JPS62127678A - Method and device for detecting rectification abnormality in DC machine - Google Patents

Abstract

PURPOSE:To detect abnormality of a commutator such as film breakage and local abrasion easily with high precision during operation by detecting vibrations of a DC machine and analyzing a segment ripple and/or pole ripple. CONSTITUTION:A vibration sensor 1 detects vibrations of the brush of the DC machine from the brush or a brush holder, etc., a constant setting circuit 4 sets the constant of a band-pass filter corresponding to the segment ripple or pole ripple, and the band-pass filter 3 extracts a signal of the frequency band of the segment ripple or pole ripple component of the DC machine to be measured from the output of an amplifier 2. An autocorrelation arithmetic and addition circuit 5 varies the function Rxx(tau) of a time difference tau and the tau to find plural maximum values Rxx(taui) of an autocorrelation function, and calculates their sum so as to grasp features of time variation of the output signal x(t) of the filter 3. Then, it is decided from a level indicated on a display device 6 whether there is abnormality or not.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a method and device for detecting rectification abnormalities in DC machines such as DC motors and DC generators, and relates to a method and device for detecting rectification abnormalities in a DC machine such as a DC motor or a DC generator. The purpose is to detect during operation.

<Prior Art> Direct current machines are widely used as variable speed, high precision drive sources and power generation sources J=L in large rolling mills and other types of mechanical equipment.

However, if a failure occurs in this DC machine, the production line will be stopped for a long time.

There is a risk of significant damage. For this reason.

Is Koishi et al.'s flow machine malfunctioning? Timely and reliable detection is required.

One of the failures of these DC machines is a commutator burnout accident called flashover.

This failure is often caused by an increase in commutator sparks due to increased commutator coating damage or uneven wear. Therefore, in the past, damage to the coating has been dealt with through regular visual inspections. On the other hand, uneven wear can be checked by disassembling and inspecting the commutator and measuring the diameter of the commutator at several locations using calipers.

However, disassembling a DC machine is a very complicated task and costs a lot of money, so inspections are often carried out over long periods or omitted, especially when it comes to the degree of uneven wear on the commutator. Early detection was difficult.

In addition, the inspection required skill and had problems with accuracy. Furthermore, there are other problems such as the need to stop the operation of the production line in order to conduct the inspection.

<Structure of the Invention> The present invention has been made in order to solve the problems of the prior art described above, and is capable of easily and highly accurately detecting abnormalities such as coating roughness and uneven wear of a commutator of a DC machine during operation. It is an object of the present invention to provide a method and apparatus.

In the present invention, vibrations of a DC machine in operation are detected.

The vibrations of this DC machine include a mixture of vibrations caused by the segments of the commutator (so-called segment) IJ tuples and vibrations caused by the poles (so-called pole ripples). As a result of various experimental studies, the present inventors have come to the knowledge that segment ripple has a strong correlation with abnormal roughness of the commutator, and pole ripple has a strong correlation with abnormal wear of the commutator. .

Therefore, in the present invention, it goes without saying that the component of the segment ripple or the pole ripple is extracted from the vibration of the DC machine; both the segment ripple and the pole ripple may be extracted), and this component is analyzed.

This is the first feature. Since each of these ripples has a unique frequency, it can be easily extracted using an appropriate bandpass filter or the like. Also, to detect vibrations, it is best to detect them from the brushes of the DC machine.

In order to further analyze segment ripples and/or pole ripples, in the present invention, ■Autocorrelation function curve■Autocorrelation function RXX (τ)=, 11-: T J'
, "x (t) This is used to diagnose rectification abnormalities. This is the second feature of the present invention, and through numerous experimental studies, it has been found that by using these, abnormalities can be detected with extremely high accuracy.

<Example> The present invention will be described in detail below based on the drawings.

FIG. 1 is a block diagram showing the configuration of the present invention. In the same figure, (1) is the brush or brush holder that controls the vibration of the brush in the DC machine.

This is a vibration sensor that detects from the brush holder mounting rod, etc. In this example, an acceleration type vibration sensor is used.

This vibration sensor (1) is.

It is connected to a bandpass filter (3) via an amplifier (2), and among the outputs of the amplifier (2), the signal in the frequency band of the segment ripple and pole ripple components of the DC machine under test is It is designed to be taken out. A constant setting circuit M (4) is connected to the band-pass filter (3), which corresponds to the segment ripple and pole ripple, respectively, based on the number of segments, the number of balls, and the rotation speed during measurement of the DC machine to be measured. The constant of the bandpass filter (3) can be set. If the purpose is to measure the roughness of the commutator, the constant of the bandpass filter is segment ripple = rotational speed (rpm) of the DC machine / 6ox x number of segments.

