aspect of the present disclosure provides a computer-readable medium storing instructions, which, when executed by a processor, causes the processor to perform the above method for detecting a fault of a gearbox according to an embodiment of the present disclosure.

According to the method and device for detecting gearbox faults disclosed in the present invention, the accuracy of gearbox fault detection can be improved and the influence of noise on gearbox fault detection can be reduced by dividing the detection data into multiple groups of sub-band data and determining the fault state of the gearbox through each group of sub-band data.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solution of the embodiment of the present invention, the following is a brief introduction to the drawings required for use in the embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

FIG. 1 shows a flow chart of a method for detecting a fault of a gearbox according to an exemplary embodiment of the present disclosure.

FIG. 2 shows a flow chart of a method for detecting a fault of a gearbox according to another exemplary embodiment of the present disclosure.

FIG. 3 shows a flow chart of a method for detecting a fault of a gearbox according to another exemplary embodiment of the present disclosure.

FIG. 4 shows a flow chart of a method for detecting a fault of a gearbox according to another exemplary embodiment of the present disclosure.

FIG. 5 shows a flow chart of a method for detecting a fault of a gearbox according to another exemplary embodiment of the present disclosure.

FIG6 shows a block diagram of an apparatus for detecting a fault of a gearbox according to an exemplary embodiment of the present disclosure.

Detailed ways

The features and exemplary embodiments of various aspects of the present invention will be described in detail below. In the detailed description below, many specific details are proposed to provide a comprehensive understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present invention by illustrating examples of the present invention. The present invention is by no means limited to any specific configuration and algorithm proposed below, but covers any modification, replacement and improvement of elements, parts and algorithms without departing from the spirit of the present invention. In the accompanying drawings and the following description, known structures and technologies are not shown to avoid unnecessary ambiguity to the present invention.

The gearbox here can be the gearbox of any device having a power transmission system. For example, the gearbox can be the gearbox of a vehicle traveling on the road.

1 , in step S110 , detection data of the gearbox is acquired.

In one embodiment, the detection data may be vibration data or sound data. For example, the detection data may be vibration data detected by an acceleration sensor for a gearbox, or sound data detected by a microphone for a gearbox. For example, the detection data may be data measured from a gearbox as the vehicle travels, that is, measurement data in the time domain.

In step S120, the detection data is divided into a plurality of groups of sub-band data, wherein each group of sub-band data corresponds to a frequency band.

In one embodiment, in step S120, the detection data may be divided into a plurality of groups of sub-band data in a plurality of frequency bands according to a predetermined octave within a predetermined frequency range.

As an example, in order to effectively divide the detection data whether it is vibration data or sound data, the above predetermined frequency range can be set to a frequency range of 20Hz to 20KHz, and the above predetermined octave can be set to 1/3 octave. In this case, the detection data can be divided into 31 groups of sub-band data corresponding to 31 frequency bands, and the lower cutoff frequency _fn and the upper cutoff frequency fn ₊₁ of each frequency band satisfy the following equation:

f _n+1 = 2 ^1/3 f _n

For example, the detection data can be filtered using filters corresponding to each frequency band to obtain each group of sub-band data of each frequency band. It should be understood that the frequency range and octave used to divide the detection data are only examples, and any other frequency range and octave can be set according to actual needs.

In this way, the influence of noise (for example, noise generated by the road and the vibration of the vehicle body during the driving of the vehicle) on the fault detection of the gearbox can be reduced.

In step S130 , a feature value of each group of sub-band data among the multiple groups of sub-band data is obtained.

Here, the characteristic value may be any characteristic value that can indicate the health state (fault state) of the gearbox.

In step S140, it is determined whether the gearbox fails based on all acquired characteristic values.

Since the characteristic values acquired in step S130 can indicate the health status of the gearbox, each characteristic value can be used in step S140 to determine whether the gearbox fails.

According to the method for detecting gearbox faults disclosed in the present invention, the accuracy of gearbox fault detection can be improved and the influence of noise on gearbox fault detection can be reduced by dividing the detection data into multiple groups of sub-band data and determining the fault state of the gearbox through each group of sub-band data.

1 and 2 , steps S110 , S120 and S140 in FIG. 2 are respectively the same as steps S110 , S120 and S140 in FIG. 1 , except that an example of step S130 in FIG. 1 is illustrated in detail in steps S131 , S132 and S133 in FIG. 2 .

