基于多项式预测滤波理论的虚拟传感器构建  

Construction of virtual sensor based on polynomial predictive filter

作　　者：董劲男[1] 秦贵和[1] 张晋东[1] 崔玥[1] 范铁虎[1] 

机构地区：[1]吉林大学计算机科学与技术学院,长春130012

出　　处：《仪器仪表学报》2008年第7期1408-1413,共6页Chinese Journal of Scientific Instrument

基　　金：国家发改委下一代互联网示范工程2005年研究开发产业化及应用试验项目(CNGI-04-1-2D)资助

摘　　要：本文提出了一种虚拟传感器的构造方法,通过预测的方法改善由于网络不确定性造成的传感器信号采样延时和丢失问题。采用多项式预测滤波理论构建虚拟的本地传感器,将网络引入的传输延时和错误对传感器信号造成的影响转化为虚拟传感器的输出误差。将传感器信号的传输与使用分离,使用户不必考虑复杂的网络特性,而只需关注虚拟传感器的输出误差。本方法不仅保证了传感器信号网络传输的实时性和可靠性,也使网络控制系统的分析与设计更加方便。The construction method of virtual sensor is presented in this paper, and the bad consequences caused b transmission uncertainties are alleviated by prediction. The virtual sensor is constructed by using polynomial predic tive filtering （PPF） theory, and the transmission uncertainties are transformed into the uncertainties of system mode consequently, Based on this, the errors caused by the transmission uncertainties and prediction are processed togeth. er as the output errors of the virtual sensor. The transmission and usage of signals are separated, and the user cal make use of these signals without considering the complex network characteristics and focus on the output errors on. ly. This method not only ensures the real-time performance and reliability of sensor signal transmission, but also makes the design and analysis of networked control system more convenient.

关 键 词：虚拟仪器 多项式预测滤波 传感器信号 网络不确定性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]