Gear fault diagnosis by monitoring the stator current of the motor

Gears are the most common transmission components in mechanical equipment and are also a consumable part. Studying gear failures makes sense for reducing the maintenance costs of mechanical equipment and improving product quality. The most direct dynamic behavior of gear defects or faults (such as tooth surface damage, cracks, excessive wear and even broken teeth) is reflected in the torsional vibration of its drive shaft. This paper focuses on the indirect measurement of the torsional vibration of the drive shaft and the measurement of the vibration signal, as well as the processing of these signals, fault feature extraction, etc., but because the direct measurement of the torsional vibration of the gear is more difficult and costly, the author proposed to monitor the stator The current is used to realize a new method of indirect measurement of the torsional vibration of the gear, and the fault diagnosis of the gear is realized by extracting the signal characteristics of the stator current. The main idea is: the torsional vibration of the gear causes the fluctuation of the air gap torque of the motor through the rotary shaft. The change of the air gap torque causes the stator current to change through the stator flux. Therefore, the gear can be realized by analyzing the frequency component of the stator current. Identification of the frequency of failure.
1 The vibration signal characteristics of the damaged gear Under normal conditions, that is, when the gear is well meshed, the vibration of the gear transmission is mainly the gear mesh excitation vibration. The main component is the meshing frequency and its harmonic components, and the response can be expressed as: x ( t) is the measured vibration signal X is the first meshing frequency harmonic component amplitude O is the first meshing frequency harmonic component phase f is the gear meshing frequency, and f is the frequency of the gear axis.
When the gear has eccentricity, pitting and broken teeth, the vibration signal produces amplitude modulation, and the amplitude modulation is caused by the influence of the tooth surface load fluctuation on the vibration amplitude. Due to fluctuations in motor speed, gear machining error, scale deformation and gear tooth bending, the cyclical variation of the overall gear stiffness is caused, and the vibration of the gear is frequency modulated. In fact, it is caused by a gear failure with a frequency of 25 Hz. After unpacking inspection, it was found that gear 1 had severe pitting wear, which was inferred from the theory.
In summary, the torsional vibration information of the gear is very sensitive to the fault of the gear. Therefore, by analyzing the torsional vibration of the stator current of the motor, the fault characteristics of the gear can be extracted. Through a large number of experiments, the variation of the stator current signal from normal to slight damage to the final damage can be found. When the signal characteristics of the stator current in the various damage stages of the gear are mastered, the fault characteristics of the early stage of the gear damage can be distinguished. So as soon as possible to take measures to troubleshoot.
In addition, when the load changes, the frequency of change can also be measured by the stator current through the transmission system. Under normal circumstances, the load change frequency is different from the gear failure frequency. In this way, after mastering the characteristic law of the fault of each gear in the stator current, it is possible to improve the characteristics of the load change in the stator current, and provide a basis for measuring the variation law of the load.
4 Conclusions This paper proves both the theoretical and practical aspects of the effectiveness and practicability of gear fault diagnosis by monitoring the indirect monitoring of the torsional vibration of the gear. A typical gear fault diagnosis method is to measure the vibration signal of the gear through an acceleration sensor mounted on the bearing housing, extract the signal characteristics, and diagnose the fault of the gear. However, this method is limited by the field, it is inconvenient to install the sensor. In addition, the path of the vibration signal transmitted from the gear to the sensor is complicated, and the mathematical model of the transfer function and the gearbox is not easy to obtain. Therefore, the methods currently used are mostly qualitative studies. . The above-mentioned problems can be solved by using the theory proposed in this paper, and remote diagnosis can also be realized, which has high use value.