Development Object Oriented Framework for Data Reconciliation of Chemical Processes

Please enable javascript in your browser.

Aghamir Mohammad Ali, Mohammad Ali | 2010

747 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 40888 (06)
University: Sharif University of Technology
Department: Chemical and Petroleum Engineering
Advisor(s): Bozorgmehry Boozarjomehry, Ramin
Abstract:
In this study, in order to enhance data contaminated with random and gross errors, the implementation of data reconciliation technique on large-scale industrial unit was investigated. Data reconciliation results on the naphtha reformer unit, revealed that uncertainty in estimation of input and output reactor temperatures, decreased up to 2% in comparison with measured ones. In addition, uncertainties in estimation of the mass flow rates have declined by nearly 30%. Also, an object-oriented framework for plant wide data reconciliation was designed. This framework was designed as an extension of plant wide identification software developed previously. Moreover, adding five classes to the previous platform, made it capable to reconcile data in linear and nonlinear steady-state systems. Also a high order linearization method was developed in order to use benefits of linear data reconciliation methods. Results depict suitability of the proposed technique to impose linear data reconciliation to nonlinear systems. This technique simplifies both pre and post data processing in nonlinear systems
Keywords:
Data Reconciliation ; Gross Error Detection ; Object-Oriented Software Framework ; Approximate Linearization

No TOC