| 张刚,侯家琛,黄笑笑,曹子涵.改进四稳态随机共振系统与MOMEDA结合的轴承故障诊断研究[J].电子测量与仪器学报,2025,39(11):119-132 |
| 改进四稳态随机共振系统与MOMEDA结合的轴承故障诊断研究 |
| Studies on improving a quad-stable stochastic resonance system’s bearingfault diagnosis combined with MOMEDA |
| |
| DOI： |
| 中文关键词: 随机共振 多稳态系统 故障特征提取 MOMEDA-PUQSR 滚动轴承 量子遗传算法 |
| 英文关键词:stochastic resonance multi-stable system fault feature extraction MOMEDA-PUQSR rolling bearing quantum genetic algorithm |
| 基金项目:重庆市自然科学基金面上（CSTB2024NSCQ MSX1060, CSTB2023NSCQ-MSX0235, CSTC2021JCYJ-MSXMX0836）项目资助 |
|
|
| Author | Institution |
| Zhang Gang | School of Communications and Information Engineering, Chongqing University of Posts and
Telecommunications, Chongqing 400065, China |
| Hou Jiachen | School of Communications and Information Engineering, Chongqing University of Posts and
Telecommunications, Chongqing 400065, China |
| Huang Xiaoxiao | School of Communications and Information Engineering, Chongqing University of Posts and
Telecommunications, Chongqing 400065, China |
| Cao Zihan | School of Communications and Information Engineering, Chongqing University of Posts and
Telecommunications, Chongqing 400065, China |
|
| 摘要点击次数: 461 |
| 全文下载次数: 358 |
| 中文摘要: |
| 致力于解决经典四稳态随机共振系统(CQSR)的输出饱和性问题，构造了一种新型分段非饱和的四稳态随机共振系统(PUQSR)。首先，通过实验信号的模拟仿真验证了PUQSR的抗饱和特性。随后，研究了PUQSR系统的势函数结构变化，根据绝热近似理论，理论推导出PUQSR系统的稳态概率密度(SPD)和功率谱放大系数(SA)，并详细分析了系统参数对其的影响。进一步地，将信噪比增益(SNRI)和SA作为衡量系统性能的指标，数值模拟仿真验证PUQSR系统在放大信号和对噪声能量的转化效率都更优越。同时，为了在强噪声背景下更有效的提取目标信号，联合多点最优最小熵解卷积(MOMEDA)方法与随机共振（SR）系统，提出一种增幅MOMEDA-PUQSR系统。最后，通过自相关谱法和量子遗传算法(QGA)寻找MOMEDA-PUQSR系统的最优参数，并应用于实际故障信号。实验结果表明，增幅后的故障信号包络表现出更明显的脉冲特性，相比于原始信号，信噪比（SNR）提升了15.404 2~26.077 8 dB。同时，相比于MOMEDA-CQSR系统，增幅信号通过MOMEDA-PUQSR系统的输出SNR提升了0.281 5~1.406 3 dB，谱峰值提升了480.144~4 314.187 3。 |
| 英文摘要: |
| Committed to solving the output saturation problem of the classical quad-stable stochastic resonance (CQSR) system, a new type of piecewise unsaturated quad-stable stochastic resonance (PUQSR) system is constructed. Firstly, the anti-saturation characteristic of PUQSR is verified by simulation of experimental signals. Then, the potential function structure variation of the PUQSR system is studied. According to adiabatic approximation theory, the steady-state probability density (SPD) and power spectrum amplification (SA) coefficient of the PUQSR system are deduced theoretically, and the influence of system parameters on them is analyzed in detail. Further, the signal-to-noise ratio improvement (SNRI) and SA are used as indicators to measure system performance, and numerical simulation verifies that the PUQSR system is better at amplifying signals and converting noise energy. At the same time, to extract the target signal more effectively in the context of strong noise, a new MOMEDA-PUQSR system is proposed by combining the multi-point optimal minimum entropy deconvolution (MOMEDA) method and the SR system. Finally, the optimal parameters of the MOMEDA-PUQSR system are found through the autocorrelation function and quantum genetic algorithm and successfully applied to the actual fault signal. The experimental results show that the increased fault signal envelope exhibits more obvious pulse characteristics, and the SNR is increased by 15.404 2~26.077 8 dB compared to the original signal. At the same time, compared with the MOMEDA-CQSR system, the output SNR of the increased signal through the MOMEDA-PUQSR system has been increased by 0.281 5~1.406 3 dB, and the spectral peak has been increased by 480.144~4 314.187 3. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