In step S131 , the envelope of each group of sub-frequency band data in the time domain may be detected to obtain envelope data of each group of sub-frequency band data.

In one embodiment, Hilbert transform may be performed on each group of sub-band data to detect the envelope of the group of sub-band data to obtain its envelope data.

In step S132, a fast Fourier transform may be performed on the envelope data of each group of sub-band data to obtain frequency domain sub-band data of each group of sub-band data.

Here, since the envelope data of each group of sub-band data is subjected to a fast Fourier transform, the frequency domain sub-band data of each group of sub-band data obtained can be distributed in a frequency band with a lower frequency, so as to facilitate subsequent processing. For example, in the case where the detection data is divided into 31 groups of sub-band data in 31 frequency bands according to 1/3 octave in the frequency range of 20 Hz to 20 KHz, the frequency domain sub-band data of each group of sub-band data obtained in step S132 can be distributed in a frequency range of several Hz to several hundred Hz.

In step S133, the modulation frequency corresponding to the frequency domain sub-band data of each group of sub-band data may be obtained. Here, the modulation frequency is the frequency with the maximum amplitude on the amplitude-frequency curve corresponding to the frequency domain sub-band data.

For example, after obtaining the frequency domain sub-band data of each group of sub-band data, an amplitude-frequency curve corresponding to the frequency domain sub-band data can be obtained, with the horizontal axis being frequency and the vertical axis being amplitude. Then, the frequency with the largest amplitude on the amplitude-frequency curve can be determined as the modulation frequency.

The modulation frequency obtained as above may be a characteristic value of the corresponding group of sub-band data. Afterwards, in step S140, it may be determined whether the gearbox is faulty according to each modulation frequency of each group of sub-band data. An example of determining whether the gearbox is faulty according to each modulation frequency is described below with reference to FIG. 3.

2 and 3 , steps S110 , S120 and S130 in FIG. 3 are respectively the same as steps S110 , S120 and S130 in FIG. 2 , except that FIG. 3 illustrates an example of step S140 in FIG. 2 in detail with steps S141 , S142 and S143 .

3 , in step S141 , it may be determined whether there is a faulty modulation frequency among all the acquired modulation frequencies. Here, the faulty modulation frequency may be a modulation frequency that is the same as a faulty frequency in a predetermined faulty frequency set.

In one embodiment, step S141 may include: determining the rotation speed of the gearbox; acquiring a fault frequency set corresponding to the rotation speed; and determining whether there is a fault modulation frequency identical to a fault frequency in the fault frequency set among all acquired modulation frequencies.

For example, different fault frequency sets corresponding to different rotational speeds of the gearbox may be pre-stored. After determining the rotational speed of the gearbox when the detection data is obtained, the fault frequency set corresponding to the rotational speed may be retrieved, and then, based on the retrieved fault frequency set, it may be determined whether there is a faulty modulation frequency in the modulation frequencies corresponding to each group of sub-band data.

In the case where it is determined in step S141 that a faulty modulation frequency exists (“Yes” in FIG. 3 ), step S142 may be executed to determine that a gearbox fault occurs.

In the case where it is determined in step S141 that there is no faulty modulation frequency (“No” in FIG. 3 ), step S143 may be executed to determine that there is no fault in the gearbox.

Here, in the case where it is determined that the gearbox has a fault, in order to further detect the fault, in one embodiment, the method for detecting a gearbox fault according to the present disclosure may also include: determining a frequency band corresponding to a fault modulation frequency, which frequency band is a frequency band among multiple frequency bands into which multiple groups of sub-band data are divided; and determining the location where the gearbox fault occurs based on the frequency band.

For example, when a fault occurs in the gears of a gearbox and the various components of the bearings of the gearbox (for example, the outer ring, the inner ring, the rollers, etc.), vibrations or sound signals in different frequency bands will be generated. Therefore, the fault location of the gearbox can be determined by determining the frequency band of the sub-band data corresponding to the fault modulation frequency determined in step S141.

In addition, in order to further accurately detect the state of the gearbox (regardless of whether the gearbox has been detected to be faulty), the method for detecting gearbox faults according to the present disclosure can also further determine the "roughness" of the detection data indicating the degree of gearbox fault, as shown in Figure 4 below.

3 and 4 , steps S110 , S120 and S140 in FIG. 4 are respectively the same as steps S110 , S120 and S140 in FIG. 3 , except that step S130 in FIG. 4 further includes step S134 , and FIG. 4 further includes step S150 .

