JPS61151478A - Apparatus for diagnosis of electromoter - Google Patents

Abstract

PURPOSE:To accurately perform diagnoisis, by performing quantitative comparison by displaying the componet ratio of the primary current of an electromotor in frequency and power source frequency changed by normality and abnormality of the rotor coil of a three-phase electromotor. CONSTITUTION:The primary current of a three-phase induction electromotor 1 is detected by current detector 2 and the rotation of the three-phase induction electromotor 9 is detected by a rotation detector 3 while the obtained signal respectively receives A/D conversion by A/D converters 4, 5. A central processing unit 7 collects wave form data outputted from the A/D converters 4, 5 to store the same in data memory 8 and performs various operations to calculate data for performing the diagnosis of the rotor coil of the three-phase induction electromotor 1 and displays the same on a display device to perform diagnosis.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application This invention is applicable to the glabellar short circuit (
The present invention relates to an electric motor diagnostic device that can diagnose abnormalities such as layer shorts, wire breaks, and rotor bar wire breaks.

2. Description of the Related Art If a rotor coil or rotor bar of a three-phase induction motor breaks or a layer short occurs, smooth rotation will not be achieved and a serious failure will occur, so it is necessary to detect abnormalities in the rotor at an early stage.

For this reason, conventionally, the current waveform of one phase of the rotating rotor coil was observed using an oscilloscope, and based on the observation results, an inspector determined whether or not there was an abnormality in the rotor. That is, when an abnormality occurs in the rotor, the detected waveform is different from the normal one, which is used to determine whether the three-phase induction motor is abnormal.

However, in such a diagnosis based on waveform observation, it is difficult to quantify the criteria for determining the presence or absence of an abnormality, and there is a problem that accurate diagnosis cannot be made.

The motor diagnostic device that forms the basis of this invention is capable of quantitatively determining abnormalities in rotor coils, etc., and detects the primary current of a three-phase induction motor with a current detector (current transformer). The rotation of the phase induction motor is detected by a rotation detector (pulse generator), and the outputs of the current detector and rotation detector are detected by the first and second A/D converters based on the sampling pulse from the fixed frequency oscillator. A/D respectively
The central processing unit collects the waveform data of the primary current output from the first A/D converter by the number of cycles (for example, about several tens of cycles) required for Fourier transform (frequency analysis). In addition to storing data in memory,
The rotation detection data output from the second A/D converter is collected for the necessary cycles and stored in the data memory, and the waveform data for several tens of cycles of the power supply frequency stored in the data memory is read out. Frequency analysis of the primary current is performed by Fourier transforming the data as one cycle, and the rotation detection data stored in the data memory is read out and the rotation speed of the three-phase induction motor is determined from this rotation detection data. The configuration is such that the beat frequency is determined by calculating the slip from a number, and the ratio between the beat frequency component and the power supply frequency component in the frequency analysis result is determined and displayed. Then, an abnormality in the rotor coil is diagnosed based on the magnitude of the above ratio. Note that the beat frequency fB is determined by the power supply frequency fo and the slip S, as f = fax (1-'2XS)...
...This is the frequency found in [1].

Problems to be Solved by the Invention The motor diagnostic device described above performs discrete Fourier transform, and if the power supply frequency and beat frequency are not synchronized with the sampling pulse of the fixed frequency oscillator, the maximum value of each component will be The image frequency is generally not synchronized with the sampling pulse. In other words, only the discrete Fourier transform can provide that value.

This invention provides an electric motor that can accurately determine the power supply frequency, beat frequency, and the values of those components, and accurately determine the ratio of the power supply frequency component to the power supply frequency component of the power spectrum, thereby increasing the accuracy of abnormality diagnosis. The purpose is to provide diagnostic equipment.

Means for Solving the Problems As shown in FIG.
Current detector (2) that detects the primary current of phase electric @ (1)
), a rotation detector (3) that detects the rotation speed of the three-phase electric motor (1), and a rotation detector (3) that collects waveform data for frequency analysis from the output of the current detector (2).
3); a data collection unit (1) that collects rotation detection data from the output of the data collection unit (1); and a frequency analysis means that performs frequency analysis by Fourier transforming the frequency analysis waveform data collected by the data collection unit (1). (II), power frequency maximum point searching means (I[[) for finding the maximum first local maximum point degree from the analysis result by the frequency analysis means (I[)); A comparison means (IV) for power frequency compares the magnitude of the values of components of orders before and after the maximum point order, and a comparison means (IV) for power frequency compares the values of the components of the first maximum point order based on the comparison result of the comparison means (IV) for power supply frequency. power supply frequency correction calculation means (V) for calculating the value of the power supply frequency component from the value and the larger value of the components of orders before and after the first local maximum point order; and the data collection unit (1). an order calculation means (Vl) that calculates an order corresponding to the beat frequency from the rotation detection data collected by; and an order corresponding to the beat frequency calculated by the order calculation means (Vl) from the analysis result by the frequency analysis means (II). A beat frequency maximum point search means (■) for finding a second local maximum point order in which the value of the component is maximum in the vicinity, and a magnitude of the values of components of orders before and after the second local maximum point order in the analysis result. Beat frequency comparison means (
Based on the comparison result of the beat frequency comparison means (■), the value of the component of the second maximum point order or the component of the order before or after the second maximum point order, whichever is larger. a beat frequency correction calculation means (IX) for calculating the value of the beat frequency component from the above, and an output means (X) for calculating and outputting the ratio between the value of the beat frequency component and the value of the power supply frequency component. It is characterized by having a configuration in which it is equipped with. In this case, the data collection section (2) includes A/D converters 4, 5 . Fixed frequency oscillator6. It is composed of a data memory 8 and the like, and the other means (1) to (X) are realized by a central processing unit (described later).

