CN101363901B - Method for detecting early failure of generator by enhancing transformations by electrical current characteristic - Google Patents

Abstract

The invention provides a method for the early failure detection of a motor using current characteristic enhancement transformation, which comprises (1) real-time collecting three-phase current of an induction machine ia, ib, and ic, and calculating a transformation matrix K based on a current power frequency fe; (2) demodulating via the following formula to obtain a torque current iq and a magnetizing current id, wherein iqd0 is equal to Kiabc; (3) performing fast Fourier transformation of torque current iq and the magnetizing current to obtain a subtle current characteristic spectrum; and (4) comparing with the failure characteristic frequency of the induction machine, and determining the failure types of the induction machine. The method has the advantages of simple principle, simple operation, and high measuring accuracy, and can extract failure characteristic component in severe power frequency background using the current characteristic enhancement transformation to detect the early failure of the induction machine.

Description

Utilize current characteristic to strengthen the transfer pair motor and carry out the method that initial failure detects

Technical field

The present invention is mainly concerned with the harmless online diagnosing technique of support shaft field of induction machine fault, refers in particular to a kind of current characteristic that utilizes and strengthens the method that the transfer pair motor carries out the initial failure detection.

Background technology

In the prior art, current characteristic analysis commonly used is monitored the state of triphase induction type motor, and its most common failure such as rotor broken bar, wire turn short circuit, gap off-centre etc. are diagnosed.In this technology, the induction machine electric current that sensor obtains becomes digital signal through data acquisition earlier, and algorithm is analyzed after diagnosing then.Analytical approach commonly used mainly is the current signal analysis of spectrum, if there is fault, the side frequency component will occur in the frequency spectrum.Usually, the frequency of side frequency component is very near the 50Hz power frequency, but the amplitude of side frequency component is far smaller than the amplitude of power frequency component, and the side frequency component of reflection fault will be covered by power frequency component and much noise as small-signal.If do not remove power frequency component, the variation of side frequency component just might be difficult to detect, and then causes more serious fault.Therefore, how under serious power frequency background, to remove power frequency component, and extract the side frequency component, just become one of key issue of induction machine current characteristic analysis.The method of elimination at present or decay power frequency component all needs hardware handles systems such as trapper usually.

Summary of the invention

The problem to be solved in the present invention just is: at the technical matters that prior art exists, the invention provides that a kind of principle is simple, easy and simple to handle, measuring accuracy is high, can utilize current characteristic to strengthen conversion extracts the Weak fault characteristic component induction machine is carried out the method that initial failure detects under serious power frequency background.

For solving the problems of the technologies described above, the solution that the present invention proposes is: a kind of current characteristic that utilizes strengthens the method that the transfer pair motor carries out the initial failure detection, it is characterized in that step is:

1. gather the three-phase electricity flow valuve i of induction machine in real time _a, i _b, i _c, and according to electric current power frequency f _eThe computational transformation matrix K;

2. by obtaining the moment of torsion current i after the following formula demodulation _qWith magnetization current i _d:

i _qd0＝Ki _abc

(i in the formula _Abc) ^T=[i _ai _bi _c] (i _Qd0) ^T=[i _qi _di ₀]

$K=\frac{2}{3}\left[\begin{array}{ccc}\mathrm{<figure-callout\; id="2"\; label="cos"\; filenames="H2008101432685E00021.png,H2008101432685E00022.png"\; state="\{\{state\}\}">cos</figure-callout>}2\mathrm{\&pi;}{f}_{e}t& \cos (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \cos (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \sin 2\mathrm{\&pi;}{f}_{e}t& \sin (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \sin (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \frac{1}{2}& \frac{1}{2}& \frac{1}{2}\end{array}\right];$

3. with the moment of torsion current i that obtains in the above-mentioned steps _qWith magnetization current i _dCarry out obtaining meticulous current characteristic spectrogram after the fast fourier transform;

4. the current characteristic spectrogram that obtains in the above-mentioned steps and the fault characteristic frequency of induction machine are compared, determine the fault type of induction machine.