In step S134, the modulation depth corresponding to the frequency domain sub-band data of each group of sub-band data may be obtained. Here, the modulation depth may be the sum of the amplitude of the modulation frequency and the amplitude of the multiple frequency of the modulation frequency on the above amplitude-frequency curve.

For example, after obtaining the amplitude-frequency curve corresponding to the frequency domain sub-band data of each group of sub-band data and obtaining the frequency with the largest amplitude on the amplitude-frequency curve (i.e., the modulation frequency), the frequency of 2 times, 3 times, ..., m times the frequency can be obtained, and the sum of the amplitudes of these frequencies can be obtained as the modulation depth. For example, m can be any value set according to actual needs, for example, m=5. It should be understood that the above values are only examples, and these values can be set to any values according to actual needs.

In step S150, the roughness of the detection data may be determined based on all the acquired modulation frequencies and all the modulation depths, and the roughness indicates the degree of fault of the gearbox.

In one embodiment, the roughness can be expressed by the following equation:

R _VA is the roughness, k is a constant coefficient determined according to the working condition of the gearbox (for example, for the gearbox of a vehicle, under the normal driving state of the vehicle, k can be set to 0.25), _fi is the modulation frequency corresponding to the i-th group of sub-band data, ΔL _i is the modulation depth corresponding to the i-th group of sub-band data, N is the total number of multiple groups of sub-band data, i and N are both integers, and 1≤i≤N.

By using the roughness R _VA determined as above, the degree of fault of the gearbox can be determined. For example, in the case of a fault in the gearbox, the greater the roughness R _VA , the more serious the fault of the gearbox. In addition, in the case of no fault in the gearbox, the degree of fault can indicate a tendency that the gearbox may fail, for example, the greater the roughness R _VA , the greater the tendency that the gearbox may fail. That is, the roughness R _VA can indicate the impact of the vibration signal or sound signal of the gearbox.

Thereby, the condition of the gearbox can be determined more precisely.

Furthermore, as an alternative embodiment, the “roughness” of the detection data may be used to determine whether the gearbox fails, as shown in FIG. 5 below.

4 and 5 , steps S110 , S120 and S130 in FIG. 5 are respectively the same as steps S110 , S120 and S130 in FIG. 4 , except that: in FIG. 5 , step S150 of determining the roughness is performed after step S130 , and then in step S140 , it is determined whether the gearbox is faulty based on the determined roughness.

Here, the operation performed in step S150 may be the same as the operation performed in step S150 described in FIG. 4 , and will not be described in detail here.

After the roughness is determined, in step S140, for example, the determined roughness can be compared with a predetermined fault threshold, and when the roughness is less than the fault threshold, it can be determined that the gearbox is not faulty, and when the roughness is equal to or greater than the fault threshold, it can be determined that the gearbox is faulty. The fault threshold here can be set to different values according to different equipment on which the gearbox is installed and/or different working conditions of the gearbox.

In addition, after determining whether the gearbox fails as described above, the degree of the failure may be further determined according to the magnitude of the roughness value as described above.

In addition, it should be understood that in this embodiment, it is also possible to determine whether a faulty modulation frequency occurs according to each modulation frequency as described above, and determine the fault location according to the frequency band corresponding to the faulty modulation frequency.

As shown in FIG. 6 , the device 100 for detecting a fault of a gearbox according to an embodiment of the present disclosure includes a data acquiring unit 110 , a data dividing unit 120 , a feature value acquiring unit 130 , and a fault determining unit 140 .

Specifically, the acquisition unit 110 is configured to acquire detection data of the gearbox.

In one embodiment, the detection data may be vibration data or sound data. As an example, the detection data is detection data in the time domain.

The data dividing unit 120 is configured to divide the detection data into a plurality of groups of sub-band data.

In one embodiment, the data dividing unit 120 may divide the detection data into a plurality of groups of sub-band data in a plurality of frequency bands according to a predetermined octave within a predetermined frequency range.

In one example, the predetermined frequency range is a frequency range of 20 Hz to 20 kHz, and the predetermined octave is a 1/3 octave.

The feature value acquisition unit 130 is configured to acquire a feature value of each group of sub-band data among the multiple groups of sub-band data.

