基于BP神经网络的超声流量计声道流速修正方法研究

Instrumentation and Equipments 仪器与设备, 2019, 7(3), 186-192

Published Online September 2019 in Hans. http://www.hanspub.org/journal/iae https://doi.org/10.12677/iae.2019.73025

Research on Correction Method

of Sound Channel Velocity of Ultrasound Flowmeter Based on BP Neural Network

Jingjing Li, Jianmin Wang, Xiaoyu Wu, Xiwen Yang, Chen Li, Qi Zhou, Zijie Teng, Qufei Shi*

Beijing Institute of Metrology, Beijing

Received: Aug. 30, 2019; accepted: Sep. 20, 2019; published: Sep. 27, 2019

ththth

Abstract

The flow velocity measured directly by the ultrasonic flowmeter is the line average velocity along the ultrasonic propagation path, and it is different from the average flow velocity of the pipeline section. In order to obtain the exact value of flow, the measured flow velocity must be corrected. In this paper, a velocity correction model of multi-channel based on BP neural network is established. According to the experiment of ultrasonic flowmeter, sufficient experimental data are obtained as data samples to train and valid the model. Through the validation of the model after training, it is found that estab-lished model can accurately correct the flow velocity in the channel of the ultrasonic flowmeter.

Keywords

Flow Measurement, Ultrasonic Flowmeter, Velocity Correction, BP Neural Network

基于BP神经网络的超声流量计声道流速修正 方法研究

李晶晶，王建民，吴晓昱，杨希文，李 晨，周 齐，滕梓洁，史去非*

北京市计量检测科学研究院，北京

收稿日期：2019年8月30日；录用日期：2019年9月20日；发布日期：2019年9月27日

摘 要

超声波流量计直接测量得到的流速为超声波传播路径上的线平均流速，它与管道截面平均流速不同，为了获得流量的准确值，必须对测量得到的流速进行修正。本文建立了基于BP神经网络的多声道线平均流

*

通讯作者。

文章引用: 李晶晶, 王建民, 吴晓昱, 杨希文, 李晨, 周齐, 滕梓洁, 史去非. 基于BP神经网络的超声流量计声道流速修正方法研究[J]. 仪器与设备, 2019, 7(3): 186-192. DOI: 10.12677/iae.2019.73025