In addition, if the purpose is to measure the degree of uneven wear of the commutator, use a bandpass filter: ball lip = rotational speed of the DC machine (rpm)/6
Set to 0X pole number.

Next, the output of the band-pass filter (3) is connected to an autocorrelation function calculation and addition circuit (5), which captures the temporal fluctuation characteristics of the output signal x(t) of the filter (3). Therefore, the function RXX (τ) of the time interval difference τ
-Year E: , fo”x(t)x (t+τ)at and the time interval difference τ is varied to obtain a plurality of local maximum values RXX (τi) of the autocorrelation function and the sum value ΣRX
Calculate X (τI). The output of this autocorrelation function calculation and addition circuit (5) is connected to a display (6).

If the level indicated on this indicator (6) is large, if the bandpass filter (3) is set to segment ripple, the commutator roughness is abnormal, and it is also set to pole ripple. If so, it can be diagnosed as uneven wear of the commutator.

The functions of the autocorrelation function calculation and addition circuit (5) described above will be further explained. The signal x(t) in Figure 2
The measurement data of the autocorrelation property of is shown. Here, the autocorrelation function RXX (τ) is 0.8, and when the time interval difference τ is further changed, the thickest values of the autocorrelation function RXX (τ1) are 0.7.0.3.0.2, respectively. .0.2. From these, the autocorrelation function R when changing the time interval difference τ
XX (addition value of maximum value of τ0 is ΣRXX(τi) = 0
．． 8-) -o, '7+o, s1+1 Obtain as 10.3+0.2+0.2=2.7. The presence or absence of an abnormality in the commutator of the DC machine to be measured can be determined using the autocorrelation function RX.
It can be determined by X(τ), but the additional value X: RXX(
According to τ1), the difference between normal and abnormal times can be determined more clearly.

FIG. 3 shows an example in which vibration of a brush in a DC machine was detected by a vibration sensor (1) attached to the brush and measured using the measurement method described above. The constant setting of the Bandhus filter (3) is based on the segment ripple [rotation speed (rpm) of the DC machine].
/6ox segment number] was measured as ±10%.

In the figure, (A) is a measurement example of a commutator manufactured by Keio Kosei Co., Ltd., and (B)
is an example of measurement of a commutator with a small film roughness, and the film roughness can be determined from multiple figures of the autocorrelation function curve, the autocorrelation function RXX (τ), and the maximum value RXX (τi) of the autocorrelation function described above. The distinction between magnitude is clear in the autocorrelation function curve.

Similarly, the difference in the autocorrelation function RXX (τ) is 0.
Although it is clear that 4510.06 = 7.5 times, the extremely thick value RXX (τI) of the above autocorrelation function is the sum of multiple values ΣRXX
It can be seen that (τi) is clearer with a difference of 4.870.1=43 times.

Figure 4 shows the constant settings of the bandpass filter within the range of pole ripple [rotation speed (rpm) of DC machine/60 x number of poles].
This is an example in which measurement was performed in the same manner as in FIG. 3 with the setting at 10%. In the figure, (A) is a measurement example of a commutator with a large degree of uneven wear, and (B) is a measurement example of a commutator with a small degree of uneven wear. The degree of uneven wear of multiple commutators is clearly distinguished by the autocorrelation function curve. Similarly, the local maximum value RXX (τI) of the autocorrelation function
It can be seen that the multiple summed values XRXX (τ0) can be clearly determined. Therefore, the autocorrelation function curve, the autocorrelation function, and the maximum value RXX (τ1) of the autocorrelation function indicated on the display (6) It is possible to understand the degree of coating roughness and uneven wear of the commutator in a DC machine.

As described above, according to the present invention, the degree of film roughness and uneven wear of a commutator in a DC machine can be determined.

Maximum value RXX (τ1) of the autocorrelation function obtained by changing the autocorrelation function curve, autocorrelation function, and time deviation τ shown on the DC machine drive medium indicator (6), and the sum value ΣR of multiple values
Diagnosis can be made based on the level of XX (τI). Therefore, for example, an abnormality in the DC machine can be detected before the film roughness, uneven wear, etc. of the commutator of a DC machine increases, and the accompanying increase in sparks causes a flashover and the drive stops. By quickly maintaining the DC machine using this method, it is possible to prevent the production line from being shut down for a long period of time. In addition, by using one device, it is possible to diagnose abnormalities regarding commutator skin damage, uneven wear, etc. without the need for conventional special techniques and experience, making it extremely easy to use. can.