In one embodiment, the characteristic value acquisition unit 130 may be configured to: detect the envelope of each group of sub-band data in the time domain to obtain the envelope data of each group of sub-band data; perform fast Fourier transform on the envelope data of each group of sub-band data to obtain the frequency domain sub-band data of each group of sub-band data; and obtain the modulation frequency corresponding to the frequency domain sub-band data of each group of sub-band data. Here, the modulation frequency is the frequency with the maximum amplitude on the amplitude-frequency curve corresponding to the frequency domain sub-band data.

In addition, in one embodiment, the characteristic value acquisition unit 130 may also be configured to: acquire the modulation depth corresponding to the frequency domain sub-band data of each group of sub-band data. Here, the modulation depth is the sum of the amplitude of the modulation frequency on the above amplitude-frequency curve and the amplitude of a predetermined multiple frequency of the modulation frequency.

The fault determination unit 140 is configured to determine whether the gearbox fails according to all acquired characteristic values.

In one embodiment, the fault determination unit 140 may be configured to: determine whether there is a faulty modulation frequency among all acquired modulation frequencies, the faulty modulation frequency being a modulation frequency identical to a faulty frequency in a predetermined fault frequency set; and determine that a gearbox fault occurs if there is a faulty modulation frequency.

In one embodiment, the fault determination unit 140 can be configured to determine whether there is a fault modulation frequency among all the acquired modulation frequencies as follows: determine the rotational speed of the gearbox; obtain a set of fault frequencies corresponding to the rotational speed; and determine whether there is a fault modulation frequency among all the acquired modulation frequencies that is the same as a fault frequency in the set of fault frequencies.

In one embodiment, the fault determination unit 140 may be further configured to: determine a frequency band corresponding to the fault modulation frequency when there is a fault modulation frequency, the frequency band being a frequency band among multiple frequency bands; and determine a location where the gearbox fault occurs according to the frequency band.

Furthermore, in one embodiment, the fault determination unit 140 may be further configured to determine the roughness of the detection data according to all the acquired modulation frequencies and all the modulation depths, wherein the roughness indicates the degree of the gearbox fault.

In other words, the apparatus 100 for detecting a fault of a gearbox according to an embodiment of the present disclosure may perform the method for detecting a fault of a gearbox according to an embodiment of the present disclosure described above with reference to FIGS. 1 to 5 .

The detection data, the division of the detection data, the acquisition of the characteristic values and the determination of the gearbox fault have been described in detail above with reference to FIGS. 1 to 5 , and will not be repeated here.

According to the device for detecting gearbox faults disclosed in the present invention, the accuracy of gearbox fault detection can be improved and the influence of noise on gearbox fault detection can be reduced by dividing the detection data into multiple groups of sub-band data and determining the fault state of the gearbox through each group of sub-band data.

According to an embodiment of the present disclosure, there is also provided a computer-readable medium storing instructions, which, when executed by a processor, may cause the processor to execute the method for detecting a fault of a gearbox according to the above embodiment of the present disclosure.

It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.

The above is only a specific implementation of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be covered within the protection scope of the present invention.

Claims

A method for detecting a fault in a gearbox, comprising:

Acquiring detection data of the gearbox;

Dividing the detection data into multiple groups of sub-band data;

Acquire a characteristic value of each group of sub-band data in the plurality of groups of sub-band data; and

It is determined whether the gearbox fails according to all the acquired characteristic values.
The method according to claim 1, wherein dividing the detection data into a plurality of groups of sub-band data comprises:

The detection data is divided into a plurality of groups of sub-band data in a plurality of frequency bands according to a predetermined octave within a predetermined frequency range.
The method according to claim 2, wherein the detection data is detection data in the time domain,

Wherein, obtaining the characteristic value of each group of sub-band data in the multiple groups of sub-band data includes:

Detecting the envelope of each group of sub-frequency band data in the time domain to obtain envelope data of each group of sub-frequency band data;

Performing fast Fourier transform on the envelope data of each group of sub-band data to obtain frequency domain sub-band data of each group of sub-band data; and

A modulation frequency corresponding to the frequency domain sub-band data of each group of sub-band data is obtained, wherein the modulation frequency is a frequency having a maximum amplitude on an amplitude-frequency curve corresponding to the frequency domain sub-band data.
The method according to claim 3, wherein determining whether the gearbox fails based on all acquired characteristic values comprises:

determining whether there is a fault modulation frequency among all the acquired modulation frequencies, wherein the fault modulation frequency is a modulation frequency that is the same as a fault frequency in a predetermined fault frequency set; and

