Automation, Control and Modelling of Basic Oxygen Steelmaking
Automation, Control and Modelling of Basic Oxygen Steelmaking
In the basic oxygen steelmaking process, the basic oxygen furnace or converter produces liquid steel by reducing the carbon content of hot metal made by the blast furnace from around 4.5 % to 0.03 % to 1.0 %. The converter blows large amount of pure oxygen into the hot metal and refines it to steel in a short period of time. Presently, the basic oxygen steelmaking process employs combined blowing (top and bottom blowing). Bottom blowing is done with the inert gas. Various materials are used by the converter during refining. Besides hot metal and iron scrap as main raw materials, the other materials used in the basic oxygen steelmaking process are calcined lime, calcined dolomite or calcined magnesite for proper slag forming and different coolants (such as ore, sponge iron etc.) during the process. The operation of the converter needs the gas temperature to be set high, and it generates a large amount of dust.
The purpose of basic oxygen steelmaking process is to refine the liquid metal (molten scrap + hot metal) and to adjust the composition and temperature of the liquid steel. For meeting this purpose, the automation and control system is used for the steelmaking process which normally consists of basic automation system and process control system.
The engineering facilities at the basic oxygen steelmaking are in reality the design and fitting together of various sub-systems. The main equipment of basic oxygen steelmaking is a refractory lined converter vessel (basic oxygen furnace) in which the process of steelmaking takes place. Besides the converter vessel, the steelmaking process has several sub-systems consisting of (i) converter vessel tilting drive, (ii) oxygen lance system, (iii) inert gas bottom stirring system, (iv) top gas (converter gas) cooling, cleaning, analyzing, and recovery system, (v) sub-lance measuring system, (vi) slopping prevention system, (vii) material handling system, (viii) scrap charging system, (ix) flux and coolant charging system, (x) ferro-alloy charging system, (xi) horizontal temperature measuring and sample taking system, (xii) automatic tapping system, (xiii) slag stopper system, (xiv) secondary dedusting system, (xv) interlocking and alarm system, and (xvi) human machine interface (HMI) system.
In addition to these sub-systems, oxygen steelmaking is to be operated in an integrated way with upstream and downstream processes. Also, the steelmaking process is to be to be linked to the external systems such as (i) steel melting shop laboratory which houses optical emission spectrometer, and x-ray fluorescent spectrometers and other analyzing equipments, and (ii) supervisory control and data acquisition (SCADA) systems.
Basic oxygen steelmaking is a complex physico-chemical process which has a large number of influencing factors. Two methods are used for the control of blowing in the converter. The first method employs indirect measurement by the exhaust gas, whereas the second method uses direct measurement by the sub-lance. In the second method there is direct measurement of the temperature (in degrees C) of liquid steel simultaneously during blowing. This method is also used for various purposes such as bath levelling, slag levelling, measurement of oxygen concentration, and slag sampling.
In the basic oxygen steelmaking process, the classical process model still holds good where the operator is required to know as much as possible about the inputs, the process parameters, and the outputs, and he needs to have this information freely available to make the required adjustment to the process, so as to produce a prime quality product. For achieving this, various control and estimation techniques are required to be used and these techniques are to function in an organized way so as to provide the required information for the operator’s action.
The sub-systems which fit into this engineering hierarchy are (i) hot metal mass measurement, (ii) hot metal analysis, (iii) inert gas bottom stirring, (iv) oxygen feed, (v) charge temperature and analysis, (vi) flux and coolant charge system, (vii) ferro-alloy charge system, (viii) process control computers, and (ix) management computers. Measurements which are needed during the process of steelmaking are (i) temperature measurement, (ii) bath carbon content, (iii) bath depth, and (iv) complete chemical analysis. This has been normally achieved by stopping the process, tilting the converter, and taking temperatures and samples manually.
Process control is an important part of the basic oxygen steelmaking operation as the heat production times are affected by it. Several steelmaking process control strategies are available today, and steel plants use strategies depending on their facilities and needs. Process control models can be broadly divided into two categories namely (i) static, and (ii) dynamic.
