E. Rose

Simulation of Sintering

Abstract The paper describes an updated model of iron-ore sintering and discusses results of computer simulation studies of the proces. A number of features are illustrated including the dependence of ignition 0n particle size and the effect on bed temperature of varying the limestone content of the raw mix. The effect of subjecting the surface of the bed to preheated air, rather than cold air, immediately after ignition is investigated. Finally, consideration is given to the use of parallel computing and the results of applying a parallel algorithm to the heat exchange problem are discussed.

A Self-Tuning Multivariable Predictor and its Application to the Control of Sinter Quality

Abstract A self-tuning multi variable predictor algorithm which is capable of providing multi step prediction of a multi variable stochastic time series is presented and applied to the problem of predicting the quality of sinter produced by an iron ore sintering process. Predicted information on strength index and FeO content of the sinter product is obtained from present and past measurements of these quantities together with coke-rate and a temperature-speed factor. The study includes the use of an extended low-pass filtering technique which produces a smaller time-lag than conventional low-pass filters. The predicted information may be used by the plant operator to effect better control of the end-product.

Sinter Strand Control; A Simulation Investment

Abstract The development of a mainly analytical model of a sinter strand process is discussed, A method of burnthrough control, effected by manipulating strand speed in accordance with a calculated average of effective pellet-diameter measurements is investigated. Simulation results show that this control strategy can lead to an increase in throughput.

An Analytical Study of Processes Occurring in an Iron Ore Sinter Bed

Abstract Fundamental heat and mass balance equations for an iron ore sintering process are presented and the method used in simulating the model is discussed. The model is validated by comparing the results of a digital computer simulation with those obtained from tests on pilot plant. The effects on heat wave propagation of varying ignition time, ignition temperature, size of mix, coke content, ore density, voidage and gas velocity are investigated. It is shown that burnthrough time is mainly dependent on gas velocity and that ignition time and temperature are not critical, provided that they exceed the values required for ignition.

Fuzzy Model Using a Modified LVQ Network for Sinter Strand Process

Abstract Rule-based fuzzy systems can be structurally mapped into Kohonens Learning Vector Quantization (LVQ) network. However the structure of the Kohonen layer is fixed in standard LVQ and the recall is in the manner of winner-takes-all. This is not suitable for the self-organizing of rule-based fuzzy systems. Modifications are made to the LVQs learning algorithm and the activity function of the Kohonen node. With the modified LVQ network the fuzzy model of an iron-ore sinter strand process is constructed. The resulting fuzzy model is dynamically self-organized from input-output observation data with very little a priori knowledge about the sinter strand process and is adaptive to the change in the environment in which the sinter strand process operates. The results of the simulation show that the model posseses very good performance in terms of prediction accuracy, adaptive ability and robustness.

IMPROVED SPEED CONTROL LAW FOR A SINTER STRAND PROCESS

Abstract The paper is concerned with the design of a feedforward control scheme aimed at minimising the waste product from a sinter strand process by manipulating strand speed. The results of simulation exercises are discussed in which the sinter bed is assumed to be composed of a series of lumped sections. The velocity of the flame-front within a section is assumed to be nonlinear and initially proportional to the raw-mix permeability. Permeability data from plant are used as input and strand speed and acceleration are constrained to lie within practically realisable limits. The results of simulation exercises using simple control laws based on straight and weighted averaging of ideal section speeds are presented. The proposed new control law involves a technique for modifying the time-shift increments associated with the straight-average law. The reduction obtained in output waste from the strand when the new control law is used is clearly demonstrated.

Lumped-Parameter Models for Control of an Iron-Ore Sinter Strand Process

Abstract The paper discusses the development of an empirical model of a sinter strand process in which the sinter bed is assumed to be divided into a convenient number of lumped sections. The results of response and control investigations carried out using a hybrid computer are presented. An alternative approach to modelling the sinter bed using fundamental physical and chemical relationships is briefly discussed and the need to divide each model section into a large number of subsections is explained. Finally the paper deals with computer control and the value of the models for computer control experimentation.

Control Investigations on a Nonlinear Lumped-Parameter Model of a Sintering Process

A digital computer model of a sinter strand process has been developed. The process involves the gradual burning of a continuous bed of material as it is conveyed along a strand. As the material passes from under the ignition hood the reaction zone gradually moves downwards through the material under a forced draught. The aim in controlling the process is to optimise the reaction rate so that the material has the required uniform properties as the process is completed.

Self-Tuning Force Control of a Flexible Structure

Abstract This paper presents the practical application of self-tuning pole-placement control to a single-actuator driven structural test rig. The controller synthesis employs a closed-loop pole-placement strategy. System identification based on recursive maximum likelihood estimation copes with the time-varying parameters and uncertainties within the system. Simulation and real-time implementation show that the system tracks a set-point load pattern whilst rejecting any noise or disturbances.

Predictive Fuzzy Control for the Sinter Strand Process

Abstract A predictive fuzzy control scheme is developed where an intermediate process model, adaptively identified from observed data, is used to perform on-line controller design to meet a closed loop performance specification which is associated with desired process output and predicted output. It is shown that the satisfactory response of the closed-loop system can be achieved by selection of the prediction horizon. The resulting separation of the adaptation system from the controller design enables learning convergency to be investigated. Simulation of the application of PFC to the control of a sinter strand process illustrates the ability of the method to control a complex industrial process.