Data Mining Research in Early Warning Model of Chlorine Gas Monitoring Wireless Sensor Network

Massive historical data are stored in chlorine gas monitoring network. So that prediction algorithm of data mining is used to dig historical data not only can make the redundant data reused, but also can forecast the network trend and improve the network early warning model. The chlorine gas monitor...

Full description

Saved in:
Bibliographic Details
Published in2014 Fourth International Conference on Instrumentation and Measurement, Computer, Communication and Control pp. 711 - 715
Main Authors Wang Rongxin, Xiu Debin, Zhou Yushan, Liu Congning, Shi Yunbo
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2014
Subjects
Data mining
Data models
Fletcher-Reeves algorithm
Monitoring
Neural networks
Prediction algorithms
Training
wireless sensor network
Wireless sensor networks
Online AccessGet full text

Cover

More Information
Summary:Massive historical data are stored in chlorine gas monitoring network. So that prediction algorithm of data mining is used to dig historical data not only can make the redundant data reused, but also can forecast the network trend and improve the network early warning model. The chlorine gas monitoring wireless sensor network based on ZigBee was designed in this paper. Then Fletcher-Reeves algorithm was added to dig historical data in the network, forecast the network trend and improve the early warning model. The predicted concentration of chlorine data were trained by data mining model. The maximum relative error between predicted concentration and measured concentration was 11.08%, and the maximum average error was 7.36%. And it can satisfy actual requirements.
DOI:10.1109/IMCCC.2014.151