
2017年第4期
文章编号:10094687(2017)04000705
车辆与动力技术 Vehicle & Power Technology
总第148期
一种涡轮增压器离心式压气机喘振自动识别算法
严满,张志强,闫学明(北京理工大学机械与车辆学院,北京100081)
摘要:喘报的早期识别对于及时采取措施避免压气机喘振或喘振控制具有重要的意义.在压气机出口处安装动态压力传感器,获取了两种型号的涡轮增压器压气机运行时的实时压力信号。对喘振信号特征进行了分析,采用匹配滤波器对数据进行预处理,从而提高了采集信号的信噪比(SNR).在此基础上设计了一种基于信号脉冲计数法的喘振自动识别算法。Matlab仿真试验表明:该算法能够及时准确地识别出涡轮增压器压气机的喘报,
关键词:增压器;喘振;自动识别;匹配滤波;脉冲计数
中图分类号:TK474.8
文献标识码:A
AutomaticRecognitionAlgorithmofSurgeinTurbocharger
CentrifugalCompressor
YANXiao,
ZHANG Zhi-qiang,
YAN Xue-ming
(School of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081, China)
Abstract:Early recognition is of great significance to take measures in time to avoid compressor surge or
to control it.The dynamic presst
re installed at their outlets of the compressors and the real-
time pressure signals of the two types of the turbochargers were obtained during the compressor in
operation. The signal features of the s
analyzed, the measured data were pretreated by using the
ratio (SNR)was improved.An automatic
matched filtering method, and then the signal-to-noise
recognition algorithm of surge was designed based on the signal pulse counting method. Simulation test in
Matlab demonstrated that this algorithm can identify the cor accurately
e
or surge of the turbocharger in time and
Key words: turbocharger; surge; automatic recognition; matched filtering; pulse counting
车用离心式涡轮增压器压气机的喘振对于其稳定运行有不利的影响[1],快速识别喘振对于确保涡轮增压器的安全运行具有重要的意义,对于涡轮增压器的喘振识别,有很多学者进行了研究[26] 对压气机喘振进行有效地识别,主要方法就是利用高精度和高灵敏度的传感器对压气机近喘振状态的相关物理量进行有效测量.检测算法主要有紊流度分析法、脉动压力方差法、压力变化速率法和相关积分法等
对文中涉及的离心压气机,孙涛等利用相关积
分法对离心压气机喘振过程进行非线性监测.以实验分析论证了相关积分值等三个监测参数在离心压气机运行工况监测、喘振强度辨识和压力信号非平稳性度量中的有效性[}],贺云南等利用振动信号成功诊断了废气涡轮增压器喘振故障[8]朱智富等采用动态测量采集系统获得了喘振过程中的多处动态压力信号,采用滑动窗口法建立压气机的喘振判断标准。得到的结果比根据经验判断获得的的临近喘振点流量波动范围明显减小(9),曹跌劫等使用了声音信号结合感知器神经网络作为判断嘴振的
收稿日期:2017-0414
作者简介:严滞(1991-),男,硕士,研究方向为叶轮机械与电子控制万方数据