As described above, the motor diagnostic device of the present invention determines and displays the motor primary current component ratio at the beat frequency and power supply frequency, which vary depending on whether the rotor coil of the three-phase motor is normal or abnormal. By quantitatively comparing the above ratios, it is possible to accurately diagnose rotor coils, etc., regardless of individual differences.

In addition, based on the frequency analysis results by Fourier transform, the maximum point calculation means, comparison means, and correction calculation means are used to calculate the
Since errors due to non-synchronization of the power supply frequency and beat frequency component values with the sampling frequency are corrected, the component values at the power supply frequency and beat frequency can be accurately determined, and diagnostic accuracy can be improved.

Embodiment An embodiment of the present invention will be described with reference to FIGS. 2 to 8. As shown in Fig. 2, this electric motor diagnostic device has the following features:
The primary current of the three-phase induction motor 1 to be diagnosed is measured using a current detector (
At the same time, the rotation of the three-phase induction motor 1 is detected by a rotation detector (pulse generator) 3, and the obtained signals are A/D converted by A/D converters 4 and 5, respectively. It is configured as follows.

The A/D conversion processing of the A/D converters 4 and 5 is performed according to the sampling pulse from the fixed frequency oscillator 6,
When diagnosing the rotor coil (during frequency analysis), set the frequency to 32 times the power supply frequency, for example. By collecting each waveform data and storing it in the data memory 8, and performing various calculations based on the waveform data stored in the data memory 8, the rotor coil of the three-phase induction motor is diagnosed (layer short).

This data is displayed on the display 9.

Note that 10 is a program memory that stores a program for diagnosing the rotor coil.

Hereinafter, the operation of this motor diagnostic device will be explained in detail with reference to FIGS. 3 to 8.

As shown in FIG. 3, the central processing unit 7 first generates the necessary number of data for the Fourier transform (several tens of cycles of the power supply frequency, for example, 64 cycles), that is, 32×64=2
048 waveform data D1.5 from the A/D converter 4.5.
and rotation detection data D2 are collected and stored in the data memory 8 (step W).

Next, the power spectrum of the primary current is obtained by reading out the waveform data D1 from the data memory 8 and performing frequency analysis (Fourier transform is performed using 2048 pieces of waveform data as one cycle) (step W2).

In this case, the frequency analysis results are as shown in Fig. 6, assuming that the rated power supply frequency rpo is 60 Hz, the sampling frequency is 60 x 321) z, and the number of discrete Fourier transform waveform data is 2048, that is, 64 cycles of the power supply frequency. As shown in , it is obtained as the value of each harmonic component with the fundamental frequency as 60/641)z, and 6'O/
641) The value of the frequency component for each z is obtained.

Next, based on the frequency analysis results in the previous step W2, calculations shown in the flowchart of FIG. (explained in relation to the figure).

This step W3 is executed because, in the frequency analysis of step W2, the power supply frequency is not constant at 60Hz, which is the rated power supply frequency, but fluctuates from time to time within the range of 59 to 61Hz, for example, when the power supply frequency is 59Hz. .51)z
This is because even when data was collected at the time of the analysis, only the value of the 64th order (m=64) component, that is, the 60)1z component, was obtained, and the value at the accurate power frequency was not obtained. , by performing the calculation in step W3 above, an accurate power supply frequency is determined, and an accurate value A of the power supply frequency component is determined.

Next, the data D2 is read from the data memory 8, the number of sampling data within a period from the timing when the data D2 becomes "1" to the timing when the data D2 becomes "1" next, for example, is calculated, and this number of sampling data and the sampling data are counted. The rotation speed N (rpm) of the three-phase induction motor 1 is determined from the frequency and the number of pulses output from the pulse generator 3 per one rotation of the three-phase induction motor 1, and further, this rotation speed N and the synchronous rotation speed Ns (rpm) are calculated. ), calculate the periphery S by performing the calculation shown in the following equation, and further calculate the order rnp of the power supply frequency f (obtained in step W3).
The order mB of the beat frequency fBo is calculated using
O is determined (step W4).

