The partial least-squares regression (PLSR) method is used to develop empirical correlations for predicting coal ash fusion temperatures (AFTs) under reducing conditions and the temperature of critical viscosity (Tcv) from the chemical composition. The database used in this work includes 433 ash samples from coals of different sources. PLSR is a powerful tool that is able to identify the most significant variables affecting coal ash thermal properties, suggesting ways of modifying the slag melting behavior. The final correlations show good predictive results with predictive standard deviations of less than 90 °C; they represent a simple and reliable tool for predicting the thermal behavior of coal ash and the effects of additives and blending of different coals.
Prediction of coal ash thermal properties using Partial Least Squares Regression
SEGGIANI, MAURIZIA;PANNOCCHIA, GABRIELE
2003-01-01
Abstract
The partial least-squares regression (PLSR) method is used to develop empirical correlations for predicting coal ash fusion temperatures (AFTs) under reducing conditions and the temperature of critical viscosity (Tcv) from the chemical composition. The database used in this work includes 433 ash samples from coals of different sources. PLSR is a powerful tool that is able to identify the most significant variables affecting coal ash thermal properties, suggesting ways of modifying the slag melting behavior. The final correlations show good predictive results with predictive standard deviations of less than 90 °C; they represent a simple and reliable tool for predicting the thermal behavior of coal ash and the effects of additives and blending of different coals.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