4. described step is used for the characteristic frequency of the embodiment induction machine rotor bar breaking fault compared and is:

f _brb＝f _e(1±2ks) k＝1，2，…

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and k is a natural number.

4. described step is used for the characteristic frequency of the embodiment shorted-turn fault compared and is:

$f_{\mathrm{st}}=f_e[\frac{n}{p}(1-s)\&PlusMinus;k]$ n＝1，2，3，… k＝1，3，5，…

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and n is a natural number, and k is a positive odd number, and p is the motor number of pole-pairs.4. described step is used for the characteristic frequency of the embodiment gap fault of eccentricity compared and is:

$f_{\mathrm{ecc}}=f_e[1\&PlusMinus;k\left(\frac{1-s}{p}\right)]$ k＝1，2，…

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and k is a natural number, and p is the motor number of pole-pairs.

Compared with prior art, advantage of the present invention just is: the present invention utilizes current characteristic to strengthen the method that the transfer pair motor carries out the initial failure detection, no longer needs to use hardware systems such as trapper; The present invention can remove power frequency component serious in the stator current, demodulates the fault signature component, and the fault signature that obtains is very obvious; Even very under the situation near the electric current power frequency, the present invention still can extract fault signature at certain fault characteristic frequency; Under the lower situation of induction machine load, the present invention still can extract faint fault signature component, detects the initial failure of induction machine.

The inventive method is that a kind of current characteristic that utilizes strengthens the method that the transfer pair motor carries out the initial failure detection, be characterized in need not to increase extra hardware system, utilize reference synchronization coordinate demodulation conversion, the three-phase current demodulation is transformed to magnetic-moment of torsion electric current equivalent space, and detect the induction machine initial failure with the spectrum signature of two magnetic-moment of torsion electric current that obtains after the demodulation, this method provides a kind of new effective technical means for solving the key issue that the fault signature component is difficult to extract under the serious power frequency background in the analysis of induction machine current characteristic.

Description of drawings

Fig. 1 is the triangle relation synoptic diagram of all electric current variablees in the induction machine;

Fig. 2 is the schematic flow sheet of fault detection method of the present invention;

Fig. 3 is supposing that constant full load of induction machine and rotating speed are when carrying out emulation under the 1416r/min, three-phase current signal figure under the induction machine rotor broken bar situation of acquisition;

Fig. 4 is that the M-T signal magnetization current that three-phase current signal demodulation conversion obtains divides spirogram;

Fig. 5 is the spectrogram of M-T signal magnetization current component after fast fourier transform;

Fig. 6-the 1st, zero load down healthy rotor is removed trend back magnetization current component figure;

Fig. 6-the 2nd, the spectrogram of signal after fast fourier transform among Fig. 6-1;

Fig. 6-the 3rd, zero load down disconnected bar rotor is removed trend back magnetization current component figure;

Fig. 6-the 4th, the spectrogram of signal after fast fourier transform among Fig. 6-3;

Fig. 7-1 is the down healthy rotor removal of 25% load trend back magnetization current component figure;

Fig. 7-the 2nd, the spectrogram of signal after fast fourier transform among Fig. 7-1;

Fig. 7-3 is the down disconnected bar rotor removal of 25% load trend back magnetization current component figure;

Fig. 7-the 4th, the spectrogram of signal after fast fourier transform among Fig. 7-3;

Fig. 8-1 is the down healthy rotor removal of 50% load trend back magnetization current component figure;

Fig. 8-the 2nd, the spectrogram of signal after fast fourier transform among Fig. 8-1;

Fig. 8-3 is the down disconnected bar rotor removal of 50% load trend back magnetization current component figure;

Fig. 8-the 4th, the spectrogram of signal after fast fourier transform among Fig. 8-3;

Fig. 9-1 is the down healthy rotor removal of 75% load trend back magnetization current component figure;

Fig. 9-the 2nd, the spectrogram of signal after fast fourier transform among Fig. 9-1;

Fig. 9-3 is the down disconnected bar rotor removal of 75% load trend back magnetization current component figure;

Fig. 9-the 4th, the spectrogram of signal after fast fourier transform among Fig. 9-3;

Figure 10-1 is the down healthy rotor removal of 100% load trend back magnetization current component figure;

Figure 10-the 2nd, the spectrogram of signal after fast fourier transform among Figure 10-1;

Figure 10-3 is the down disconnected bar rotor removal of 100% load trend back magnetization current component figure;

Figure 10-the 4th, the spectrogram of signal after fast fourier transform among Figure 10-3.

