计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (25): 78-80.DOI: 10.3778/j.issn.1002-8331.2008.25.024

• 研发、设计、测试 • 上一篇    下一篇

神经网络在小管径超声波流量测试中的应用

陈 平,王 俏,耿庆军   

  1. 山东理工大学 计算机科学与技术学院,山东 淄博 250049
  • 收稿日期:2007-10-09 修回日期:2008-03-31 出版日期:2008-09-01 发布日期:2008-09-01
  • 通讯作者: 陈 平

Application of ultrasonic flow measurement for small diameter pipe based on Neural Network

CHEN Ping,WANG Qiao,GENG Qing-jun   

  1. School of Computer Science and Technology,Shandong University of Technology,Zibo,Shandong 255049,China
  • Received:2007-10-09 Revised:2008-03-31 Online:2008-09-01 Published:2008-09-01
  • Contact: CHEN Ping

摘要: 介绍了一种基于神经网络的改进型超声波流量测量系统的研制,该系统针对时差法超声波流量计在小管径应用上时差难以测量的不足,采用动量BP算法,实现了对各种非线性影响因素的补偿,通过参数修正有效地提高了系统的测量精度。与传统时差法超声波流量计相比,该系统稳定性好,精度高,具有良好的应用价值和推广价值。

关键词: 超声波流量, 神经网络, 动量BP, 小管径

Abstract: Introduce that a design of ameliorate ultrasonic flow measurement system.In order to overcome the insufficiency of the ultrasonic flow meter for small diameter pipe in travel time,an momentum BP arithmetic is used to realize the compensate of variety of non-linearity influence factors.And the accuracy of measurement has been improved by parameter amending.Compared with the traditional ultrasonic flow meter,the system has well stability,precision and practicality.

Key words: ultrasonic flow, neural network, momentum BP, small diameter pipe