Accuracy, precision and information of the adaptive Kalman filter in chromatography
β Scribed by Yuzuru Hayashi; Sarah C. Rutan
- Publisher
- Elsevier Science
- Year
- 1993
- Tongue
- English
- Weight
- 815 KB
- Volume
- 271
- Category
- Article
- ISSN
- 0003-2670
No coin nor oath required. For personal study only.
β¦ Synopsis
The accuracy and pregslon of the adaptwe Kalman filter were examined usmg computer slmulatlons of chromatograptnc sltuatlons, where a known (target) peak overlaps unth an unknown (mterferent) peak The peak overlap, area ratlo and noise level m the chromatograms slgmficantly affect the rehabdlty of the concentration estimates obtamed from the filter The accuracy (blas of the estimates) and preclslon (relative standard devlatlon, SD) for the known peak have smular tendencies as a function of peak overlap and area ratio For an area ratlo of 1, a nose level of 0 2% of the peak maxItnum and a peak resolution of 10, the bias for the known peak 1s 0 1% and the SD 1s 0 042%, If the peak resolution IS 15, then the bias IS 0 004% and the SD 1s 0 036% The function of mutual mformatlon (FUMI) 1s shown to represent approximately the rehatuhty @Ias and standard dewatlon) of the concentration estimates obtamed from the adaptive Kalman filter &~o&.s Liquid chromatography, Optimlzatlon methods, Accuracy, Kalman filter
π SIMILAR VOLUMES
Background signals must be removed from analyte responses before reliable qualitative and quantitative results can be obtained. For cases in which a reliable estimate for the background response cannot be obtained, the combination of two mathematical approaches yields accurate quantitative results.