Embodiment

Below with reference to the drawings and specific embodiments the present invention is described in further details.

Referring to shown in Figure 2, the inventive method is that a kind of current characteristic that utilizes strengthens the method that the transfer pair motor carries out the initial failure detection, the steps include:

1. gather the three-phase electricity flow valuve i of induction machine in real time _a, i _b, i _c, and according to electric current power frequency f _eThe computational transformation matrix K,

$K=\frac{2}{3}\left[\begin{array}{ccc}\mathrm{<figure-callout\; id="2"\; label="cos"\; filenames="H2008101432685E00021.png,H2008101432685E00022.png"\; state="\{\{state\}\}">cos</figure-callout>}2\mathrm{\&pi;}{f}_{e}t& \cos (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \cos (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \sin 2\mathrm{\&pi;}{f}_{e}t& \sin (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \sin (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \frac{1}{2}& \frac{1}{2}& \frac{1}{2}\end{array}\right];$

2. by obtaining the moment of torsion current i after the following formula demodulation _qWith magnetization current i _d:

i _qd0＝Ki _abc

In the formula, (i _Abc) ^T=[i _ai _bi _c], (i _Qd0) ^T=[i _qi _di ₀], i _Qd0Represent one by i _q, i _d, i ₀Three column vectors that element constitutes, (i _Qd0) ^TExpression i _Qd0Transposition, T represents vectorial transposition, i ₀Be to be independent of f _eNull variable;

3. with the moment of torsion current i that obtains in the above-mentioned steps _qWith magnetization current i _dCarry out obtaining meticulous current characteristic spectrogram behind the fast Fourier transform (FFT);

4. the current characteristic spectrogram that obtains in the above-mentioned steps and the fault characteristic frequency of induction machine are compared, determine the fault type of induction machine.

In the principle of said method of the present invention, the triangle relation of all electric current variablees can obtain following formula as shown in Figure 1 in view of the above in the induction machine:

i _a＝i _qcos2πf _et+i _dsin2πf _et (1)

$i_b=i_q\cos (2\mathrm{\&pi;}{f}_{e}t-\frac{2\mathrm{\&pi;}}{3})+i_d\sin (2\mathrm{\&pi;}{f}_{e}t-\frac{2\mathrm{\&pi;}}{3})---\left(2\right)$

$i_c=i_q\cos (2\mathrm{\&pi;}{f}_{e}t+\frac{2\mathrm{\&pi;}}{3})+i_d\sin (2\mathrm{\&pi;}{f}_{e}t+\frac{2\mathrm{\&pi;}}{3})---\left(3\right)$

Just:

$\left[\begin{array}{c}{i}_a\\ i_b\\ i_c\end{array}\right]=\left[\begin{array}{cc}\mathrm{<figure-callout\; id="2"\; label="cos"\; filenames="H2008101432685E00021.png,H2008101432685E00022.png"\; state="\{\{state\}\}">cos</figure-callout>}2\mathrm{\&pi;}{f}_{\mathrm{<figure-callout\; id="2"\; label="e\; t\; sin"\; filenames="H2008101432685E00021.png,H2008101432685E00022.png"\; state="\{\{state\}\}">e</figure-callout>}}t& \sin 2\mathrm{\&pi;}{f}_{e}t\\ \cos (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \sin (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})\\ \cos (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})& \sin (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\end{array}\right]\&CenterDot;\left[\begin{array}{c}{i}_q\\ i_d\end{array}\right]=w\left(t\right)\&CenterDot;\left[\begin{array}{c}{i}_q\\ i_d\end{array}\right]---\left(4\right)$

In the formula, i _a, i _b, i _cBe actual three-phase current variable, i _q, i _dBe the electric current variable that conversion obtains, W is a transformation matrix, f _eBe the electric current power frequency, t is for starting the motor run-time variable of counting from this detection method.

