矿井突水监测科氏力质量流量计的设计  

Design of Coriolis mass flowmeter for monitoring water inrush inmine

作　　者：童敏明[1] 秦海鹏[1] 

机构地区：[1]中国矿业大学信息与电气工程电学院,徐州221008

出　　处：《仪器仪表学报》2009年第8期1613-1617,共5页Chinese Journal of Scientific Instrument

基　　金：国家重点基础研究发展规划项目973计划(2007CB209407)资助项目

摘　　要：介绍了一种实时在线监测矿井底板突水的科氏力质量流量计。它以FPGA,STM32F103为核心,采用基于神经网络的信号处理算法进行信号处理,可以精确分析水流量质量信号的频率和相位差,消除了检测电路的零飘及干扰信号的影响。实验数据表明,通过样本训练后的流量计在测量模拟矿井突水水源的质量流量时相对误差小于2.2%,密度相对误差小于4.8%。This paper introduces the design of a Coriolis mass flowmeter for the on-line monitoring of water inrush in mine. FPGA and STM32F103 are used in the design. Neural network is used in the signal processing for accurate measurement of water flow. The neural network algorithm could accurately track the signal frequency; calculate the phase difference between two signals, and is insensitive to circuit zero drift and interference. Experiment results show that the relative error of the measuring system is 2.2% for mass flow and 4.8% for water density in detecting the source of water inrush after sample training.

关 键 词：矿井突水 科氏力质量流量计:信号处理 RBF神经网络 

分 类 号：TB971[一般工业技术—计量学]