Note that the present invention is not limited to the above embodiments.

For example, in the above embodiment, as a means for diagnosing the degree of film roughness and degree of uneven wear of a commutator, an autocorrelation function curve, an autocorrelation function RXX(τ), and a local maximum value RXX(τ
i) Three means for indicating the i number of added values ΣRXX (τi) have been shown, but only one of them may be used.

In addition, as a means of diagnosing the degree of coating roughness and degree of uneven wear of the commutator, a band-pass filter (3
) is rectified by the rectifier circuit aη, and this value is displayed on the display (
b) Diagnosis may be made by displaying the trowel. In addition, the output of the autocorrelation function calculation and addition circuit (5) is processed by a detection circuit or rectification circuit e), the output is integrated by an integrator circuit, and the output of the integration circuit 8 (a) is The diagnosis may be made by displaying the information on the display. In addition, the power spectrum of the output of the band pass filter (3) or the output of the autocorrelation function calculation and addition circuit ¥6 (5) is calculated by the power spectrum calculation circuit M (3).
x) and displaying the output on the display (32) for diagnosis.

In addition, various modifications can be made without departing from the gist of the present invention.

<Effects of the Invention> As detailed above, according to the present invention, the degree of coating roughness, uneven wear, etc. of a commutator in a DC machine can be easily detected without stopping the function of the DC machine. Therefore, for example, the degree of coating roughness and uneven wear of the commutator increases, and the resulting increase in sparks causes a flashover, and the abnormality can be detected before the drive is stopped. This has the effect of allowing machine failure to be detected early, easily, and with high precision.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention, FIGS. 2, 3, and 4 are explanatory diagrams of the effects, and FIG. 5 is a block diagram showing another embodiment of the present invention. It is a diagram. (1)... Vibration sensor → J', (2)... Increaser] device,
(3)...Band pass filter, (4)...Constant setter, (5)...Autocorrelation function calculation and addition circuit, (
6)...Indicator. Patent applicant Nippon Kokan Co., Ltd. Inventor Hiroshi Okitsu Joined the company
Nakamoto-oriented Yakushiji Kyokudo Seki 1) Naruo Patent Attorney Attorney Masao Yoshihara Sando Takahashi Seido Attorney Hiroko Yoshihara Procedural Amendment (l-
) July 77, 1939 [(Mr. Michibe Uga, Commissioner of the Patent Office (Patent Office Examiner Exhibition 9)
1. Indication of the incident 1985 Patent Application No. 267≠2Bu No. 2
, Title of the invention: Direct current machine, General, Article Cr) Tumor return ρ'chi's ridge compensation 3, Person making the amendment Relationship with the planning case Applicant (412) Japan f, ] Ceremony company, agent 5, date of amendment order -+can-now--→←-yun←6, subject of amendment e2'...i Amendment contents 1 Specification 1, page 5, line 6 Correct the text ``from the brush'' to ``from the brush, brush holder, etc.''. In the 11th line of page 5 of the same book, ``Head Office'' is replaced with ``Lim.''
Correct 8 T-+o. Lim 3, Dochi, page 7, line 9, replace “ ” with “T-+
(G) Correct it as tim Jl. T-1ψ ■ Correct the phrase "autocorrelation function" in the third line of page 9 of Shiro to "autocorrelation function." From the end of line 4 to the beginning of line 5 on page 9 of the same book on the left, the text ``From the box,'' has been corrected to ``From Figure 3.'' In the 5th line of page 10 of the same book, the phrase ``additional value'' is replaced by ``additional value.''
I am corrected.

Claims (1)

[Claims] 1. The present invention is characterized by detecting vibrations of a DC machine during operation, extracting components related to segment ripples and/or pole ripples from this detection signal, and detecting rectification abnormalities based on the components. How to detect rectification abnormality in a DC machine. 2. Detect the vibration of the DC machine during operation, extract components related to segment ripples and/or pole ripples from this vibration, and change the autocorrelation function curve, autocorrelation function R_X_X(τ), and time interval difference τ of this signal. The sum value Σ^ of the local maximum value R_X_X(τi) of the autocorrelation function obtained by
A method for detecting rectification abnormality in a DC machine, characterized in that rectification abnormality is detected by determining at least one of n_i_=_1R_X_X(τi). 3. A sensor that detects vibrations of an operating DC machine, a filter that extracts a signal in a predetermined frequency band from a signal from the sensor, and an autocorrelation function calculation device that calculates an autocorrelation function of the signal that has passed through the filter. A rectification abnormality detection device for a DC machine, comprising: