Linear constrain relations in biochemical reaction systems III. Sequential application of data reconciliation for sensitive detection of systematic errors.

摘要: This article presents a method to test the presence of relatively small systematic measurement errors; e.g., those caused by inaccurate calibration or sensor drift. To do this, primary measurements—flow rates and concentrations—are first translated into observed conversions, which should satisfy several constraints, like laws conservation chemical elements. study considers three objectives: 1.Modification commonly used balancing technique improve error sensitivity be able detect errors. this end, is applied sequentially in time. 2.Extension enable direct diagnosis errors measurements instead diagnosing conversions. was achieved analyzing how individual are expressed residual vector. 3.Derivation new quantitatively determine error, that size at expected value chisquare function equals its critical value. The industrial data demonstrating effectiveness approach. It shown that, for most possible sources, 2% 5% could detected. In given application, variation N-content biomass appointed cause © 1994 John Wiley & Sons, Inc.