Open this publication in new window or tab >>2006 (English)Doctoral thesis, comprehensive summary (Other scientific)
Abstract [en]
An increasing number of power plants in Scandinavia are beginning to use biofuel instead of coal or oil. The material in the new fuel is a mixture of woodchips, mostly Pine, Spruce and Salix, bark, GROT (tops and branches from felling waste) and sawdust from sawmills. It is heterogeneous, having a moisture content varying from 15% up to 65%. The moisture content affects the combustion of the fuel and therefore its commercial value. The industry is now interested in obtaining a method for measuring the moisture content of biofuel, quickly and reliably; preferably on delivery at the power plant.
The measuring technique presented in this thesis is the first reported in the literature capable of measuring the moisture content of a large sample of such an heterogeneous material as biofuel. The equipment is today calibrated for a sample volume of 0.1 m3. A radio frequent signal is supplied from an antenna and penetrates the biofuel. Its reflection is modeled using partial least squares.
As part of the work presented in this thesis, a new type of measuring rig and an analysis method for measurement of the moisture content of large samples of heterogeneous material have been developed. A statistical model for moisture content measurements of five different biofuel materials using radio waves has been built, having a root mean square error of prediction of 2.7. The interactions between biofuels and radio frequent signals have been demonstrated, indicating a variation of the reflection with varying types of biofuel material and variation in the reflection and delay of the signal with varying moisture content.
Place, publisher, year, edition, pages
Institutionen för Samhällsteknik, 2006. p. 147
Series
Mälardalen University Press Dissertations, ISSN 1651-4238 ; 24
Keywords
radiofrequency measurement, moisture content measurement, fuel characterization, biofuel, time domain measurement, Bulk measurement, Multivariate data analysis
Identifiers
urn:nbn:se:mdh:diva-130 (URN)91-85485-08-X (ISBN)
Public defence
2006-01-27, Lambda, 13:15 (English)
Opponent
Supervisors
2008-11-172008-11-17Bibliographically approved