Robust Data Reconciliation Based on A Generalized Objective Function

Abstract

A robust data reconciliation method is proposed. This approach is based on the minimization of a generalized objective function, which is obtained from residuals probability density function estimation. Existing approaches are based on the minimization a pre-defined objective to take into account random and gross errors. The proposed approach is optimal in the sense of maximum likelihood estimation, allows the use of various model constraints and can be used for both steady state and dynamic systems. The performance of the proposed method is illustrated by a chemical engineering example.