Application of macroscopic balances to the identification of gross measurement errors

摘要: A systematic method is presented which capable of both detecting the presence grossly biased measurement errors and locating source these in a bioreactor through statistical hypothesis testing. Equality constraints derived from material energy balances are employed for detection data inconsistencies subsequent identification suspect measurements by process analysis rectification. Maximum likelihood techniques applied to estimation states parameters after have been eliminated. The level if significance specified experimenter while assumed be randomly, normally distributed with zero mean known variances. Two different approaches analysis, batchwise sequential, that lead consistent set adjustments on experimental values, discussed. Several examples based fermentation taken literature sources demonstrate utility proposed method, one solved numerically illustrate computational aspect algorithm