The simplest form of process control is based upon a static process model. It consists of a set of balances for heat, oxygen, iron, and slag, combined with an equation of state. The latter describes the relationship between the iron content in the slag, the actual contents of manganese and carbon in steel, and the basicity of the slag. Static models determine the amount of oxygen to be blown and the charge to the furnace, given the initial and final information about the heat, but yield no information about the process variables during the oxygen blow. Static models are basically like shooting of an arrow. There is no further control once the arrow once it leaves the bow.
In case of dynamic process control, accurate information of the actual state of the blowing process is needed. Ideally, continuous information on the steel, slag, and gas compositions as well as the temperature is to be available and used on-line for process supervision. Any deviation from the progress of the anticipated process can then be detected and, based upon the models, the oxygen supply can be adapted or additional flux can be added into the converter. In a basic steelmaking converter, this is possible only in an ideal situation. In practice, the situation is totally different. Especially in the basic oxygen steelmaking process, there are strong practical limitations for continuous measurements, for example vibration, dust, high temperature and liquid metal and slag phases. Dynamic models make adjustments during the oxygen blow based on certain in-blow measurements.
The requirements of a dynamically controlled process are (i) not to interrupt the process and (ii) to get the real time measurements. A sub-lance system which can handle the process conditions and utilizing disposable sensors in the lance-tip is used for this purpose. The different sensors are characterized by their measurement functions, the most important ones are (i) bath temperature measurement, (ii) bath carbon measurement, and (iii) bath level measurement. Any combination can be used.
The main functions of the basic automation system include oxygen lance control, material control, bottom stirring control, sub-lance detecting control, and end-point control. The process control system carries out production management, control models, process control, and data management. The process control system is used to control the basic automation system. Firstly, it collects information about the melting process and detections made by the sub-lance. Then it judges the status of the melting process according to the results of model calculations. Finally, it sends signals to the basic automation system to control the adjusted parameters.
The automation and control of basic oxygen steelmaking not only considers the converter specific process functions, but also takes into account the relevant parameters of the charging materials, including hot metal preparation, scrap yard management, and scheduling logistics. Process optimization (Level-2) solutions are based on advanced algorithmic equations, which accurately represent the complex thermodynamic metallurgical reactions. The solutions are mainly suitable for a wide range of operating conditions, e.g. variable scrap to hot metal ratios, minimum slag practice, and varying phosphorus content.
The key objectives of the automation and control of basic oxygen steelmaking process are (i) meeting the requirements of the steelmaking, and (ii) providing the operational assistance. Further, the automation and control of the steelmaking process is an effective way (i) to provide comprehensive and consistent process information for the guidance of the operator, (ii) to ensure standardized operations for homogeneous quality of liquid steel, (iii) to improve the performance of the process, (iv) to improve the accuracy of the endpoint control, (v) to shorten the heat cycle, (vi) to enhance the productivity through optimized steelmaking, and (vii) to reduce the production costs by using process models for optimized material usage and energy input. The automation and control relies mainly on computers and is inseparable from the mechanization of the steelmaking process.
The general architectural structure of the automation and control of basic oxygen steelmaking process incorporates (i) corporate information system, (ii) steel melting shop management information system, (iii) process control, and (iv) field instrument and equipment.
As can be deducted from the various sub-systems and the interfacing which exists between them it becomes clear that the required interlinking cannot be achieved with conventional (analog) circuits.Wide use has hence to be made of digital process control equipment which offers various advantages such as (i) additions and system changes can be easily accommodated, (ii) advanced control strategies can be handled, (iii) intelligence can be programmed into the system, (iv) effective backup facilities can exist, (v) CRT (cathode ray tube) operators interface can be incorporated with a large format of display options, (v) existence of stored data, (vi) easy access to information and stored data, and (vii) communication between higher and lower hierarchy. Fig 1 shows the basic automation and process control system of basic oxygen steelmaking.
Fig 1 Basic automation and process control system of basic oxygen steelmaking
Increased speed and capacity of computers, adoption of programmable logic controllers (PLCs) in electrical and control systems, and switch over from analog to digital instrumentation have resulted into remarkable improvement in the control accuracy. Further, the application of direct digital control in recent times has accelerated the automation of the steelmaking process.
