KR102102518B1 - Body of revolution fault analysis apparatus and method using transmission error - Google Patents

Abstract

The present invention relates to an apparatus and method for analyzing a rotating body defect using a transmission error, and more specifically, to determine not only a defect of a rotating body such as a gear, but also a defect type by applying an Ensemble Empirical Mode Deposition (EEMD). It relates to an apparatus for analyzing a defect of a rotating body using a transmission error that can be performed.

Description

Body of revolution fault analysis apparatus and method using transmission error

The present invention relates to an apparatus and method for analyzing a defect of a rotating body using a transmission error, and more specifically, to apply an ensemble empirical mode deposition (EEMD) to a transmission error (TE) to rotate gears, etc. The present invention relates to an apparatus and method for analyzing a defect of a rotating body using a transmission error that can determine not only the entire defect but also the defect type.

In general, the mechanical system is driven under physical friction conditions, so maintenance and repair can be prevented only by frequent maintenance.

As such, since the mechanical system requires frequent maintenance, there is a problem in that maintenance cost and time are high.

Due to these problems, condition-based maintenance (CBM), which performs maintenance only when there is a possibility of failure, has recently attracted attention. State-based maintenance performs maintenance only when there is a possibility of failure, thus reducing operating costs and increasing profits through reduced maintenance and time.

A field that can be expected to have a great effect by applying fault diagnosis can be a system that requires high cost for maintenance, and representatively, a gearbox such as a high-speed rail and a wind turbine.

Since the gearbox is a basic element used to transmit rotational power in a mechanical system, research into detecting and diagnosing defects through various sensor signals has been actively conducted.

Typically, gear defects include two spalls that generate vibration when a relatively large part of the surface falls off and gear engagement occurs, and cracks in which fatigue cracks grow in the vertical direction of the surface due to repeated bending stress. It can be.

Of these, cracks should be dealt with as a particularly important defect because once they occur, they rapidly lead to fracture (breakage) and, as a result, chain damage and overall gearbox damage.

As a method for diagnosing a gear defect, a method using a transmission error (TE) and a method using a vibration signal are used.

Transmission error (TE) means the error between the actual and ideal rotation angles that occur due to manufacturing errors or deformation due to loads when the gears rotate with each other as shown in Equations 1 and 2.

Here, θ ₁ , θ ₂ are the rotation angles of the two spur gears of the gearbox, and R ₁ and R ₂ are the radius of the spur gears.

Here, K _g is gear mesh stiffness (GMS), and T _out is driven gear torque.

This error causes the gearbox to vibrate, and it produces a whining noise. And since the shape change of the gear teeth changes the gear mesh stiffness (GMS), and this causes a change in the transmission error, research has been conducted to analyze the shape change of the tooth through the transmission error.

However, the method using the conventional transmission error has a problem in that it is possible to identify only the presence or absence of a defect through the transmission error or to grasp the degree of failure for a single defect, and not to determine whether the failure is caused by a spall or a crack.

In addition, there have been attempts to diagnose defects and identify types of defects using conventional vibration signals, but there are problems in that defects cannot be diagnosed or errors are increased due to the characteristics of vibration signals that identify various vibration factors as one signal.

Published Patent No. 10-2001-0044393 (2001.06.05.)

Therefore, the object of the present invention is to apply a transmission error (TE) and ensemble empirical mode decomposition (Ensemble Empirical Mode Deposition: EEMD) to apply a rotational body using a transmission error that can determine not only whether a defect such as a gear, but also a defect type It is to provide a defect analysis apparatus and method.

A rotating body defect analysis apparatus using a transmission error according to the present invention for achieving the above object is connected to a gear of a gearbox and rotates to output rotation signals for two or more of the gears rotating in the gearbox Encoder; A TE measuring unit which calculates a transmission error by a difference between a rotation signal output from the rotation encoder and a rotation signal for a normal gear, and outputs a transmission error signal; An RTE conversion unit for calculating a residual transmission error (RTE) according to the transmission error signal output from the TE measurement unit and outputting the residual transmission error signal accordingly; After inserting at least one random white noise that is different from the residual transmission error signal and performing empirical mode decomposition (EMD) processing, an ensemble average is obtained for the corresponding eigenmode function (IMF) signals for each white noise. An EEMD processing unit for outputting a residual transmission error mode signal that is a mode function (IMF) signal; And a defect determination unit for calculating a crest factor for a plurality of residual transmission error mode signals inputted from the EEMD processing unit and determining whether a gear outputting the rotation signal is normal or defective based on the calculated crest factor. Is done.