As everyone knows, actual three-phase current i _a, i _b, i _cCan directly measure, but i _q, i _dCan not directly measure, only can be by i _a, i _b, i _cEstimate.When measuring i _a, i _b, i _cAfterwards, the inverse transformation according to following formula is shown below, and can obtain i _q, i _d

$\left[\begin{array}{c}{i}_q\\ i_d\end{array}\right]={\left[W{\left(t\right)}^{T}W\left(t\right)\right]}^{-1}W{\left(t\right)}^T\left[\begin{array}{c}{i}_a\\ i_b\\ i_c\end{array}\right]=W^{+}\left[\begin{array}{c}{i}_a\\ i_b\\ i_c\end{array}\right]---\left(5\right)$

In the formula, W ⁺It is the pseudoinverse of W.

Theory according to Paul C.Krause etc. has a reference synchronization coordinate transform framework can realize the conversion of three-phase steady-state current variable to any reference coordinate.Reference synchronization coordinate transform framework and top two formula are combined, above the demodulation conversion process can be described below:

i _qd0＝Ki _abc (6)

In the formula

(i _abc) ^T＝[i _ai _bi _c]

(i _qd0) ^T＝[i _qi _di ₀]

$K=\frac{2}{3}\left[\begin{array}{ccc}\mathrm{<figure-callout\; id="2"\; label="cos"\; filenames="H2008101432685E00021.png,H2008101432685E00022.png"\; state="\{\{state\}\}">cos</figure-callout>}2\mathrm{\&pi;}{f}_{e}t& \cos (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \cos (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \mathrm{<figure-callout\; id="2"\; label="sin"\; filenames="H2008101432685E00021.png,H2008101432685E00022.png"\; state="\{\{state\}\}">sin</figure-callout>}2\mathrm{\&pi;}{f}_{e}t& \sin (2\mathrm{\&pi;}{f}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \sin (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \frac{1}{2}& \frac{1}{2}& \frac{1}{2}\end{array}\right]$

Abc reference frame variable i _a, i _bAnd i _cIn characteristic component modulated by power frequency component, according to top reference synchronization coordinate demodulation transformation for mula (6), can be with characteristic component from i _a, i _b, i _cDemodulation transforms to qd0 reference frame variable i _q, i _dIn, that is to say, at i _q, i _dIn, power frequency component has been eliminated by demodulation.Therefore, can better detect initial failure than abc component, this core also of the present invention just place by analyzing the qd0 component.The i that three-phase current signal obtains through reference synchronization coordinate demodulation conversion _q, i _dBe called moment of torsion electric current and magnetization current, be referred to as magnetic-moment of torsion (M-T) electric current.Two M-T electric currents after the demodulation can extract meticulousr fault signature, can more effectively detect the induction machine initial failure.

Method of the present invention is applicable to the early detection of faults such as induction machine rotor broken bar, wire turn short circuit, gap off-centre.The principal character frequency that is used for detecting these faults is near the additional side frequency component that exists the stator current frequency spectrum electric current power frequency.Difference according to the side frequency component is diagnosed specific fault.For example, the characteristic frequency of diagnosis rotor broken bar, wire turn short circuit and gap fault of eccentricity can be passed through formula (7) respectively, calculate with (9) (8).

The characteristic frequency of rotor bar breaking fault is

f _brb＝f _e(1±2ks) k＝1，2，… (7)

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and k is a natural number.

The characteristic frequency of shorted-turn fault is

$f_{\mathrm{st}}=f_e[\frac{n}{p}(1-s)\&PlusMinus;k]$ n＝1，2，3，… k＝1，3，5，… (8)

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and n is a natural number, and k is a positive odd number, and p is the motor number of pole-pairs.The characteristic frequency of gap fault of eccentricity is

$f_{\mathrm{ecc}}=f_e[1\&PlusMinus;k\left(\frac{1-s}{p}\right)]$ k＝1，2，… (9)

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and k is a natural number, and p is the motor number of pole-pairs.

