摘要
针对故障特征信号在诊断时经常淹没在噪声中难以提取问题,利用Woods-Saxon型单稳态模型与混合型双稳态模型结合提出一种混合型三稳态随机共振系统,该系统不仅保留了Woods-Saxon对故障信号易于检测的优点又利用了三稳态对噪声利用率高的特点。利用信噪比增益为衡量指标,提出寻找最优系统参数的自适应算法;对α噪声背景下的谐波振动信号、调幅信号、周期脉冲衰减信号进行检测;提出一种变分模态分解与混合三稳态结合的信号检测方法,并将其应用于实际轴承故障检测;经过仿真实验表明混合型三稳态随机共振模型以及组合模型在故障信号的检测中检测结果清晰可靠、性能优越。
Aiming at the problem that fault feature signals are often submerged in noise so that difficult to extract,a hybrid tri-stable stochastic resonance system,combining the Woods-Saxon monostable model with the hybrid bistable model was proposed.The system not only retains the advantages of Woods-Saxon’s easiness to detect fault signals but also takes advantage of the high noise utilization characteristic of the tri-stable model.First,the signal-to-noise gain was used as a measure to create an adaptive algorithm for finding the optimal system parameters.Then,harmonic vibration signals,amplitude modulation signals and periodic pulse attenuation signals were detected under the background ofαnoise.Finally,a signal detection approach using the variational mode decomposition combined with the hybrid tri-stable model was put forward and applied to the actual bearing faults detection.The simulation results show that the hybrid tri-stable stochastic resonance model and the composed model can achieve clear and reliable test results and superior performances in fault signal detection.
作者
张刚
李红威
ZHANG Gang;LI Hongwei(School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Chongqing Key Laboratory of Signal and Information Processing,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2019年第18期9-17,共9页
Journal of Vibration and Shock
基金
国家自然科学基金(61771085)