The defect determination unit checks whether the crest factor of each of the plurality of residual transmission error mode signals falls within a normal crest factor range, a spall fault crest factor range, or a crack fault crest factor range, and determines whether the gear is normal or spall fault, It is characterized by determining whether it is a crack defect.

The RTE converter may include a notch filter that notches and filters a gear meshing frequency signal and a harmonic frequency signal from the transmission error signal output from the TE measuring unit; And a high pass filter for filtering a low frequency signal having a predetermined low frequency or less from the notch filtered transmission error signal and outputting a residual transmission error signal.

The EEMD processing unit includes a noise insertion unit that inserts and outputs random white noise different from the residual transmission error signal; An EMD processing unit that outputs a plurality of residual transmission error mode signals, which are IMF signals for each, by performing EMD on each residual transmission error signal in which white noise is inserted; And an average processing unit that performs ensemble averaging and outputs residual transmission error mode signals corresponding to each white noise.

The apparatus is characterized in that it further comprises an output unit for outputting the defect determination result information determined by the defect determination unit.

The defect determination unit is characterized in that the crest factor is calculated by the following equation (9).

[Equation 9]

 Rotating body defect analysis method using a transmission error according to the present invention for achieving the above object is: a rotary encoder is connected to the gear of the gearbox to rotate signals for two or more of the gears rotating in the gearbox Output gear rotation encoding process; A transmission error measurement process in which a TE measurement unit calculates a transmission error by a difference between a rotation signal output from the rotation encoder and a rotation signal for a normal gear, and outputs a transmission error signal; A residual transmission error acquisition process in which the RTE conversion unit calculates a residual transmission error (RTE) according to the transmission error signal output from the TE measurement unit, and outputs a residual transmission error signal accordingly; After the EEMD processor inserts at least one random white noise that is different from the residual transmission error signal and performs empirical mode decomposition (EMD) processing, the ensemble average of the corresponding eigenmode function (IMF) signals for each white noise An EEMD processing process of taking and outputting a residual transmission error mode signal that is an eigenmode function (IMF) signal; And a defect determination unit calculating a crest factor for a plurality of residual transmission error mode signals input from the EEMD processing unit, and determining whether the gear for the rotation signal is normal or defective based on the calculated crest factor. It is characterized by.

The defect determination process includes: a normal determination step of determining whether the gear is normal by examining whether the crest factor is within a normal crest factor range; A spall defect determination step of determining whether the gear is a spall defect by examining whether the crest factor falls within a spall defect crest factor range; And a crack defect determining step of determining whether the gear is a crack defect by inspecting whether the crest factor falls within the crack defect crest factor range.

The remaining transmission error acquiring process includes: a notch filtering step of notch filtering and outputting a gear engagement frequency signal and its harmonic frequency signal from a transmission error signal output from the TE measurement unit; And a high pass filtering step in which the high pass filter filters low frequency signals below a certain low frequency from the notch filtered transmission error signal to output a residual transmission error signal.

The defect determination process includes: a noise insertion step in which a noise inserting unit inserts different random white noise into the residual transmission error signal and outputs the random white noise; An EMD processing step in which the EMD processing unit EMDs the respective residual transmission error signals in which white noise is inserted to output a plurality of residual transmission error mode signals, which are IMF signals for each; And an average processing step of ensemble averaging the residual transmission error mode signals corresponding to each white noise and outputting the average processing unit.

The method is characterized in that it further comprises an output process of outputting the defect determination result information determined by the defect determination unit as one or more of text, icons, and graphics.

In the defect determination process, the defect determination unit is characterized in that it calculates the crest factor by the following equation (9).