Next, based on the frequency analysis result in step W2 and the order mso of the beat frequency rso obtained in the previous step W4, calculations shown in the flowchart of FIG. 5 are performed to obtain an accurate beat frequency rB and the value of its frequency component. A8
(Step W5: Detailed explanation will be given in conjunction with FIG. 5).

This step W5 is executed because the beat frequency f8 is not an integral multiple of 60/64 Hz, and an accurate value at the beat frequency fB cannot be obtained, so the calculation in step W is performed. The accurate beat frequency is obtained by performing the following steps, and the accurate beat frequency component value A
We are looking for 8.

Next, the ratio between the value AB of the beat frequency component and the value A of the power supply frequency component is calculated by the following equation, and this ratio is output to an output device such as a printer (CRT) together with the beat frequency f8.
, printer, etc.) 9 (step W6).

Next, the calculation in step W3 in FIG. 3 will be explained based on the flowchart in FIG. 4. Ma-? 4ksp
%,'mm island fin r+whJfr-j-tsu bite, InWh
Find the order where the value of the component is maximum around -(=64). Then, let this order be Inl (ml is an integer), its frequency be rl, and the value of its component be V (ml).

Then, the value of the next (m I-1) component V(m+1)
and (m, +1) and the value of the next component V (mt+1), and when V (ml-1) ≧ ■ (m1 + 1) - Old... (2), k = V (ml) / V ( m, -1) -old-+31"
d=1/ (k+1)...
・(4) Ap = V (ml) / (yc d/a
+nπd) −-+51f, = f1 (1-67m1)
-Old-(er1mP=m1 (1-67m1
) ・・・・・・Perform each operation in (7) (rnp is a real number) and find the accurate power supply frequency r, this power supply frequency f,
Find the value A of the component and its order m.

On the other hand, as a result of the above comparison, V (ml -1)<V (m+, + 1)...
...When (8), = V (m,) / V (ml + 1) ... Self - (
(1) d=1/(k+1)
...QllAp = V (ml)
/ (πd/a+nrtd) −Qυr,=f,
(1-d/m,) -------021
mp=ml (1d/m1) --Qm
(rnp is a real number) and calculates the power supply frequency r.

The value Δ32nd order m of the component of the power supply frequency [, is determined.

Next, the calculation in step W in FIG. 3 will be explained based on the flowchart in FIG. 5. First, an order whose component value is maximum near the order mho corresponding to the beat frequency fBO is found.

Then, let this order be m2 (m2 is an integer), its frequency be f2, and the value of its component be V, (m-2).

Then, the value of the next (m2-1) component V (rn 1-1
) and the value of the (m2+1) degree component V (m2+l
), and V (rr+z 1 ) ≧V C'rn2 + 1
)......041, k - V (m2) / V (m2 1.)
River −α d=1/(k+1)
・・・・・・Scream AB =V (m2) / (
πd/5ind) -...-Q71fB= f2
(1-d/m2) . . . -Ql is performed to obtain the accurate beat frequency fB and the value A8 of the component of this beat frequency f.

On the other hand, as a result of the above comparison, V (m2-1)<V (m2+ l) ・・−
−−α When the odor is present, k=V (m2) /V (m2 +1) −−Q
md=1/ (k+1) ...(
21) AB = V (m2) / <nd/aln
πd) −(22)r, = r2 (1-d/m2)
・ Perform each demonstration of (23) to find the beat frequency fP
, the value A8 of the component of the beat frequency fB is determined.

Next, the ratio between the power spectrum value P (fB) at the beat frequency f8 and the power spectrum value P (f,) at the power supply frequency f, and the layer short of the rotor coil. The relationship with disconnection will be explained. The use of the test motor is as follows.

Rated output 3KW Rated voltage: 400V Rated current: 12A Number of poles = 6 Rated rotation speed: 1) 44 rpm Synchronous rotation speed: 1200 rpm Power frequency: 60 Hz Figure 7 (A) shows the case where the rotation of the 3-phase induction motor is normal. FIG. 2 is a diagram of the primary current waveform of the first phase of FIG.

FIG. 7(B) shows the power spectrum, where the vertical axis shows the percentage of the effective value and the horizontal axis shows the frequency.

FIG. 8(A) is a first phase primary current waveform diagram when the rotor winding is unbalanced, and FIG. 8(B) is its power spectrum.

