You need JavaScript to view this

Modelling of structure and properties of soft carbons with application to carbon anode baking

Abstract

This work deals with topics related to modelling and control of ring furnaces for the baking of carbon anodes used in aluminium electrolysis. Anodes made of a granular coke and coal tar pitch are used in aluminium electrolysis. The anode properties are imperative for successful operation of the aluminium smelters. After mixing and forming the anode paste, heat treatment of the carbon blocks takes place in so-called ring furnaces. A ring furnace consists of a series of heat treatment sections where each section is loaded with a batch of anodes. The heat treatment of the anodes in a section consumes a lot of energy, and the anode properties partly depend on the heat treatment program. Previous work in the field of ring furnace modelling, operation and control is shortly reviewed. Both petroleum coke and coal tar pitch belong to the group of soft carbons. Models for structural parameters and porosity of soft carbons are developed. Furthermore, a new model for pyrolysis of coal tar pitch is proposed. Based on the models for pyrolysis, structure and porosity, new models for properties of single phase carbons and composite anodes are developed. These models are suitable for use in optimization of the baking  More>>
Publication Date:
Nov 01, 1998
Product Type:
Thesis/Dissertation
Report Number:
NEI-NO-1341
Resource Relation:
Other Information: TH: Thesis (Dr. Ing.); 313 refs, 174 figs., 34 tabs.; PBD: Nov 1998
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; ENERGY; ALUMINIUM; ANODES; HEATING; ELECTRIC FURNACES; METAL INDUSTRY; SMELTERS; PYROLYSIS; MATHEMATICAL MODELS
OSTI ID:
20185885
Research Organizations:
Norges teknisk-naturvitenskapelige universitet, Trondheim (Norway)
Country of Origin:
Norway
Language:
English
Other Identifying Numbers:
Other: ISBN 82-471-0269-2; TRN: NO0105187
Availability:
Available to ETDE participating countries only(see www.etde.org); commercial reproduction prohibited; OSTI as DE20185885
Submitting Site:
NW
Size:
580 pages
Announcement Date:
Mar 15, 2002

Citation Formats

Gundersen, Oeyvind. Modelling of structure and properties of soft carbons with application to carbon anode baking. Norway: N. p., 1998. Web.
Gundersen, Oeyvind. Modelling of structure and properties of soft carbons with application to carbon anode baking. Norway.
Gundersen, Oeyvind. 1998. "Modelling of structure and properties of soft carbons with application to carbon anode baking." Norway.
@misc{etde_20185885,
title = {Modelling of structure and properties of soft carbons with application to carbon anode baking}
author = {Gundersen, Oeyvind}
abstractNote = {This work deals with topics related to modelling and control of ring furnaces for the baking of carbon anodes used in aluminium electrolysis. Anodes made of a granular coke and coal tar pitch are used in aluminium electrolysis. The anode properties are imperative for successful operation of the aluminium smelters. After mixing and forming the anode paste, heat treatment of the carbon blocks takes place in so-called ring furnaces. A ring furnace consists of a series of heat treatment sections where each section is loaded with a batch of anodes. The heat treatment of the anodes in a section consumes a lot of energy, and the anode properties partly depend on the heat treatment program. Previous work in the field of ring furnace modelling, operation and control is shortly reviewed. Both petroleum coke and coal tar pitch belong to the group of soft carbons. Models for structural parameters and porosity of soft carbons are developed. Furthermore, a new model for pyrolysis of coal tar pitch is proposed. Based on the models for pyrolysis, structure and porosity, new models for properties of single phase carbons and composite anodes are developed. These models are suitable for use in optimization of the baking process. A detailed mathematical model of a part of the heat treatment process is formulated in three spatial dimensions. The model is based on first principle descriptions of fundamental physical and chemical phenomena and the resulting model appears as a set of partial differential equations. The spatial differential operators are discretized by using the finite volume approach. In this way, a high dimensional nonlinear state space model is obtained. The model has been simulated using the method of lines. A vector of quantities which describes the anode properties is defined. This property vector constitutes a systematic definition of anode quality where the quality parameters are calculated as nonlinear transformations of the state space vector. Models are derived for some anode properties. Anode properties are not directly measurable during normal process operation but they can be calculated by using the proposed property model. In this way, it has been possible to study anode property evolution as function of the furnace operation strategy. The ring furnace model has not been verified, but there seems to be a satisfactory correspondence between measurements and model calculations. A model based control strategy based on the proposed model is presented. Open loop optimization of process economy is suggested for calculation of the nominal heat treatment trajectory. Since a certain heat treatment program can be used for many batches of anodes, the solution of the optimization problem can be solved off line. In real process operation, deviations from the nominal disturbances may occur. A control corrector based on classic control theory is suggested to cope with non-nominal disturbances. The main contributions in this study are as follows: (1) A model of pyrolysis of coal tar pitch derived with basis in first principles. (2) Models of carbon properties derived from fundamental principles. (3) A detailed first principles mathematical model of a part of an anode baking furnace. (4) A two-level model based control strategy for the ring furnace: (a) Off-line optimization of process economy and anode quality, (b) A basic level control corrector for on-line tracking along the optimal trajectory. One of Hydro Aluminium's ring furnaces in Aardal has been considered during the development of the mathematical model for the ring furnace. Still, however, many of the results presented in this study have general significance.}
place = {Norway}
year = {1998}
month = {Nov}
}