[Equation 9]

.

.

The present invention has the effect of separating the difference in the transmission error signal of the gear according to the type of the gear defect through the ensemble empirical mode decomposition (EEMD) to determine not only the gear defect but also the defect type.

The present invention has the effect that it is possible to know the type of the defect, so it is possible to more accurately determine whether or not the gear has failed, and thus, it is possible to perform an appropriate and quick response according to the gear failure.

1 is a view showing the configuration of a rotor defect analysis apparatus using a transmission error according to the present invention.
2 is a view showing a spall defect and a crack defect of a gear classified according to the present invention.
3 is a view showing a gear contact state according to the present invention.
4 is a view showing the states of the spalls and cracks, transmission errors and residual transmission error waveforms according to the present invention.
5 is a view showing a waveform of a transmission error and a residual transmission error of a spool obtained by filtering according to the present invention.
6 is a view showing a waveform of a transmission error and a residual transmission error of a crack obtained by filtering according to the present invention.
7 is an enlarged view of the transmission error and residual transmission error waveforms of the spalls and cracks according to the present invention.
8 is a waveform diagram showing the IMF signal of the spall transmission error according to the present invention.
9 is a waveform diagram showing an IMF signal of a crack transmission error according to the present invention.
10 is a view showing the waveforms of the residual transmission error of the normal gear and the spall and the cracked gear according to the present invention.
11 is a view showing a crest factor graph of IMF signals according to the present invention.
12 is a flowchart showing a method for analyzing a defect in a rotating body using a transmission error according to the present invention.

Hereinafter, a configuration and operation of a rotating body defect analysis apparatus using a transmission error according to the present invention will be described with reference to the accompanying drawings, and a method for analyzing a rotating body defect in the apparatus will be described.

1 is a view showing the configuration of a rotor defect analysis apparatus using a transmission error according to the present invention, FIG. 2 is a view showing a spall defect and a crack defect of a gear classified according to the present invention, and FIG. FIG. 4 is a view showing a state of spalls and cracks, a transmission error (TE), and a residual transmission error (RTE) waveform according to the present invention, and FIG. 5 is filtered according to the present invention. It is a diagram showing the transmission error and residual transmission error waveforms of the obtained spool, and FIG. 6 is a diagram showing the transmission error and residual transmission error waveforms of cracks obtained by filtering according to the present invention, and FIG. 7 is a spool according to the present invention 8 is a waveform diagram showing an IMF signal of a spall transmission error according to the present invention, and FIG. 9 is a crack transmission error according to the present invention. A waveform diagram showing the IMF signal of Figure 10 is a diagram showing the waveform of the residual transmission error of the normal gear and spall and cracked gear according to the present invention, Figure 11 is a crest factor graph of the IMF signals according to the present invention It is the figure shown. This will be described below with reference to FIGS. 1 to 11.

An apparatus for analyzing a rotating body defect using a transmission error according to the present invention includes a gearbox 10, a rotation encoder 20, a transmission error (TE) measurement unit 30, a residual transmission error (RTE) conversion unit 40, an ensemble. And an empirical mode decomposition (EEMD) processing unit 50, a defect determination unit 60, and an output unit 70.

The gearbox 10 transmits rotational power from a mechanical system, and includes a pair of gears, motors, and powder brakes to analyze defects. In the following description, the spur gear configured in the gearbox 10 has a gear module (Module: length between gears and teeth) of 4 mm, a pitch diameter of 280 mm, 140 mm, and a servo motor having a rated output of 2.9 kW. It will be explained. Preferably, the motor is controlled so that the gear is rotated in units of 0.1 deg.

As shown in (a) of FIG. 3, the gear pair has a force transmitted from 3-1 and a contact portion thereof and a force from 3-1 and 3-2 as shown in FIG. 3 (b). It occurs alternately when there are two parts of the contacted tooth.

The rotation encoder 20 outputs a rotation signal according to the rotation and rotation amount of each gear of the gearbox. The rotation signal will fluctuate by including a fault signal according to a crack such as (b) and a spool as shown in (a) of FIG. 2.