Along with the advancement of process computers and peripheral measuring technology, blowing control for converter has shifted from a static control system to a dynamic or fully automatic operational control system. Further, because of the technological advances made in the electrical and control systems, controllers have moved from instrument panels to CRT displays allowing operators to monitor and control the steelmaking process on the CRT screens. Also, with the use of mathematical models and expert system (using artificial intelligence programmes) the automation and control of the steelmaking process has become more operator-friendly.
Control systems for different sub-systems are frequently configured as DCS (distributed control system) and PLC (programmable logic controller) which seamlessly connects to the DCS of the basic oxygen furnace and provide integrated monitoring and control. The unique advantage of this integrated approach is that it covers the aspects of process stability, product quality, operation flexibility, and improved working environment while safeguarding efficiency and cost effectiveness.
The distributed control instrumentation accommodates (i) production operators’ consoles, (ii) live mimics, (iii) instrument displays and control, (iv) trend graphs, (v) and logs. Distributed computers on a data-highway are used with the required I/O (input / output) to handle (i) water systems, (ii) weighing systems, (iii) bottom stirring system, (iv) oxygen systems, and (v) communication with the host computer. Management information / control computer is normally a large capacity system which is primarily used to (i) supply information i.e. shift / day / month reports, (ii) handle interactive production, (iii) scheduling between downstream / upstream plants, (iv) prepare for charge (pre-Ioading of scrap etc.), (v) accommodate the static models such as heat balance, determining the fluxes (lime/ dolomite) and cooling agents (sponge iron / iron ore), quantities as well as when to charge, and oxygen balance (determine the rate, duration and blow pattern), (vi) accommodate the dynamic model which come into operation after the sub-lance has provided real-time information. The system generates an active display which enables the operator to end the process on target, to calculate final results, and to suggest minor modifications and to add the final alloys.
The control models are the core part of the automatic steelmaking control system. They integrate the knowledge of melting mechanism, mathematical statistics, expert principles, and adaptive learning. The control equations are derived using the knowledge of melting mechanism and the key control parameters are defined by mathematical statistics and expert principles. Moreover, these control parameters can be regularly modified through adaptive learning. The control models refer to the static control model, main materials model, slag forming model, temperature model, oxygen consumption model, dynamic control model, slopping model, alloy model, and end-point model etc. Further, there is also an adaptive learning model. Different detection equipments used are sub-lance, mass spectrometer, flame spectrometer, microwave distance meter, and monitoring device for oxygen lance vibrations etc.
Further, presently there are several control models, such as mechanism model, statistic model, and incremental model, etc. The mechanism model is based on the heat and mass conservation. It determines the relationship among variables by mathematical derivation. However, it is not suitable for application due to the complexity of the melting process. The statistic model is based on the black-box theory. The physico-chemical process is ignored in this model. It is only concerned with the statistical relationship between input and output parameters. The calculation accuracy of this model cannot be maintained as long as the melting condition is changed. Using the incremental model, the operating parameters can be refined by comparing with the productivity data recorded. It can overcome the influence caused by the changes in the melting conditions. However, the main shortcoming of this model is low calculation accuracy. Fig 2 shows the functions of the control systems and process models.
Fig 2 Functions of the control systems and process models
End-point carbon prediction
End-point carbon prediction has been initially relied on the experience and skill of the operator. It is well known that this method is inefficient and difficult, especially for the medium high carbon steel melting process. With the development of computer and information technology, the study on computer control of the basic oxygen steelmaking has been carried out. The static charge model based on computer calculation was first exploited by Jones & Laughlin Steel Corporation, and was used to calculate the amount of charged hot metal, scrap, and slagging materials and guide endpoint carbon control of liquid steel.
With the rapid development of auto-detection methods, mathematical models, and algorithms, dynamic and intelligent end-point carbon prediction has become available for steelmaking process. Based on the features of collected data, which has been used to calculate the end-point carbon content, the end-point carbon prediction is divided into three stages, such as static prediction, dynamic prediction, and intelligent prediction.
