摘要
空化检测对于保障离心泵运行的安全性和可靠性具有重要意义,已有研究侧重于信号采集和特征提取,对于空化诱发的振动噪声形成机理研究不够深入。为了实现离心泵空化状态的准确表征和有效识别,本文建立了基于信号调制理论的流体机械振动噪声信号模型,将流体激振信号和调制信号视为空化表征的有效信息成分,在此基础上提出了一种基于频带能量和峭度的主导频带时频分析方法,并结合卷积神经网络实现空化状态智能识别.最后,仿真信号和实际数据的分析结果验证了流体机械信号模型的合理性,也证明了所提出的主导频带时频分析方法的有效性。
Cavitation detection is important in ensuring the safety and reliability of centrifugal pumps,the existing researches focus on signal acquisition and feature extraction.However,the signal formation mechanism of vibration and noise induced by cavitation hasn’t been investigated fully.In order to realize cavitation state characterization and identification effectively,this study establishes the vibration and noise signal model of fluid machinery based on amplitude-modulated(AM) signal theory.The carrier wave signal induced by hydraulic excitation force and modulation signal are regarded as the active signal components.Thus,this study proposes a dominant frequency bands time-frequency analysis method(DFTF) based on the calculation of energy and kurtosis of frequency bands.Further,cavitation state intelligent identification is realized by combining DFTF with deep convolutional neural network.Finally,the rationality of the proposed signal model and the validity of DFTF are demonstrated by simulation signal and real data.
作者
伍柯霖
钱全
邢允
初宁
武鹏
李诗佯
吴大转
WU Ke-Lin;QIAN Quan;XING Yun;CHU Ning;WU Peng;LI Shi-Yang;WU Da-Zhuan(Institute of Process Equipment.,College of Energy Engineering.Zhejiang University,Hangzhou 310027,China;The 705th Research Institute Kunrning,CSIC,Kunming 650118,China)
出处
《工程热物理学报》
EI
CAS
CSCD
北大核心
2021年第1期106-114,共9页
Journal of Engineering Thermophysics
基金
国家自然科学基金资助项目(No.61701440)
浙江省重点研发计划(No.2019C01147)。
关键词
空化检测
调幅信号
时频分析
智能诊断
cavitation det ection
amplitude-modulated signal
time-frequency analysis
intelligent diagnosis
