3 Bearing fault detection and diagnosis
At present, the method of detecting bearing faults in China generally has methods for measuring temperature, noise and vibration parameters. The method of measuring temperature and noise is relatively simple, but it lacks effective preventive effect. Once the bearing temperature is too high and noise is high during operation, It is generally believed that the locomotive bearings have reached a relatively serious degree of damage. The method of measuring vibration parameters is to test the bearing when the locomotive is in the middle and small repairs. Because it is convenient and simple to obtain and collect information in the test, it is more suitable for the use of the repair section, which is a simple and practical detection method. .
3.1 Diagnostic methods for locomotive
bearings
There are two main methods for locomotive bearing: (1) simple diagnosis of locomotive bearings, and (2) locomotive bearing resonance demodulation spectrum analysis technology-----precise diagnosis.
3.1.1 Simple diagnosis is that in the vibration detection, the size of some vibration parameters is generally tested, and the state of the bearing is judged by comparison with the standard value (threshold value). In the simple diagnosis of the locomotive bearing, the main selection time is The KV, and the acceleration effective value parameter Xrms of the domain wave kurtness coefficient parameter are used as the fault judgment parameters. The kurtosis index KV is a dimensionless parameter, which is sensitive to the early fault of the bearing and is not sensitive to the operating conditions. When the working surface of the bearing tooth surface fails, the impact pulse at the working surface defect will be generated every revolution. The larger the fault, the larger the amplitude of the impact response, the faster the KV value rises, but the serious deterioration of the bearing fault later. The KV value is reduced, while the opposite is the opposite. It is not sensitive enough to early failures, but it rises with the bearing failure and the stability is better. In short, it is appropriate to obtain a reliable diagnosis of simple diagnosis, and it is appropriate to apply the kurtosis coefficient and the comparison value. It takes into account the sensitivity and stability of the diagnostic parameters.
Traction motor bearing ground detection threshold
Dongfeng 4 locomotive axlebox bearing top wheel detection threshold
Simple diagnosis has a better side, but there are also many shortcomings. When the bearing is easily diagnosed, the following situations will occur; (1) The bearing parameter value is faulty, but the fault location cannot be determined. (2 The bearing itself is faultless but due to failures during assembly (such as rotor dynamic imbalance, shaft misalignment, etc.). (3) When the sensor is placed in the machine base, the measured fault signal is strong and weak (the inner ring is the smallest, the roller is strong, and the outer ring is large), so the error is judged according to the parameters. Through the analysis of the above three cases, the simple diagnosis does not determine the method of the bearing component failure, and can not solve the problem of "depending on the situation". Therefore, it is not enough to judge the presence or absence of the fault by simple diagnosis. It is necessary to carry out precise diagnosis.
3.1.2 Precision diagnosis
When the bearing surface is damaged, such as fatigue peeling, local wear, surface corrosion, etc., the bearing and the rolling and rotating rotation will take turns to crush these surface damages. The high-speed inner ring rotation will cause the crushing to impact. The longitudinal wave caused by the impact is emitted outward at the speed of sound before the material has been deformed. It has a steep frontal waveform and an extremely rich spectrum. With the internal damping of the material, the longitudinal wave of the shock is sharply attenuated, so that the sensor receives the impact force. Pulse signal. The impact pulse waveform is approximately rectangular, the fault frequency of the rectangular pulse is extremely wide, and the natural frequency of the bearing system and various types of sensors is very low, so the natural frequency of the bearing system and various types of communication chamber sensors are affected by the fault pulse frequency. Covered, the fault frequency provokes the resonance of the bearing system and various types of sensors, and the use of a passband filter to filter out unwanted low- and high-frequency noise (such as various types of noise such as mechanical vibration) only causes the fault shock pulse to provoke the bearing outer ring or The sensor resonance response waveform passes, and the low frequency fault shock signal is amplified and increased to a higher frequency response attenuation oscillation of the vibration system, and the envelope demodulation amplifies the higher frequency response into an expanded low frequency signal, after frequency analysis. The instrument is transformed into a low spectrum, which is the demodulation spectrum.
4 locomotive bearing diagnosis case
The following case is the fault case detected by the fourth left axlebox bearing of Dongfeng 7D0034 locomotive on January 27, 2002. The bearing type is 752732. Tested using a JL-501 bearing test stand with a speed of 500 rpm.
The diagnostic parameters detected by the time domain waveform have a kurtosis coefficient of 6.63, and the high kurtosis is also seriously exceeded, so the test result is unqualified. At the same time, the peak of the demodulation spectrum is 60HZ, and the peak interval of 1, 2, and 3 is 60HZ. The frequency of the
bearing fault is calculated according to the locomotive bearing fault frequency table: 500×0.12=60HZ From the table, the faulty part is found to be the outer ring of the bearing. In the subsequent disassembled bearing, the outer ring raceway is found to be somewhat eroded and crushed.
5 Conclusion
The locomotive rolling bearing diagnostic method described in this paper has been carried out in the past two years. I have two sets of equipment: JL-201 locomotive bearing diagnostic instrument and JL-501 locomotive bearing diagnostic table. JL-201 is mainly used for traction motor bearing ground detection and locomotive bearing top wheel inspection. JL-501 is mainly used for dynamic detection of rolling bearings of medium repair locomotives. In the past two years, 2,318 sets of bearings have been tested for other sections and this section of the locomotive. A total of 189 sets of unqualified bearings have been detected. Among them, from 2001 to the present, there has been no bearing failure or machine breakage caused by bearing failure. It has reached a significant fruit.