In the induction machine theory, the corresponding distinctive characteristic frequency of every kind of fault, and also these characteristic frequencies have the corresponding calculated formula, as formula (7), (8) and (9).In the engineering, can judge the fault type that induction machine exists by the characteristic frequency component that exists on the stator current spectrogram according to this corresponding relation of fault and its characteristic frequency.When the characteristic frequency of induction machine fault during, be difficult to from the frequency spectrum of former three-phase current signal, find out this fault signature very near the electric current power frequency.But there is not this limitation in the frequency spectrum of the M-T signal that is obtained by former three-phase current signal conversion, because the electric current power frequency component has been eliminated by demodulation in the M-T signal.This is the advantage place of the inventive method just.

With the induction machine rotor bar breaking fault is example, and the checking of its emulated data is as follows: utilize three-phase current signal (i under the induction machine rotor broken bar situation that emulation shown in Figure 3 obtains _a, i _b, i _c), according to reference synchronization coordinate demodulation transformation for mula (6), obtain M-T signal magnetization current component shown in Figure 4 by the three-phase current signal conversion.According to the analysis of front, the demodulation of M-T signal reply power frequency.For this reason, the M-T signal is made FFT and provided frequency spectrum as shown in Figure 5.Obviously, as can be seen from Figure 5 50Hz power frequency component and odd harmonics thereof are eliminated, and characterize the 5.6Hz characteristic frequency (f of rotor broken bar _Brb) component and 2 subharmonic thereof are fairly obvious.Therefore, the frequency spectrum of M-T signal can reflect induction machine rotor bar breaking fault feature well, has verified the validity of the inventive method.

Be example with actual sensed formula rotor broken bar fault below, verify the inventive method elimination power frequency component and extract the fault signature component to be used for the validity that the induction machine initial failure detects.As for other fault, the step that detects with this method is the same, and just the characteristic frequency value that finally obtains on spectrogram is in different size.

Testing used induction machine is three-phase Y-connection 4 utmost point 12kW induction machines, and electric current is about 25A at full capacity.Induction machine adopts the power supply of 50Hz power supply, and experiment is to carry out for 0.0633 time at full load, rotating speed 1404r/min, unit revolutional slip, and load is provided by Dyn. and variable load station.

Be respectively at load under 0,25%, 50%, 75% and 100% the situation, gathered the phase current signal of healthy and rotor broken bar two states induction machine respectively.To 5 kinds of operating modes of 2 class induction machines, with sample frequency 51200Hz, gathered 10 groups of data of three-phase current respectively, every group of number of data points is 161946.The variable load condition test shows that the dependence of diagnostic measures value and load is more little, and it is reliable that monitoring parameter is got over robust.On the basis of above-mentioned data, the data set that has generated at 4096 every group is used for following analysis.According to above-mentioned acquisition parameter, it is as shown in table 1 to obtain relevant current characteristic estimated parameter.

The current characteristic estimated parameter (△ f is a frequency resolution) that table 1 is relevant

Load (%)	Electric current power frequency f e(Hz)	f eLeft side frequency component (Hz)	Actual axle is (Hz) frequently	The revolutional slip s of unit	Rotor bar breaking fault characteristic frequency f brb(Hz)
						0	49.9525	24.9762	24.9763	0	0
25	49.9525	25.2924	24.6601	0.0127	1.2688±△f
						50	49.9525	25.6085	24.3440	0.0253	2.5276±△f

75	49.9525	25.9247	24.0278	0.0380	3.7964±△f
						100	49.9525	26.557	23.3955	0.0633	6.3240±△f

By reference synchronization coordinate demodulation conversion, Fig. 6 series-Figure 10 series has provided healthy rotor respectively and disconnected bar rotor is the magnetic after the phase current conversion-moment of torsion current component and FFT analysis of spectrum result thereof under 0,25%, 50%, 75% and 100% situation at the induction machine load.

Referring to Fig. 6-1,6-2,6-3,6-4, the not significant difference of the spectrum signature of healthy rotor and disconnected bar rotor.Because under the zero load situation, the revolutional slip s=0 of unit, the characteristic frequency of rotor bar breaking fault can not occur.