The TE measuring unit 30 calculates a transmission error TE from each rotation signal output from the rotation encoder 20 and a rotation signal for a known normal gear, and FIG. 4 (c) according to the calculated transmission error And (D), (A) of FIG. 5, (A) of FIG. 6, and the like. The transmission error signal includes a repetition of a gear mesh frequency (GMF) signal, a low frequency signal, and a defect signal according to a defect.

When a spall occurs in the gear as shown in (a) of FIG. 2, the peak value increases as shown in (c) of FIG. 4, and the phase of the transmission error signal increases compared to the transmission error signal of the normal gear. You can see it moves.

And it can be seen that the peak value increases and the phase does not move in comparison with the transmission error of the normal gear, as shown in (B) of FIG.

The RTE conversion unit 40 calculates and outputs the residual transmission error using the transmission error signal input from the TE measurement unit 30 and the transmission error signal of a normal gear.

In the present invention, the RTE conversion unit 40 filters the transmission error signal having the spall and crack defects as shown in FIGS. 5 and 6 to subtract the normal transmission error from the transmission error signal output from the TE measurement unit 30 and remains Calculate the transmission error and output the residual transmission error signal as shown in Fig. 4 (E) and Fig. 4 (F).

The residual transmission error signal due to crack defect has a smaller peak value than the residual transmission error signal due to spall defect and has an asymmetric peak.

The RTE conversion unit 40 includes a notch filter 41 and a high-pass filter 42 to notch and high-pass the transmission error signal to generate and output a residual transmission error signal.

Specifically, the notch filter 41 notches the gear meshing frequency (GMF) signal of 35 Hz and its harmonic frequency signal from the transmission error signal input from the TE measuring unit 30 according to the spall defect and crack defect. The filtered transmission error signals as shown in FIGS. 6B and 7B are output.

The high-pass filter 42 removes a low-frequency signal of 10 Hz or less of the notch-filtered transmission error signal, and according to spall defects and crack defects, Figs. 5 (c), 6 (c), 7 (e), and The residual transmission error signal as shown in FIG. 8 (bar) is output.

As shown in Fig. 4 (E), the residual transmission error signal for spall defect is characterized by a change in the magnitude of the peak value, and as shown in Fig. 4 (F), the residual transmission error signal for crack defect is characterized by the sharpness of the asymmetric vertex. Lose.

The EEMD processing unit 50 receives the residual transmission error signal from the RTE conversion unit 40, and changes the peak (amplitude) value of the residual transmission error signal due to spalling defects and changes the sharpness of the residual transmission error signal due to cracking defects. Performs an operation for more clearly distinguishing, and outputs a residual transmission error mode signal, which is an IMF signal as shown in FIGS. 8 and 9 accordingly.

Specifically, the EEMD processing unit 50 includes a noise insertion unit 51, an EMD processing unit 52, and an average processing unit 53.

The noise inserting unit 51 inserts at least one random white noise into the residual transmission error signal input from the RTE conversion unit 40 to connect the disconnection section of the signal to be a continuous residual transmission error signal. Output. This is to prevent the EMD processing unit 52 from failing to decompose the IMF signal with respect to the residual transmission error signal. In other words, when a non-contiguous signal such as an impact signal enters the signal, mode mixing is not distinguished and signal mixing occurs in different IMF signals, so insertion of white noise prevents mode mixing. It is to do.

The EMD processing unit 52 inserts different white noises (i = 1,2,3, ..., n) output from the noise insertion unit 51, and each residual transmission error signal includes various modes of signals. Decomposes for each mode, and outputs a residual transmission error mode signal c _{i, m} .

)를 구성하고 모든 극솟값을 3차 스플라인으로 잇는 하위 포락선(lower envelope:

)를 구성하여 하기 수학식 3과 같이 두 개의 평균을 구한다. Specifically, the EMD processing unit 52 is a technique for finitely decomposing time series data in which multiple modes of nonlinearity are mixed into a vibration function called an intrinsic mode function (IMF) including each mode. The IMF signal is defined as a function that satisfies two conditions. The first condition is that the number of extreme values of the signal and the number of zero-crossings are the same or only one difference exists, and the second condition consists of the maximum value of the function. This means that the average value of the envelope consisting of the envelope and the minimum value should always be zero. The IMF signal is created and separated from the original signal by the following process. When there is an original signal, an upper envelope that connects all the maximum values of the signal to a third-order spline.

