摘要
为削减扩展工况传递路径分析在信号采集过程中干扰噪声的影响,提高分析精度,提出了一种结合自适应变分模态分解和巴氏距离的优化OPAX方法。该方法考虑到多尺度模糊熵能够较好地表征非稳态复杂信号,将其作为适应度函数。采用模拟退火粒子群算法和粒子群算法分别对信号进行自适应变分模态分解,并通过模拟信号证明了模拟退火粒子群算法的泛化性和准确性。最后采用巴氏距离对分解信号进行筛选,以区分相关模态和非相关模态,将相关模态进行重组实现OPAX采集信号的有效去噪。在确定去噪算法的准确性后,以某轻型客车为例,建立了振动测点到驾驶员右耳的传递路径分析模型,在90 km/h匀速行驶工况下采集整车主要结构点悬置主被动端的加速度信号,并对采集信号进行了降噪处理。以驾驶员右耳声压的测试值为参考,对比了OPAX优化前后计算结果。结果表明:优化后OPAX计算值的信噪比提高了71.2%,均方根误差减小了66.9%,在峰值频率处的误差均控制在5%以内;该方法能有效滤除OPAX在信号采集过程中的干扰噪声,有效保留信号的完整性,提高OPAX方法的分析精度。
In order to reduce the influence of the interference noise of OPAX in signal acquisition process and improve the analysis accuracy,an optimized OPAX method combining adaptive variational mode decomposition and Bhattacharyya distance is proposed.This method takes into account that the multi-scale fuzzy entropy can better characterize the unsteady complex signal,and uses it as the fitness function.The adaptive variational mode decomposition of the signals are performed by using the simulated annealing particle swarm algorithm and the particle swarm algorithm respectively,and the generalization and accuracy of the simulated annealing particle swarm algorithm are proved by simulated signals.Finally,the decomposed signals are screened by the Bhattacharyya distance to distinguish the related modes from the non-correlated modes,and the related modes are reorganized to realize the effective denoising of the OPAX acquisition signals.After determining the accuracy of the denoising algorithm,taking a light passenger car for example,the analysis model of the transmission path from vibration measuring point to driver’s right ear is established,the acceleration signals of the suspend active and passive ends of the main structural points of the vehicle are collected at 90 km/h constant speed,and the collected signals are denoised.Referencing the test value of the sound pressure on the driver’s right ear,the calculation results before and after OPAX optimization are compared.The result shows that(1)the signal-to-noise ratio of the OPAX calculated value after optimization is improved by 71.2%,the root mean square error is reduced by 66.9%,and the error at the peak frequency is controlled within 5%;(2)this method can effectively filter out the interference noise of OPAX in the signal acquisition process,effectively retain the integrity of the signal,and improve the analysis accuracy of the OPAX method.
作者
张俊红
贾宏杰
周启迪
朱小龙
林杰威
ZHANG Jun-hong;JIA Hong-jie;ZHOU Qi-di;ZHU Xiao-long;LIN Jie-wei(State Key Laboratory of Engine Combustion,Tianjin University,Tianjin 300072,China;Tianjin Renai College,Tianjin 301636,China;Weichai Power Co.,Ltd.,Weifang Shandong 261061,China)
出处
《公路交通科技》
CAS
CSCD
北大核心
2021年第7期138-144,158,共8页
Journal of Highway and Transportation Research and Development
基金
国家自然科学基金项目(51705357)
天津市自然科学基金项目(18JCYBJC20000)。
关键词
汽车工程
扩展工况传递路径分析
多尺度模糊熵
自适应变分模态分解
巴氏距离
去噪
automobile engineering
Operational-X Transfer Path Analysis(OPAX)
multi-scale fuzzy entropy
adaptive variational mode decomposition
Bhattacharyya distance(BD)
denoising
