基于位域运动误差观测器的齿轮断齿故障诊断

Due to high speed, heavy load or other poor working conditions, the gear is often prone to tooth crack, surface damage, or even the tooth broken. As an important component of transmission mechanism, the fault of gear has a great impact on the safe operation of the entire system. Therefore, it is necessary to monitor the gear condition and diagnosis of the gear fault. Compared with the traditional vibration method and noise method, the gear fault diagnosis based on the motor drive system needs no additional sensors, becoming the latest research hotspot. For the frequent acceleration and deceleration in practical conditions such as servo system, this paper presents a new gear fault diagnostic method in spatial domain. A cascaded observer was designed firstly that can obtain the kinematic error of the gear transmission system without sensors on the load-side. Then using the spatial domain resampling technique, the fault daignosis no longer depended on the speed operating condition. Furthermore, the signal to noise ratio (SNR) of gear fault was effectively improved with the spatial domain synchronous averaging method. Finally, the algorithm was validated on the gear tooth-broken fault diagnosis platform. The results show that this method can implement the gear fault diagnosis in a large range of variable speed situation.