Referring to Fig. 7-1,7-2,7-3,7-4,, but still can find out this characteristic frequency component (shown in Fig. 7-4) although cause the amplitude of rotor bar breaking fault characteristic frequency component very little because load is low.Also provable thus, the inventive method can be than detecting early stage rotor bar breaking fault under the underload.

Along with load increases gradually, rotor bar breaking fault characteristic frequency component is more and more significant, for example 50% load condition (shown in Fig. 8-1,8-2,8-3,8-4) and 75% load condition (shown in Fig. 9-1,9-2,9-3,9-4).

From Figure 10-1,10-2,10-3,10-4 as can be seen, under high capacity or full load, the characteristic frequency and the higher hamonic wave thereof of rotor bar breaking fault are fairly obvious.

These above-mentioned actual sensed formula electrical fault examples have been verified the inventive method elimination power frequency component and have extracted the fault signature component to be used for the validity that the induction machine initial failure detects.

The above only is a preferred implementation of the present invention, and protection scope of the present invention also not only is confined to the foregoing description, and all technical schemes that belongs under the thinking of the present invention all belong to protection scope of the present invention.Should be pointed out that for those skilled in the art in the some improvements and modifications that do not break away under the principle of the invention prerequisite, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (4)

1. one kind is utilized current characteristic to strengthen the method that the transfer pair motor carries out the initial failure detection, it is characterized in that step is:

1. gather the three-phase electricity flow valuve i of induction machine in real time _a, i _b, i _c, and according to electric current power frequency f _eThe computational transformation matrix K,

$K=\frac{2}{3}\left[\begin{array}{ccc}\cos 2{\mathrm{\&pi;f}}_{e}t& \cos ({2\mathrm{\&pi;f}}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \cos ({2\mathrm{\&pi;f}}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \sin {2\mathrm{\&pi;f}}_{e}t& \sin (2{\mathrm{\&pi;f}}_{e}t-\frac{2}{3}\mathrm{\&pi;})& \sin (2\mathrm{\&pi;}{f}_{e}t+\frac{2}{3}\mathrm{\&pi;})\\ \frac{1}{2}& \frac{1}{2}& \frac{1}{2}\end{array}\right];$

2. by obtaining the moment of torsion current i after the following formula demodulation _qWith magnetization current i _d:

i _qd0＝Ki _abc

(i in the formula _Abc) ^T=[i _ai _bi _c] (i _Qd0) ^T=[i _qi _di ₀], i ₀Be to be independent of f _eNull variable;

3. with the moment of torsion current i that obtains in the above-mentioned steps _qWith magnetization current i _dCarry out obtaining meticulous current characteristic spectrogram after the fast fourier transform;

4. the current characteristic spectrogram that obtains in the above-mentioned steps and the fault characteristic frequency of induction machine are compared, determine the fault type of induction machine.

2. according to claim 1ly utilize current characteristic to strengthen the transfer pair motor to carry out the method that initial failure detects, it is characterized in that, 4. described step is used for the characteristic frequency f of the embodiment induction machine rotor bar breaking fault compared _BrbFor:

f _brb＝f _e(1±2ks) k＝1，2，…

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and k is a natural number.

3. according to claim 1ly utilize current characteristic to strengthen the transfer pair motor to carry out the method that initial failure detects, it is characterized in that, 4. described step is used for the characteristic frequency f of the embodiment shorted-turn fault compared _StFor:

$f_{\mathrm{st}}=f_e[\frac{n}{p}(1-s)\&PlusMinus;k]$ n＝１，2，3，… k＝1，3，5，…

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and n is a natural number, and k is a positive odd number, and p is the motor number of pole-pairs.

4. according to claim 1ly utilize current characteristic to strengthen the transfer pair motor to carry out the method that initial failure detects, it is characterized in that, 4. described step is used for the characteristic frequency f of the embodiment gap fault of eccentricity compared _EccFor:

$f_{\mathrm{ecc}}=f_e[1\&PlusMinus;k\left(\frac{1-s}{p}\right)]$ k＝1，2，…

In the formula, f _eBe the electric current power frequency, s is the unit revolutional slip, and k is a natural number, and p is the motor number of pole-pairs.

Images