多传感器数据融合技术在气门动态特性试验中的应用  

Application of Multi-Sensor Data Integration Technique to the Experiment of Valve Dynamic Characteristic

作　　者：刘忠民[1] 俞小莉[1] 李建锋[1] 沈瑜铭[1] 齐放[1] 

机构地区：[1]浙江大学机械与能源工程学院,浙江杭州310027

出　　处：《内燃机学报》2005年第3期264-268,共5页Transactions of Csice

摘　　要：压电式加速度传感器的漂移问题在气门动态特性试验中难以避免。气门加速度信号的微小漂移,将导致积分的速度信号严重失真。试验采用压电式加速度传感器和电涡流位移传感器测量气门动态特性,并将多传感器数据融合技术应用于气门动态试验数据处理中。由加速度信号积分和位移信号微分获得速度信号,分析两种传感器不同的频响特性,采用小波分解、融合并重构,提取相对准确的气门速度信号。It is unavoidable for the shifting problem of piezoelectric acceleration sensor in the experiment of valve dynamic characteristic. Small shifting of acceleration signal will result in the abnormal alteration of valve velocity. In this paper, the Multi-sensor data integration technique was applied in the experiment of valve dynamic characteristic where the piezoelectric acceleration sensor and eddy current displacement sensor were used, and valve velocity signals were acquired by acceleration integral and displacement differential. From the analysis of the difference between frequency response characteristics of the two sensors, the accurate valve velocity signals were obtained by signal decomposition, integration and reconstruction by the wavelet.

关 键 词：多传感器 数据融合 气门 小波分解与重构 

分 类 号：TK413.4[动力工程及工程热物理—动力机械及工程]