基于随机共振的齿轮箱早期故障诊断研究

随机共振在微弱故障诊断中相对线性系统具有明显优势，能显著增强信号的信噪比。将随机共振与共振解调相结合来检测齿轮箱的早期裂纹故障，首先通过带通滤波选取共振频带，然后通过Hilbert进行解调，最后采用归一化方法调节随机共振的系统参数得到故障特征输出。对仿真信号和QPZZ Ⅱ齿轮故障系统实测数据的处理结果表明，该方法与传统共振解调相比具有明显优势，为齿轮箱故障诊断提供了一种更为有效的途径。

Research on Early Fault Diagnosis of Gearbox Based on Stochastic Resonance

Stochastic resonance has obvious advantages in weak fault diagnosis for linear system, since it can significantly enhance the signal to noise ratio. In this paper, the stochastic resonance is combined with the resonance demodulation to detect early crack fault of the gearbox. The band pass filter was selected by the resonant frequency band, then the Hilbert enveloping demodulation was used for the filtered signals, and finally the normalized stochastic resonance method by means of adjusting system parameters was applied to get the ideal characteristic output for the gearbox faults. Compared with the traditional resonance demodulation, simulated signals and QPZZ II Gear fault system measured data were processed and analyzed, and the results show that the stochastic resonance method for the gearbox fault diagnosis is effectual for the weak signal detection and it provides a novel way for the default diagnosis for the rotational machines.