DocumentCode
3281227
Title
Research on data fusion technology for integrated composite sensor for mining
Author
Jie, Yang
Author_Institution
Inst. of Electr. & Control Eng., Xi´´an Univ. of Sci. & Technol., Xi´´an, China
fYear
2011
fDate
15-17 April 2011
Firstpage
1623
Lastpage
1626
Abstract
The integrated composite sensor for mining integrates the primary switching circuit or links for detecting the safe environmental parameters of mine and selects a proper microcontroller to realize the time sharing detection of different environmental parameters. The parallel secondary data fusion algorithm is applied. After considering the time and space, the data fusion method combining weighted arithmetic mean and recursive estimation is applied for the single sensor. BP network data fusion algorithm is applied for different sensors of different types at the same detection position. In addition, relevance fusion algorithm is applied to judge the reliability of sensor output and improve the detection accuracy of the integrated composite sensor. Data simulation is performed for data fusion algorithm through tests, which verify the accuracy and feasibility of data fusion algorithm and are of great value for reference.
Keywords
backpropagation; computerised instrumentation; microcontrollers; mining industry; recursive estimation; reliability; sensor fusion; BP network data fusion algorithm; data simulation; integrated composite sensor; microcontroller; mining; primary switching circuit; recursive estimation; relevance fusion algorithm; secondary data fusion algorithm; sensor reliability; time sharing detection; weighted arithmetic mean; Decision support systems; BP network data fusion algorithm; data fusion; integrated composite sensor; the data fusion method combining weighted arithmetic mean and recursive estimation;
fLanguage
English
Publisher
ieee
Conference_Titel
Electric Information and Control Engineering (ICEICE), 2011 International Conference on
Conference_Location
Wuhan
Print_ISBN
978-1-4244-8036-4
Type
conf
DOI
10.1109/ICEICE.2011.5777658
Filename
5777658
Link To Document