) And a lower envelope that connects all the minima to a third-order spline.

) To calculate the two averages as shown in Equation 3 below.

를 하기 수학식 4와 같이 원 신호

에서 빼줌으로써 첫 번째 h(t)를 획득한다. Saved

The original signal as in Equation 4 below

The first h (t) is obtained by subtracting from.

가 IMF 신호의 조건을 만족한다면 첫 번째 IMF 신호가 되는 것이고 그렇지 않다면

가 IMF조건이 만족될 때 까지 시프팅(sifting) 과정을 반복한다. 반복되는 시프팅 과정은 앞선 과정과 같이 IMF 조건을 만족하지 못한

를 원 신호처럼 두고

의 극댓값으로 구성된 upper envelope와 극솟값으로 구성된 lower envelope를 통해 구한

을 빼주는 것을 말한다. 이를 하기 수학식 5와 같이 나타낼 수 있다.if

Is the first IMF signal if it satisfies the condition of the IMF signal.

Repeat the shifting process until the IMF condition is satisfied. The repeated shifting process does not satisfy the IMF conditions as in the previous process.

Leave it like the original signal

Calculated from the upper envelope consisting of the maximum value of and the lower envelope consisting of the minimum value.

Refers to subtracting. This can be expressed as Equation 5 below.

는

로 표시한다. 원 신호

에서 첫 번째 IMF인

를 빼주게 되면 그 결과를 첫 번째 residue라고 하고 하기 수학식 6과 같이

로 나타낼 수 있다.Through the shifting process, IMF conditions were satisfied.

The

It is indicated by. Signal

The first IMF in

If subtracted, the result is called the first residue, and as shown in Equation 6 below.

Can be represented as

에서 다시 시프팅 과정을 거쳐 두 번째 IMF

를 얻게 되고 이후 다시

에서 앞선 과정의 반복을 통해 N번째 IMF까지 얻는다. 시프팅 과정은 마지막 잔여물(residue)인

가 단조함수가 될 때 마치게 된다. 따라서 원 신호

는 다음의 수학식 7로 표현 할 수 있다.

The second IMF after shifting again

And then again

From the previous process, it is obtained up to the Nth IMF. The shifting process is the last residue

Ends when becomes a monotonic function. Hence the original signal

Can be expressed by the following equation (7).

The average processing unit 53 substitutes the corresponding IMF signals for each white noise (i) into Equation 8 below, takes an ensemble average, and outputs a plurality of residual transmission error mode signals according to spall defects and crack defects.

For example, when the number of different white noises inserted in the residual transmission error signal is 5 and 10 (i = 1,2, ..., 10) IMF signals are output for each residual transmission error signal, the first white Ensemble for the first IMF signal of noise, the first IMF signal of second white noise, the first IMF signal of third white noise, the first IMF signal of fourth white noise, and the first IMF signals of fifth white noise Take the average and output the first residual transmission error mode signal (c ₁ ), the second IMF signal of the first white noise, the second IMF signal of the second white noise, the second IMF signal of the third white noise, the fourth for the second signal and the five IMF IMF second signal of the second white noise of a white noise by taking the ensemble average and outputting a second residual error transmission mode signal (c ₂₎ By taking the ensemble average for the IMF signal corresponding to white noise by all and outputs a residual error transfer mode signal (c _i).

8 and 9, a large amplitude characteristic of the transmission error signal due to spalling defects has a great influence on the residual transmission error mode signal in a relatively low frequency region, while the peak of the transmission error signal due to crack defects is relatively As can be seen, it affects the residual transmission error mode signal in the high frequency region.

