首页 | 期刊介绍 | 编委会 | 道德声明 | 投稿指南 | 常用下载 | 过刊浏览 | In English
引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  下载PDF阅读器  关闭
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 24次   下载 0  
分享到: 微信 更多
基于LS-RBF神经网络温度补偿的溶解氧检测系统设计
綦声波1, 王榕2, 尹保安1, 张阳1
1.中国海洋大学;2.中国海洋大学 (联系方式 17865326597 邮箱 wangrongrong312@163.com)
摘要:
在溶解氧检测过程中,为克服温度对溶解氧传感器的影响,本文对极谱型溶解氧检测系统的溶解氧电极激励源、高精度信号采样、软件标定和温度补偿等方面进行研究。通过对极谱型溶解氧传感器工作原理进行分析,设计了极谱型溶解氧传感器检测电路;根据溶解氧电极的温度特性,设计了基于NTC热敏电阻的硬件温度补偿电路,并利用最小二乘法及RBF神经网络构建了软件温度补偿模型;同时针对溶解氧检测系统在实际应用中所面临的复杂问题,进一步分析并验证采用BP神经网络的优势。利用饱和蒸馏水进行温度补偿实验,结果表明:经温度补偿后,该溶解氧检测系统的相对误差及采样波动均在1%以内,大大减小了传感器的非线性误差,检测精度可满足应用要求。
关键词:  溶解氧  电极激励源  温度补偿  RBF神经网络  软件标定
DOI:10.11759/hykx20191016003
分类号:
基金项目:中央高校基本科研业务费专项(201964014)
Design of dissolved oxygen detection system based on LS-RBF neural network temperature compensationrk
Qi Shengbo,Wang Rong,Yin Baoan,Zhang Yang
Ocean University of China
Abstract:
In the process of dissolved oxygen detection, in order to overcome the influence of temperature on dissolved oxygen sensor, the polarographic dissolved oxygen detection system is studied in this paperabout the dissolved oxygen electrode excitation source, high-precision signal sampling, software calibration and temperature compensation. By analyzing the working principle of the polarographic dissolved oxygen sensor, a polarographic dissolved oxygen sensor detection circuit is designed. According to the temperature characteristics of dissolved oxygen electrode, a hardware temperature compensation circuit based on NTC thermistor is designed. The software temperature compensation model is established by least squares method and RBF neural network. Temperature compensation experiments are carried out using saturated distilled water. Meanwhile dissolved oxygen detection system for complex problems in practical applications faced further analyze and verify the advantages of using BP neural network. The experimental results show that the relative error and sampling fluctuation of the dissolved oxygen detection system are within 1% after temperature compensation, which greatly reduces the nonlinear error of the sensor. The measurement accuracy can meet the application requirements.
Key words:  dissolved oxygen  electrode excitation source  temperature compensation  RBF neural network  software calibration
版权所有 《海洋科学》编辑部 Copyright©2008 All Rights Reserved
主管单位:中国科学院 主办单位:中国科学院海洋研究所
地址:青岛市南海路七号  邮编:266071  电话:0532-82898755  E-mail:marinesciences@qdio.ac.cn
技术支持:北京勤云科技发展有限公司