A study was conducted to assess the feasibility of using Raman spectroscopy and statistical models to develop an in-situ, on-line melter monitoring system (OMMS) for waste glass and commercial glass melters. This article focuses on developing statistical models that relate Raman spectral characteristics to melt temperature and glass composition. The statistical models, based on classical calibration and inverse calibration methods, were developed using high-temperature data reported in the literature for sodium-alumino-silicate glasses. Within the composition and temperature regimes for which the models were developed, the predicted values for temperature and glass composition agree very well with the actual values. Hence, the feasibility study was considered a success, and additional work to develop an OMMS is planned.
Revised: July 23, 2001 |
Published: June 1, 2001
Citation
Piepel G.F., H. Li, C.F. Windisch, Y. Su, and M.L. Elliott. 2001.Statistical Modeling of Raman Spectroscopy Data from High-Temperature Glass Melts for On-Line Monitoring of Temperature and Composition.Quality Engineering 13, no. 4:667-677.PNNL-SA-34690.