Method for identifying a fault in an electrical machine

1. A method for identifying a fault in an electrical machine having a rotor and a stator, the method comprising the steps of:

carrying out a first vibration measurement on vibrations of the stator in a first radial direction of the stator;

carrying out a second vibration measurement on vibrations of the stator in a second radial direction of the stator;

determining, on the basis of at least one of the first vibration measurement and the second vibration measurement, a first vibration frequency;

determining, on the basis of the first vibration measurement and the second vibration measurement, a mode shape of the vibration at the first vibration frequency; and

using a combination of the first vibration frequency and the mode shape to identify a fault condition of the electrical machine.

2. The method according to claim 1 , wherein the method comprises the steps of:

carrying out a plurality of vibration measurements in a plurality of different radial directions;

determining, on the basis of at least one of the plurality of vibration measurements, a first vibration frequency; and

determining, on the basis of the plurality of vibration measurements, a mode shape of the vibration at the first vibration frequency.

3. The method according to claim 1 , wherein the fault condition is identified when a vibration amplitude at the first vibration frequency exceeds a predetermined threshold value.

4. A method for identifying a fault in an electrical machine having a rotor and a stator, the method comprising the steps of:

carrying out a first vibration measurement on vibrations of the stator in a first radial direction of the stator;

carrying out a second vibration measurement on vibrations of the stator a second radial direction of the stator;

determining, on the basis of at least one of the first vibration measurement and the second vibration measurement, a first vibration frequency;

determining, on the basis of the first vibration measurement and the second vibration measurement, a mode shape of the vibration at the first vibration frequency;

carrying out vibration measurements with a first load and with a second load of the machine;

determining a difference in vibration amplitudes with a first load and with a second load at the first vibration frequency; and

using a combination of the first vibration frequency, the mode shape and the difference in vibration amplitudes to identify a fault condition of the electrical machine.

5. The method according to claim 1 , wherein the fault condition is one of the following: a broken rotor bar, dynamic eccentricity, static eccentricity, inter-turn short circuit, inter-coil short circuit.

6. The method according to claim 1 , wherein the method comprises the step of:

determining, on the basis that the first vibration frequency ƒ and the mode shape m fulfil one of the following conditions: ƒ=n·ƒ r or ƒ=n·ƒ r ±2·s·ƒ s and m=n, wherein n=(1, 3, 5, . . . ), ƒ r =rotation frequency of the motor, s=rotor slip and ƒ s =supply frequency, that a rotor bar is broken.

7. The method according to claim 1 , wherein the method comprises the step of:

determining, on the basis that the first vibration frequency ƒ and the mode shape m fulfil one of the following conditions: ƒ=2·ƒ r and m=2; ƒ=2·ƒ s −ƒ r and m=2·p−1; ƒ=2·ƒ s +ƒ r and m=2·p+1, wherein ƒ r =rotation frequency of the motor, ƒ s =supply frequency and p=number of stator pole pairs, that the rotor is dynamically eccentric.

8. The method according to claim 1 , wherein the method comprises the step of:

determining, on the basis that the first vibration frequency ƒ and the mode shape m fulfil the following conditions: ƒ=2·ƒ s and m=2·p+1 or m=2·p−1, wherein ƒ s =supply frequency and p=number of stator pole pairs, that the rotor is statically eccentric.

9. The method according to claim 1 , wherein the method comprises the step of:

determining, on the basis that the first vibration frequency ƒ and the mode shape m fulfil one of the following conditions: ƒ=2·k·ƒ s and m=(2, 4, 6, . . . ), wherein k=(1, 2, 3, . . . ) and ƒ s =supply frequency, that the stator coils have either an inter-turn short circuit or an inter-coil short circuit.

10. The method according to claim 9 , wherein the method comprises the steps of:

carrying out vibration measurements with a first load and with a second load of the machine, the a first load being smaller that the second load;

determining a difference in vibration amplitudes with a first load and with a second load at the first vibration frequency; and

determining, on the basis that the vibration amplitude increases with an increasing load and that the increase of the vibration amplitude exceeds a predetermined threshold value, that the stator coils have an inter-turn short circuit.

11. The method according to claim 1 , wherein the electrical machine is an induction motor.

12. A monitoring system for identifying a fault in an electrical machine having a rotor and a stator, the monitoring system comprising:

a first sensor arranged to measure vibrations of the stator in a first radial direction of the stator,

a second sensor arranged to measure vibrations of the stator in a second radial direction of the stator,

a processor receiving measurement signals from the first sensor and the second sensor, the processor comprising a first algorithm for detecting from the measurement signals a first vibration frequency and a mode shape of the vibration at the first vibration frequency, and the processor further comprising a second algorithm for identifying a fault condition of the electrical machine from the combination of the first vibration frequency and the mode shape.

13. The monitoring system according to claim 12 , wherein the monitoring system comprises a plurality of sensors arranged to measure vibration at a plurality of radial directions of the stator, and the processor receives measurement signals from the plurality of sensors.

14. The monitoring system according to claim 12 , wherein the sensors are accelerometers.

15. An induction motor comprising a monitoring system according to claim 12 .