JPS6066647A - Diagnosing method of motor - Google Patents

Abstract

PURPOSE:To accurately diagnose the propriety of a winding by obtaining the maximum amplitude value and the average value of ordinary vibration component from the vibration waveform at the starting time. CONSTITUTION:A piezoelectric acceleration sensor 2 is mounted in a casing of a 3-phase induction motor 1 to detect the vibration at the starting time of the motor 1. The detection signal of this vibration waveform is inputted through a signal processor 3 to a calculator 4. This calculator 4 reads out the input signal at the prescribed sampling pitch to sample the maximum amplitude value as a peak value for the prescribed period. The maximum peak value is determined from the peak values, and the average value of the peak values in the ordinary vibration period is calculated. Further, the sum of the products of the respective peak values and the time is obtained in the entire range of the sample collecting time. The normal or abnormal state of the motor 1 is discriminated by the ratio of the maximum peak value to the average value and the sum in the entire range.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for determining breakability of a motor, and for example, to a method for diagnosing the quality or looseness of the insulation of motor windings based on the vibration waveform that occurs when the motor is started.

If of the motor winding! + Edge defects are directly related to the lifespan of the electric motor, so it is necessary to have it checked periodically.

One method for diagnosing the insulation of electric motors was to record the vibration waveform of a single weight using an electromagnetic oscilloscope, etc. when it was started, and based on the recorded results, the inspector determined whether the insulation of the windings was defective. . In other words, when the insulation of the windings deteriorates and the insulators wither (volume reduction), the windings become loose, and the amplitude of the motor's vibration waveform recorded by the electromagnetic oscilloscope becomes the normal amplitude. appears significantly larger than the For example, the first
Compared to the normal vibration waveform shown in Figure (A), the vibration waveform shown in Figure 1 (B) when insulation is defective or loosens (hereinafter referred to as abnormality) is during the transient vibration period Since the amplitude appears extremely large both during the period from 1 to 3 until the vibration reaches a steady state) and during the steady vibration period Y (the steady state period that appears after the transient vibration period), it is possible to determine whether the motor is normal or abnormal.

However, when using this method, it is difficult to quantify the criteria for determining whether lightning is true or unusual, and because it relies on the experience and intuition of the inspector, there is considerable variation among individual inspectors, making it difficult to make an accurate diagnosis. 9°I [! :
,i,kaa)go.

The present invention has been made in view of such circumstances, and is
From the vibration waveform at startup of iIJla, the maximum amplitude value in the transient depth period, the average value of the amplitude in the steady vibration period, and the area of the part containing the damping component (both small) can be used to determine the quality or looseness of the insulation of the motor winding. It is an object of the present invention to provide a method for diagnosing an electric motor that can accurately distinguish 111.

A method for diagnosing a motor according to the present invention is a method for detecting vibration that occurs when the motor is started, and diagnosing loosening of the motor windings based on the detection result. 1ζThe vibration is detected, and based on the waveform of the detected vibration, the maximum amplitude value of the vibration waveform that falls on the transient vibrator from the time of startup until the vibration reaches a steady state, and the steady state vibration when the vibration reaches a steady state. The present invention is characterized in that the average value of the amplitude of the vibration waveform over a period of time and the area of a portion of the vibration waveform that includes at least a damping component are determined, and the looseness of the winding is determined based on these.

Hereinafter, the present invention will be explained in detail based on drawings showing embodiments thereof.

FIG. 2 is a zonal diagram showing the implementation state of the method of the present invention, FIG. 3 is a signal waveform diagram for explaining its operation, and FIG.

A piezoelectric acceleration sensor 2 is attached to the casing of the three-phase induction motor 1. The acceleration sensor 2 is for detecting vibrations when the electric motor 1 is started up, and manually inputs a detection signal having a waveform as shown in FIG. 3(a) to the signal processing circuit 3. The signal processing circuit 3 amplifies and rectifies this input signal.
The arithmetic unit 4 outputs a waveform output signal as shown in the figure (b).
Enter.

