Year

1992

Degree Name

Doctor of Philosophy

Department

Department of Materials Engineering

Abstract

Online Realtime Dynamic Analysis of the Blast Furnace Process has been investigated to include the concurrent application of Aerodynamics, Thermodynamics and Metallurgical fundamentals applied to the Iron Formation Processes.

The aim of this study is to establish a Process Control System to control the Instantaneous Iron Production Rate for the Blast Furnace. The system depends upon the ability to continuously monitor, control, and locate the various Iron forming reactions throughout the Process. Three methods for control are outlined beginning with the simple Isotherm based version through to the more refined method with five classes of identification of the Zone Volume Interface boundaries based on streamline Gas and Particulate knowledge.

Experiments using Instrumented Mt. Newman Iron Ore Particles subjected to heating patterns which simulate Burden descent Streamline flow paths, provide a method of classification of the various Blast Furnace Zone Volumes and Iron Production Rates in terms of the Iron forming reactions that occur in the various Zone types. Specially designed equipment to provide a controlled environment to simulate Blast Furnace conditions also allowed Visual observations and Video recordings of Instrumented Iron Ore Particles to be made as they were reduced to Liquid Iron.

These Visual observations added a new dimension to the Aerodynamic and Metallurgical Knowledge of the Process, namely the suddenness with which the Liquidus mechanisms occur in only a few seconds, and secondly the accompaning Volumetric change effects that also occur. The importance of these Process mechanisms were not previously appreciated or even evident using Instrumented recording methods. Using a combination of Visual observations, Isotherm recordings and Quenched sample Ore Metalography, a number of Aerodynamic, Thermodynamic, and Metallurgical concept were developed and then used to more accurately designate the various Zone Volume Boundaries.

The concept of a Flame Front Reduction Characteristic (Ffrench Zone) was the first Visualised Reaction mechanism developed in line with the Aerodynamic viewpoint of the reaction observations applied to a uniformly distributed packed bed, as for example in a Sintering bed. This Ffrench zone concept was further refined by applying concurrently Aerodynamic and Thermodynamic Knowledge to the Ore Layers of the Burden. This introduced the concept of the Ore Layer Stratification effects where reactions occur on individual particles in and around the Iron Liquidus Isotherms (where liquid Iron first forms) as particles are affected by the characteristic flow streamline contours resulting from the preferential gaseous flow pathways through the alternate burden layers of Coke and Ore. Additionally, Individual particle Stratification effects also apply within the proximity of each particle where the Iron Liquidus Isotherm condition exists. The Rate of available Heat also determines the Metallurgical nature of the reactions that occur. A further gaseous flow induced effect results from the interparticle Gas jetting impingement (Gas Lithology) onto the successive particles and produces localised forced reactions on the particle surface from the concentrated Heat in the jet stream. These combined effects provide key dynamic components for a more accurate location of Zone interface boundaries and Iron formation mechanisms, and all of which contribute to providing more accurate knowledge for Online Realtime Process Control and Process Simulation Systems.

Instantaneous Iron Production Rate for the Blast Furnace Process trialed at No.5 Blast Furnace Port Kembla was visualised initially in relation to zone volumes classified by "S" Curve direct reduction zone profiles based on the Ffrench Zone concept. Various other Blast Furnace Process Control Systems were also developed to take advantage of the technological advancement that occurred during the period of this study including Online Operator Visualisation and programming concepts. Inherent technological shortcomings of these systems led to the development of the Knowledge based Visualisation concept design which is considered necessary to concurrently utilise the voluminous Streamline based Process knowledge for Online Realtime Process Control use.

Future Process control systems providing Online Realtime Process Control are envisaged as using streamline based analogue Instrumentation sensors,with neural data array processing of gas and Particulate streamline flow concepts to provide the desired operation of the Blast Furnace Process, thus yielding the quality and quantity of saleable product required.

A theoretical internal fluid motion physical property designated Harmonology was developed as a fundamental Process Unit Dynamic mechanism responsible for "creating" matter motion of fluids at the atomic level, and subsequently are manifested as Visible flow patterns. This concept evolved as a result of concurrently reconciling the matter migration at high temperature as observed in Metalography studies of Iron Ore reduced samples when internal voids are formed (Newman Effervescent Voids).

In future systems, Process Visualisation in terms of "Aurora" displays modelled on human visual and Knowledge concepts (Humanoids) are envisaged to provide "multi-personality" dynamic insights of the "moods" of the Process to meet the challenge of providing concurrent knowledge Visualisation for the indispensable human operator comprehension and mood "matching" for Online Process Control purposes. The Humanoid Visualisation incorporates complementary mood and comprehension "matching " personality traits designed to maintain Human involvement and motivation with the dynamic behaviour of the Visualisation Aurora Imagery.

In an overall sense theoretical Process considerations suggest that Particulate matter can inherently be more accurately blended and flow rate controlled as a single continuous flow stream of Powder. Subsequently this powder is Transformed to Liquid Product in a Blended Powder Direct Reduction Transformer.

The single most important outcome of this thesis was the reinforcement of the concept that concurrency of knowledge and knowledge crosspollination is needed in a dynamic reasoning and visualised fashion. This approach is considered necessary to both understand, and online predictively control the quantity and quality of the Process and its Production Products in terms of the raw materials from which they are produced.

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.