L.V.D. : a software for mass balance equilibration and data validation

Didier Maquin (1) , José Ragot (1)

1 CRAN - Centre de Recherche en Automatique de Nancy

This paper deals with the problem of data validation in large scale steady state linear systems which are described by algebraic equations. The conditions for observability and redundancy are defined and a procedure is presented for classifying the system variables into observable, unobservable, redundant and no redundant variables. Reconciliation of the redundant process measurements is developped using a recurrent estimation technique which is well suited in real time data processing. Statistical tests based on the residuals or imbalance of the model constraints either individually or collectively allow to identify the sources or locations of the gross errors. The authors then present a software called L.V.D. as "Logiciel de Validation de DonnŽes" (Data Validation Software), which take into account all the major difficulties in the field of data validation for linear systems

Domaines: Automatique / Robotique

Didier Maquin, José Ragot. L.V.D. : a software for mass balance equilibration and data validation. European symposium on computer application in the chemical industry, Apr 1989, Erlangen, Germany. pp.555-560. ⟨hal-00293582⟩