The detected rotation speed during normal operation was 1) 44 rpm, so the slip SI was 0.0467 as shown in equation (24), and the beat frequency fB1 was 5 as shown in equation (25).
4.41) z.

rBl=60x (L-2x0.0467)=54.4
Hz...(25) As is clear from Figure 7 (A), no beat waveform appears in the primary current waveform in this case, and from Figure 7 (B), 54.4 Hz
It can be seen that the level of is 0.

On the other hand, the detected rotation speed at the time of abnormality was 1) 10 rpm, so the slip S2 was 0.075 as shown in equation (26).
become. Then, the beat frequency fB2 becomes 51 Hz as shown in equation (27).

f s2-6Qx (12X0.O75)=51Hz
...127) As is clear from Fig. 8 (A), the primary current wave is 1-, 2, 9 rM oil cramped chome-j1 negative QrEfl (Q) end 18-day HD-. As shown, a power spectrum of about 9% is seen at 51 Hz, and from this, D=9% is known and an abnormality in rotation can be detected.

In this way, the beat frequency can be accurately known, and abnormalities in the rotation of the motor can be detected from the ratio with the power spectrum at the power supply frequency. Therefore, if a criterion value obtained through experimentation or actual results is determined, a quantitative determination of normality or abnormality can be made by directly comparing the detection result with the criterion value. As a result, it is possible to diagnose the electric motor, which differs from person to person, and to prevent serious failures of the electric motor.

In addition, since the Fourier transform results are further processed to accurately determine the values of the power supply frequency, beat frequency, and their components, errors due to non-synchronization between the sampling frequency of the Fourier transform and the power supply frequency and beat frequency can be eliminated. The ratio of each frequency component can be determined extremely accurately, and the reliability of abnormality diagnosis can be improved.

Effects of the Invention The motor diagnostic device of the present invention determines and displays the component ratio of the motor primary current at the beat frequency and power supply frequency, which vary depending on whether the rotor coil of the three-phase motor is normal or abnormal. By making quantitative comparisons, it is possible to accurately diagnose rotor coils, etc., regardless of individual differences.

In addition, the Fourier transform results are further processed to accurately determine the values of the power supply frequency, beat frequency, and their components, eliminating errors caused by asynchrony between the sampling frequency of the Fourier transform and the power supply frequency and beat frequency. The ratio of each frequency component can be determined extremely accurately, and the reliability of abnormality diagnosis can be improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the present invention, FIG. 2 is a block diagram of an embodiment of the invention, FIG. 3 is a flowchart showing the main program of the central processing unit, and FIG. 4 is the same as that shown in FIG. A detailed flowchart of the main parts, Fig. 5 is a detailed flowchart of the main parts in Fig. 3,
Figure 6 is an explanatory diagram of the Fourier transform results, Figures 7 and 8
The figure is an explanatory diagram of the relationship between frequency analysis results and whether the rotor coil is normal or abnormal. 1... 3-phase induction motor, 2... current detector, 3...
・Rotation detector, ■...Data collection unit, ■...Frequency analysis means, ■...Local maximum point search means for power frequency, ■...
... Comparison means for power supply frequency, ■ ... Correction calculation means for power supply frequency, ■ ... Order calculation means, ■ ... Maximum point search means for beat frequency, ■ ... Comparison means for beat frequency, ■ . . . Beating frequency correction calculation means,

Claims (1)

[Claims] A current detector (2) that detects the primary current of a three-phase motor (1).
), a rotation detector (3) that detects the rotation speed of the three-phase electric motor (1), and a rotation detector (3) that collects waveform data for frequency analysis from the output of the current detector (2).
a data collection unit (1) that collects rotation detection data from the output of 3), and a frequency analysis means (II) that performs frequency analysis by Fourier transforming the frequency analysis waveform data collected by this data collection unit (I). and the frequency analysis means (
power frequency maximum point searching means (III) for finding a first local maximum point order in which the value of a component is maximum near the order corresponding to the rated power frequency from the analysis result according to II); power frequency comparison means (IV) that compares the magnitude of the values of the components of the orders before and after the local maximum point order; power frequency correction calculation means (V
), an order calculation means (VI) for calculating the order corresponding to the beat frequency from the rotation detection data collected by the data collection section (I), and an order calculation means (VI) for calculating the order corresponding to the beat frequency from the rotation detection data collected by the data collection section (I); ), a beat frequency maximum point search means (VII) for finding a second local maximum point order in which the value of the component is maximum near the order corresponding to the beat frequency obtained by Beat frequency comparison means (V
III) and the comparison result of this beat frequency comparison means (VIII), the value of the component of the second local maximum point order or the value of the component of the orders before or after the second local maximum point order, whichever is larger. a beat frequency correction calculation means (IX) for calculating the value of the beat frequency component from the above, and an output means (X) for calculating and outputting the ratio of the value of the beat frequency component and the value of the power supply frequency component. Equipped with electric motor diagnostic equipment.