In the presence of the fault modulation frequency, it is determined that the gearbox is faulty.
The method according to claim 4, wherein determining whether there is a faulty modulation frequency among all the acquired modulation frequencies comprises:

determining a rotational speed of the gearbox;

Acquire a set of fault frequencies corresponding to the rotation speed; and

It is determined whether there is a fault modulation frequency that is the same as a fault frequency in the fault frequency set among all the acquired modulation frequencies.
The method according to claim 4, further comprising:

In the case where the fault modulation frequency exists, determining a frequency band corresponding to the fault modulation frequency, wherein the frequency band is a frequency band among the multiple frequency bands; and

A location where a fault of the gearbox occurs is determined based on the frequency band.
The method according to claim 3, wherein obtaining the characteristic value of each group of sub-band data in the multiple groups of sub-band data further comprises:

A modulation depth corresponding to the frequency domain sub-band data of each group of sub-band data is obtained, wherein the modulation depth is the sum of the amplitude of the modulation frequency on the amplitude-frequency curve and the amplitude of a predetermined multiple of the modulation frequency.
The method according to claim 7, further comprising:

The roughness of the detection data is determined according to all the acquired modulation frequencies and all the modulation depths, wherein the roughness indicates the degree of failure of the gearbox.
The method according to claim 8, wherein the roughness is expressed by the following equation:

Among them, R VA is the roughness, k is a constant coefficient determined according to the working conditions of the gearbox, fi is the modulation frequency corresponding to the i-th group of sub-band data, ΔL i is the modulation depth corresponding to the i-th group of sub-band data, N is the total number of the multiple groups of sub-band data, i and N are both integers, and 1≤i≤N.
A device for detecting a fault in a gearbox, comprising:

A data acquisition unit, configured to acquire detection data of the gearbox;

A data division unit, configured to divide the detection data into a plurality of groups of sub-band data;

a feature value acquiring unit configured to acquire a feature value of each group of sub-band data in the plurality of groups of sub-band data; and

The fault determination unit is configured to determine whether the gearbox fails according to all acquired characteristic values.
The apparatus according to claim 10, wherein the data partitioning unit is configured to:

The detection data is divided into a plurality of groups of sub-band data in a plurality of frequency bands according to a predetermined octave within a predetermined frequency range.
The device according to claim 11, wherein the detection data is detection data in the time domain,

Wherein, the characteristic value acquisition unit is configured as follows:

Detecting the envelope of each group of sub-frequency band data in the time domain to obtain envelope data of each group of sub-frequency band data;

Performing fast Fourier transform on the envelope data of each group of sub-band data to obtain frequency domain sub-band data of each group of sub-band data; and

A modulation frequency corresponding to the frequency domain sub-band data of each group of sub-band data is obtained, wherein the modulation frequency is a frequency having a maximum amplitude on an amplitude-frequency curve corresponding to the frequency domain sub-band data.
The apparatus according to claim 12, wherein the fault determination unit is configured to:

Determine whether there is a fault modulation frequency among all the acquired modulation frequencies, wherein the fault modulation frequency is a modulation frequency that is the same as a fault frequency in a predetermined fault frequency set; and

In the presence of the fault modulation frequency, it is determined that the gearbox is faulty.
The apparatus according to claim 13, wherein the fault determination unit is configured to determine whether there is a faulty modulation frequency among all the acquired modulation frequencies as follows:

determining a rotational speed of the gearbox;

Acquire a set of fault frequencies corresponding to the rotation speed; and

It is determined whether there is a fault modulation frequency that is the same as a fault frequency in the fault frequency set among all the acquired modulation frequencies.
According to the apparatus of claim 13, the fault determination unit is further configured:

In the case where the fault modulation frequency exists, determining a frequency band corresponding to the fault modulation frequency, wherein the frequency band is a frequency band among the multiple frequency bands; and

A location where a fault of the gearbox occurs is determined based on the frequency band.
The apparatus according to claim 12, wherein the feature value acquisition unit is further configured to:

A modulation depth corresponding to the frequency domain sub-band data of each group of sub-band data is obtained, wherein the modulation depth is the sum of the amplitude of the modulation frequency on the amplitude-frequency curve and the amplitude of a predetermined multiple of the modulation frequency.
According to the apparatus of claim 16, the fault determination unit is further configured to:

The roughness of the detection data is determined according to all the acquired modulation frequencies and all the modulation depths, wherein the roughness indicates the degree of failure of the gearbox.