论文部分内容阅读
针对转子系统的裂纹-碰摩耦合故障,提出了一种基于融合信息熵距的转子振动故障诊断方法.利用转子实验台模拟转子系统裂纹故障、碰摩故障及裂纹-碰摩耦合故障,并采集三种典型故障的振动加速度信号.利用时域的奇异谱熵、频域的功率谱熵、时-频域的小波能谱熵以及小波空间特征谱熵,计算融合信息熵距实现故障诊断.实例研究表明:这四种信息熵形成了综合评价转子振动状态的特征指标,多测点、多转速下的信息熵距曲线较好地区分了单一故障和耦合故障,有效地提高了转子振动故障诊断的准确性.测试信号与其对应的振动故障之间的信息熵距最小,信息熵距曲线位于坐标轴的最下方,达到了诊断单一故障和耦合故障的目的.
Aiming at rotor-system coupling failure of crack-rubbing, a rotor vibration fault diagnosis method based on fused information entropy distance was proposed. The rotor system was used to simulate the coupling failure of rotor system with crack fault, rub-impact fault and rub- Three Typical Fault Vibration Acceleration Signals Based on singular spectral entropy in time domain, power spectrum entropy in frequency domain, wavelet energy spectrum entropy in time-frequency domain and spectrum entropy of wavelet feature spectrum, fusion information entropy distance is calculated to realize fault diagnosis. The results show that these four kinds of information entropy form the characteristic index for comprehensively evaluating the vibration state of the rotor. The information entropy curves at multi-measuring points and multi-speed can distinguish the single fault and the coupling fault well, effectively improving the rotor vibration fault diagnosis The information entropy distance between the test signal and its corresponding vibration fault is the smallest and the information entropy distance curve is located at the bottom of the coordinate axis so as to achieve the purpose of diagnosing single fault and coupling fault.