Abstract
The scope of the project, entitled ``Determination and Correlation of Coal Data``, was to investigate the combustion reactivity of coals as measured by different laboratory scale methods, and if possible correlate these with basic coal parameters. Through correlation of advanced and more simplistic methods, the project aim was to depict methods that could improve the understanding and prediction of coal char burnout in pulverized coal fired power plant boilers. It has been demonstrated that a valid statistical model can be created from the present span of variables, and within the variance covered by the eight coals included. The study has limited the numbers of variables needed to predict the char burnout and if more coals were included, the number of variables could have been reduced further, without loosing the predictive ability of the models. The statistical analysis has shown interesting characteristics and correlations, especially with respect to the southern hemisphere coals with respect to their inertinite constituents. The study has further shown that a simple classification or understanding of coal composition cannot be applied, and that further research must be conducted in that respect. To validate the present models, it is recommended to perform tests with representative coals actually used
More>>
Citation Formats
Rosenberg, P, and Thomsen, E.
Determination and correlation of coal data. Multivariate analysis.
Denmark: N. p.,
1993.
Web.
Rosenberg, P, & Thomsen, E.
Determination and correlation of coal data. Multivariate analysis.
Denmark.
Rosenberg, P, and Thomsen, E.
1993.
"Determination and correlation of coal data. Multivariate analysis."
Denmark.
@misc{etde_10141221,
title = {Determination and correlation of coal data. Multivariate analysis}
author = {Rosenberg, P, and Thomsen, E}
abstractNote = {The scope of the project, entitled ``Determination and Correlation of Coal Data``, was to investigate the combustion reactivity of coals as measured by different laboratory scale methods, and if possible correlate these with basic coal parameters. Through correlation of advanced and more simplistic methods, the project aim was to depict methods that could improve the understanding and prediction of coal char burnout in pulverized coal fired power plant boilers. It has been demonstrated that a valid statistical model can be created from the present span of variables, and within the variance covered by the eight coals included. The study has limited the numbers of variables needed to predict the char burnout and if more coals were included, the number of variables could have been reduced further, without loosing the predictive ability of the models. The statistical analysis has shown interesting characteristics and correlations, especially with respect to the southern hemisphere coals with respect to their inertinite constituents. The study has further shown that a simple classification or understanding of coal composition cannot be applied, and that further research must be conducted in that respect. To validate the present models, it is recommended to perform tests with representative coals actually used in the power plants. It is recommended that the present limited database is extended in order to cover a wider range of coal types used in Danish power plants. This would ensure that more universally applicable and statistical valid methods would be made available to the power plants, in order to provide simple tools for rapid and inexpensive prediction of the burning behavior of the various coals. (EG) (10 refs.)}
place = {Denmark}
year = {1993}
month = {Aug}
}
title = {Determination and correlation of coal data. Multivariate analysis}
author = {Rosenberg, P, and Thomsen, E}
abstractNote = {The scope of the project, entitled ``Determination and Correlation of Coal Data``, was to investigate the combustion reactivity of coals as measured by different laboratory scale methods, and if possible correlate these with basic coal parameters. Through correlation of advanced and more simplistic methods, the project aim was to depict methods that could improve the understanding and prediction of coal char burnout in pulverized coal fired power plant boilers. It has been demonstrated that a valid statistical model can be created from the present span of variables, and within the variance covered by the eight coals included. The study has limited the numbers of variables needed to predict the char burnout and if more coals were included, the number of variables could have been reduced further, without loosing the predictive ability of the models. The statistical analysis has shown interesting characteristics and correlations, especially with respect to the southern hemisphere coals with respect to their inertinite constituents. The study has further shown that a simple classification or understanding of coal composition cannot be applied, and that further research must be conducted in that respect. To validate the present models, it is recommended to perform tests with representative coals actually used in the power plants. It is recommended that the present limited database is extended in order to cover a wider range of coal types used in Danish power plants. This would ensure that more universally applicable and statistical valid methods would be made available to the power plants, in order to provide simple tools for rapid and inexpensive prediction of the burning behavior of the various coals. (EG) (10 refs.)}
place = {Denmark}
year = {1993}
month = {Aug}
}