The defect determination unit 60 obtains the absolute value of the crest factor to the residual transmission error mode signal, which is the IMF signal output from the EEMD processing unit 50, and determines whether the corresponding gear is normal or spall according to the range in which the absolute value is formed. It is determined whether it is a defective gear or a gear having a crack defect, and as a result, the defect determination result information is output.

 Specifically, the defect determination unit 60 calculates the crest factor according to Equation 9 below, which is obtained through the absolute value of the largest peak relative to the RMS value of the discrete data, and determines whether the calculated crest factor falls within a preset normal crest factor range. Defective crest factor range or crack defects Determines whether the gear crest factor falls within the crest factor range.

The normal crest factor range is 3 <c1 <5 as shown in FIG. 11, can be defined between 3 <c3 <4, and the spall defect crest factor range is 4 <c1 <7 and can be defined as 4.5 <c3 <5.5, and cracks. The defect crest factor range is 6 <c1 <8 and could be defined as 3 <c3 <4.

 The output unit 70 displays the defect determination result information output from the defect determination unit 60 in text, icon, graphic, or the like, or outputs it in voice.

12 is a flowchart showing a method for analyzing a defect in a rotating body using a transmission error according to the present invention.

Referring to FIG. 12, the rotation encoder 20 performs encoding according to rotation of the gears of the gearbox 10 and outputs rotation signals according to rotation of at least one gear of the gearbox 10 (S111).

The TE measuring unit 30 receives the rotation signal output from the rotation encoder 20, compares the rotation angle (amount) of the normal gear with the rotation angle (amount) of the gear according to the actual rotation signal, and calculates a transmission error. The transmission error (TE) signal is output (S113).

The RTE converter 40 performs notch filtering to filter the gear meshing frequency signal and the harmonic signal of the gear meshing frequency signal included in the transmission error signal output from the TE measuring unit 30 (S117), and notch filtered High-pass filtering is performed to filter low-frequency signals of 10 Hz or less from the residual transmission error signal (S119), and a filtering process is performed (S115).

When the residual transmission error signal by filtering is output, the EEMD processing unit 50 inserts white noise into the residual transmission error signal input from the RTE conversion unit 40, and then performs the above-described EMD to include the residual transmission error signal. A plurality of residual transmission error mode signals, which are IMF signals according to modes, are generated, and EEMD is performed by averaging and outputting the generated residual transmission error mode signals (S121).

The defect determination unit 60 receives an average processed residual transmission error mode signal from the EEMD processing unit 50, calculates a crest factor for each residual transmission error mode signal, and calculates a crest factor range in which the calculated crest factor is formed and a preset normal gear. Whether it is included in the normal crest factor range, the spall crest factor range, or the crack crest factor range is determined, and one of gear normal, spall defect, and crack defect is determined (S123). The above ranges may be defined by experiments.

On the other hand, the present invention is not limited to the typical preferred embodiments described above, but can be carried out by improving, changing, replacing or adding in various ways without departing from the gist of the present invention. Anyone who has a will easily understand. If the implementation by such improvement, modification, replacement or addition falls within the scope of the appended claims, the technical idea should also be regarded as belonging to the present invention.

10: gearbox 20: rotary encoder
30: TE measurement unit 40: RTE conversion unit
41: notch filter 42: high pass filter
50: EEMD processing unit 51: noise insertion unit
52: EMD processing unit 53: average processing unit
60: defect determination unit 70: output unit

Claims (12)

