摘要
航空发动机静电监测技术表现出了较高的故障预警能力,但原始静电信号常包含较多噪声,为提高故障信息提取的准确性,必须对静电信号进行降噪处理。本研究首先介绍了静电监测技术的原理,分析了信号的噪声的来源和主要构成;针对静电信号耦合噪声滤除问题,引入了信号稀疏表达和经验模态分解理论,研究了模态分量的筛选依据和相关准则,并提出了一种基于模态分量优化重构和稀疏表达的联合降噪算法和具体流程;利用所提方法对涡扇发动机试车实验中采集的实际静电信号进行了降噪效果验证,并与其它方法进行了对比。结果表明本文方法在滤除随机噪声以及工频干扰的同时能更高程度的保留有用异常颗粒信号,稀疏迭代次数在设置为20~50时均能够较好提取异常信号。
The aero-engine electrostatic monitoring technology shows a high fault warning capability.However,the raw electrostatic signal usually contains unexcepted noise.To improve the accuracy of the information extraction,the electrostatic signal should be denoised.This study firstly introduces the principle of electrostatic monitoring technology and analyzes the sources and principal components of the noise.To address the problem of coupling noise filtering of electrostatic signal,the theories of sparse signal representation and empirical mode decomposition are introduced.The basis and criteria for mode functions selection are investigated,and a joint denoising algorithm and process based on the mode function optimized reconstruction and sparse representation are proposed.The actual electrostatic signal acquired in the turbofan engine test run is used to evaluate the denoising effectiveness of the proposed algorithm.Meanwhile,it is compared with other classic methods.The results show that the proposed method can effectively remove the random noise and power frequency interference while remaining the useful particle signal to a large degree.The abnormal signal can be extracted well when the number of sparse iterations is set from 20 to 50.
作者
殷逸冰
文振华
Yin Yibing;Wen Zhenhua(School of Mechanical and Automotive,Qingdao University of Technology,Qingdao 266520,China;School of Aeroengine,Zhengzhou University of Aeronautics,Zhengzhou 450015,China)
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2022年第2期196-204,共9页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(51975539)
航空科学基金(2018ZD55008)项目资助
关键词
航空发动机
静电监测
稀疏表达
信号处理
传感器
aero-engine
electrostatics monitoring
sparse representation
singal processing
sensor
