Premium
Principles of dynamic balancing
Author(s) -
Almasy Gedeon A.
Publication year - 1990
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.690360905
Subject(s) - kalman filter , dynamic balance , residual , state variable , computer science , control theory (sociology) , mathematics , mathematical optimization , algorithm , statistics , engineering , physics , control (management) , thermodynamics , mechanical engineering , artificial intelligence
The principles of static balancing, also called measurement error reconciliation (MER), have been generalized to transient conditions. An exactly linear dynamic balance model has been stated where flow variables are assumed to be stochastic processes with independent increments. With this model the residual error of static MER due to dynamic effects can be estimated. Dynamic balancing is proposed as the estimation of flow and inventory variables by Kalman filtering applied to the balance model. The error of estimates obtained in this way is significantly less than that for the static MER estimates, even if the process is nearly in steady state. The main problem of dynamic balancing is that it requires not only the redundant observation of flow variables but also the observation of inventory variables. Compatibility of MER and dynamic balancing has been shown.