The arithmetic unit 4 reads this input signal with a certain sampling pinch, and reads the input signal at a certain period L (or a certain number of read data).
The maximum amplitude value for the peak value a representing that period
Sampling processing is performed with k (k=1.2...n).

If the driving frequency of the motor 1 is (iQIIz), L is 1 cycle of IG, 67m5, and the sampling time T is 2
It is best to set the time to ~4 seconds. FIG. 3(c) conceptually shows the data obtained in this way. This will be explained below with reference to this graph.

Next, the /iii calculating device 4 determines the largest peak value l from among the peak values ak (corresponding to a3 in FIG. 3(c)).

Next, the calculation shown by the following +11 formula is executed to calculate, for example, 20 peak values (al'1
-IQ l an-IRl ”' an-11all
), that is, the average value A2 of the peak values in the steady vibration period Y shown by the two-dot chain line in FIG. 3(c) is calculated.

2υ And the ratio between the maximum peak value A1 and the average value A2, A
Turn on I/A2.

Next, the sampling time of the product of each peak value ak and time [
Sum 5a over the entire range of r = (al → −a2 → −...+
an) Add L. Here, it corresponds to the area of all the components within the sampling time of 1'' of the detected vibration, and physically it is the time integral value of the energy of all the components. Also, this sum S
In order to reduce production by the product of the average value A2 and time T shown in formula (1) from a, the area of the region where the peak value shown by diagonal lines in FIG. 3 (c) is larger than the average value A2 Sd is calculated using the following equation (2).

Sd = (al +22+-+an) L-A2
T-f21 This area Sd corresponds to the area of the damping component of the detected vibration, and is physically a time integral value of the energy of the damping component. Since the information contents of Sa and Sd are almost equivalent, they can be used selectively depending on the vibration waveform situation. For example, when the peak/tot/peak value of a transient oscillator is large, Sd is close to zero. In such cases, it is better to use Sa.

Next, the arithmetic unit 4 records the maximum peak value AI, the ratio AI/A2 therebetween, the area Sa of all components, or the area Sd of the attenuation component on the printer 5.

Next, the effects of the method of the present invention will be explained based on Examples.

Ten electric motors were used for testing, and their specifications are shown in Table 1. In addition, in order to investigate the effect on the vibration waveform at startup when there is looseness in the winding, No. 1, No. 3
The stator windings of these motors were not impregnated with varnish at the time of manufacture.

Similarly to the above-mentioned motors, motor No. 2 and No. 4 do not have their stator windings impregnated with varnish, and furthermore, one pole of the U phase is short-circuited to -V, and motor No. 6 is short-circuited. Therefore, there is no way to increase the amount of vibration generated.

Here, the ratio /'+/Az between the maximum peak value A1 and the average value A2 and the total component area Sa are used as indicators for determining whether the electric motor is normal or abnormal.

Table 2 shows the ratio AI/of each motor detected by the method of the present invention.
A2 is a table showing the total component area Sa and sampling time T, and FIG. 5 shows the ratio AI/A2 on the vertical axis and SaX on the horizontal axis.
100/T, and is a graph plotted based on the experimental results shown in Table 2 for each electric motor.

The degree of loosening simulated here according to Table 2 and Figure 5,
That is, the fact that varnish is not impregnated is difficult to detect based on the ratio AI/A2, and it is difficult to detect the fact that it is not impregnated with varnish. Changes in A2°A r /A2 over time, etc. are also required.

On the other hand, it can be observed that the ratio A + /A 2 of the electric motor (positive 2, 4.6) in which a layer short occurred in the stator winding is small, and the area Sa is large.

From this, it can be considered that detection of the area Sa alone can serve as an index for determining field lightning. In other words, when a layer short occurs, an abnormality can be easily determined based only on the area Sa, which vibration is obtained.

Classifying the above detection results in terms of observed waveforms, the 1st, 6th
As shown in Figure (A), the average value A2 during the lightning oscillation period is very small compared to the maximum peak value A1 during the hyper-i oscillation period, that is, the value of AI/A2 is large, and the area Sa is small. thing·
...Electric FI & 1.

3.5 2. As shown in Figure 6 (b), the value of 1 to the maximum peak value is small and approximately equal to the average value A2, the ratio AI/A2 is small, and the area Sa is small...Electric motor floor 7 ,8,9.1
0 3. As shown in Figure 6 (c), the maximum peak value A1 and the average value A2 are both large and approximately equal, 1LA+/Az
is small and the area Sa is large...Electric motor positive 2. 4. 6 Based on the above considerations, when the vibration waveform shown in FIG. 6(c) appears, it can be determined that an abnormality in the motor, that is, loosening of the windings or poor insulation has occurred.

Also, from Figure 5, as the plot moves from top to bottom or left to right, the ratio A 1 / A 2 becomes smaller,
Furthermore, it can be determined that the abnormality is progressing as the area Sa increases, and this should be taken into consideration in diagnosis by monitoring changes over time.

In the case of the method of the present invention as described above, the maximum peak value A1
, the average value A2 and the area Sa or Sd can be quantified, so
The normal and abnormal enrichment regions shown in Fig. 5, which are obtained through experiments and actual results, are determined in advance as judgment criteria, and the actual measurement results of the motor to be judged as normal and abnormal are plotted, or the normal and abnormal rich regions are plotted or Quantify the values in the same way,
By directly comparing the measured value and the normal value, it is possible to accurately determine whether the motor is normal or abnormal.

In addition, the above-mentioned maximum peak value A++ + average value A2, area Sa
Or, in addition to Sd, the average value P of the sum of squares of each peak value ak, which is an index of the overall magnitude of vibration at the time of startup, as shown in the following formula (3), is also used as a criterion for determining whether it is normal or abnormal. Good too.

This average value of the sum of squares P also increases as the insulation deteriorates.

Furthermore, in the above-described embodiments, the method of the present invention is applied to a three-phase induction motor, but it goes without saying that the present invention can also be applied to other AC motors such as a single-phase induction motor.

As detailed above, the electric motor diagnosis method according to the present invention analyzes the vibration of the electric motor at startup, and estimates the maximum amplitude value, the average value of the amplitude of the lightning vibration component, the area of all components, or the area of the damping component. Therefore, the quality of the insulation of the motor windings,
Furthermore, the present invention has excellent effects, such as being able to determine whether or not there is any loosening due to manufacturing reasons, making accurate diagnosis possible.

[Brief explanation of the drawing]

Fig. 1 is a signal waveform diagram for explaining the conventional method, Fig. 2 is a block diagram showing the implementation state of the method of the present invention, Fig. 3 is an explanatory diagram for explaining its operation, and Fig. 4 is a flowchart 1. -1
FIG. 5 is a graph showing the experimental results of the electric motor under test, and FIG. 6 is a waveform chart showing the types of vibration waveforms of the electric motor. ]...Three-phase electric hammer 2...Acceleration sensor 3...
・Signal processing circuit 4...6AW, device characteristics Applicant Sumitomo Metal Industries Co., Ltd. Agent Patent attorney Noboru Kono
Figure 2 Figure 2 C A) (B) Figure 1 Figure 2 Time A number 3 [1 Figure 4 Figure 5 (C) Figure 6

Claims (1)

[Claims] (2) In a method of detecting the vibration that occurs when the motor starts up and diagnosing the loosening of the motor winding based on the detection result, the vibration is detected with a vibration detector attached to the motor, and the waveform of the detected vibration is detected. Based on the maximum amplitude value of the vibration waveform during the transient vibration period from startup until the vibration reaches a steady state
Calculate the average value of the amplitude of the vibration waveform during the steady vibration period when the vibration has reached a steady state, and the area of the part of the vibration waveform that includes at least the reduced* component, and use these to reduce the loosening of the winding.
A method for diagnosing an electric motor, characterized by classifying P1.