A rotary encoder connected to a gear of a gearbox and outputting a rotation signal for at least one of gears rotating in the gearbox;
A TE measuring unit for calculating a transmission error and outputting a transmission error signal based on a rotation signal output from the rotation encoder and a rotation signal for a preset normal gear;
An RTE conversion unit for calculating a residual transmission error (RTE) using the transmission error signal output from the TE measurement unit and a transmission error signal of a normal gear, and outputting the residual transmission error signal accordingly;
After inserting at least one random white noise that is different from the residual transmission error signal and performing empirical mode decomposition (EMD) processing, an ensemble average is obtained for the corresponding eigenmode function (IMF) signals for each white noise. An EEMD processing unit for outputting a residual transmission error mode signal that is a mode function (IMF) signal; And
Calculating the crest factor for a plurality of residual transmission error mode signals input from the EEMD processing unit, and determining whether the gear outputting the rotation signal is normal or defective based on the calculated crest factor, but the plurality of residual transmission error modes And a defect determination unit for determining whether the gear is normal, spall defective, or crack defective by examining whether the crest factor of each signal belongs to the normal crest factor range, the spall defect crest factor range, or the crack defect crest factor range. A device for analyzing a defect of a rotating body using a transmission error.
delete According to claim 1,
The RTE conversion unit,
A notch filter configured to notch-filter a gear engagement frequency signal and a harmonic frequency signal from the transmission error signal output from the TE measurement unit; And
And a high pass filter for filtering a low frequency signal below a certain low frequency from the notch filtered transmission error signal and outputting a residual transmission error signal.
According to claim 1,
The EEMD processing unit,
A noise inserting unit inserting different random white noise into the residual transmission error signal and outputting the random white noise;
An EMD processing unit that outputs a plurality of residual transmission error mode signals, which are IMF signals for each, by performing EMD on each residual transmission error signal in which white noise is inserted; And
And an average processing unit which ensemble-averages and outputs residual transmission error mode signals corresponding to each white noise.
According to claim 1,
And an output unit configured to output information on defect determination results determined by the defect determination unit.
According to claim 1,
The defect determination unit,
A rotating body defect analysis apparatus using a transmission error, characterized in that the crest factor is calculated by the following equation (9).
[Equation 9]

A gear rotation encoding process in which a rotation encoder is connected to a gear of a gearbox and outputs rotation signals for at least one of the gears rotating within the gearbox;
A transmission error measurement process in which a TE measurement unit calculates a transmission error based on a rotation signal output from the rotation encoder and a rotation signal for a normal gear, and outputs a transmission error signal;
A residual transmission error acquisition process in which the RTE conversion unit calculates a residual transmission error (RTE) according to the transmission error signal output from the TE measurement unit, and outputs a residual transmission error signal accordingly;
After the EEMD processing unit inserts at least one random white noise that is different from the residual transmission error signal and performs empirical mode decomposition (EMD) processing, ensemble averaging for the corresponding eigenmode function (IMF) signals for each white noise An EEMD processing process of taking and outputting a residual transmission error mode signal that is an eigenmode function (IMF) signal; And
The defect determination unit includes a defect determination process for calculating a crest factor for a plurality of residual transmission error mode signals input from the EEMD processing unit, and determining whether the gear for the rotation signal is normal or defective based on the calculated crest factor. ,
The defect determination process,
A normal determination step of determining whether the gear is normal by examining whether the crest factor is within a normal crest factor range;
A spall defect determination step of determining whether the gear is a spall defect by examining whether the crest factor falls within a spall defect crest factor range; And
And a crack defect determining step of determining whether the gear is a crack defect by inspecting whether the crest factor falls within a crack defect crest factor range.
delete The method of claim 7,
The remaining transmission error acquisition process,
A notch filtering step in which the notch filter notches and filters the gear engagement frequency signal and its harmonic frequency signal from the transmission error signal output from the TE measurement unit; And
And a high-pass filtering step in which the high-pass filter filters a low-frequency signal having a predetermined low frequency or less from the notch-filtered transmission error signal to output a residual transmission error signal.
The method of claim 7,
The defect determination process,
A noise inserting step in which a noise inserting unit inserts different random white noise into the residual transmission error signal and outputs it;
An EMD processing step in which the EMD processing unit EMDs each of the residual transmission error signals in which white noise is inserted to output a plurality of residual transmission error mode signals that are IMF signals for each; And
The average processing unit includes an averaging processing step of ensemble averaging the residual transmission error mode signals corresponding to each white noise, and outputting the rotor defect analysis method using the transmission error.
The method of claim 7,
And an output process of outputting the defect determination result information determined by the defect determination unit as one or more of text, icons, and graphics.
The method of claim 7,
In the defect determination process,
The defect determining unit calculates the crest factor by the following equation (9) Rotating defect analysis method using a transmission error.
[Equation 9]

Images