Static prediction – During the entire process of basic oxygen steelmaking, operators are normally assisted by a computer-based guidance system, which proposes process parameters and operator actions based on calculations of mass and energy balance and thermodynamic calculations. The static end-point carbon prediction mainly relies on the mathematical model established based on mass and heat balance, which can calculate the end-point carbon content in the liquid steel based on the initial charge parameters (such as charged hot metal and scrap, and composition and temperature of hot metal). Fig 3 shows the static model for end-point prediction for basic oxygen steelmaking.
Fig 3 Static model for end-point prediction for basic oxygen steelmaking
The key point of static end-point carbon prediction is the reasonable establishment of mathematical mode and acquisition of initial amount data. Compared with the randomness and uncertainty of end-point carbon prediction based on the experience and skill of the operator, the static end-point carbon prediction can perform a quantitative calculation of blown oxygen and end-point carbon content, which improves the prediction accuracy of the end point carbon. The normally used mathematical models for static end-point carbon prediction mainly include theoretical model, and statistical model
The theoretical model can calculate the quantity of blown oxygen and end-point carbon content based on the calculations of mass and heat balance during the steelmaking process. Due to the complex interaction between various influential factors in the basic steelmaking process, the calculation of mass and heat balance is normally completed with empirical values and inaccurate, hence, the theoretical model shows a relatively poor performance on basic steelmaking converter end-point carbon prediction.
The statistical model only concerns the relation between input variables and output variables using statistical analysis of collected data without considering the chemical reaction mechanism in the liquid bath, which is depicted by the equation X = F (W, S, T, t, Z), where ‘F’ is a linear or nonlinear function, ‘W’ is the charged weight of hot metal and scrap,’ S’ are the target values of end-point composition in liquid steel, ‘T’ is the initial temperature of hot metal, ‘t’ is the oxygen blowing time, and ‘Z’ are other important influential factors (such as top lance height and oxygen pressure).
As a kind of statistical model, the back propagation neural network combined with different algorithms is widely applied to the end-point prediction for basic oxygen steelmaking in recent years. Compared with the theoretical model, the neural network is specialized in analyzing random deviation and eliminating the influence of random factors and it can provide a more reliable reference for end-point carbon prediction. Fig 4 shows back propagation neural network for end-point carbon prediction.
Fig 4 Back propagation neural network for end-point carbon prediction
However, the theoretical and statistical models described in the above section are only built on consideration of initial conditions and static process data (a small dataset without time-series feature cannot represent the actual production), making static endpoint carbon prediction models unsuitable for actual production since the prediction accuracy is limited. A particular challenge of static end-point carbon prediction is the reasonable establishment of prediction models based on a large production dataset which has a time series feature. Based on the above challenge, the dynamic end-point carbon prediction is rapidly developed based on static prediction.
Dynamic prediction – Differed from the static control, the dynamic end-point carbon prediction can predict the end-point carbon content in liquid steel and fulfill on-line adjustment of operating parameters with the calculation of dynamic models established on the time-series data (lance movements, carbon mono-oxide and carbon di-oxide levels of the off-gas, spectral features of the flame) collected by monitoring devices. Presently, the sub-lance system, off-gas analysis system, and flame spectrometric analysis system are the primary methods which are applied to dynamic end-point carbon prediction for basic oxygen steelmaking. Fig 5 shows the dynamic end-point carbon prediction with the sub-lance system.
Fig 5 Dynamic end-point carbon prediction with the sub-lance system
The dynamic end-point prediction with the sub-lance system directly measures the carbon content of liquid steel at a later stage of the blowing process, build the on-line prediction model, and dynamically predict the carbon content with different blowing times. With the application of a sub-lance system, the influence of initial deviation on charged materials can be reduced, and the end-point carbon prediction is more accurate and precise compared with static prediction. Some Japanese steel melting shops are achieving a carbon predictive accuracy of more than 90 % with an error tolerance of + / – 0.02 %.
Through the monitoring the off-gas information (carbon mono-oxide and carbon di-oxide content change during oxygen blowing), the carbon content of liquid steel can be dynamically inferred with a mathematical model based on off-gas information, and the end-point carbon content can be predicted and controlled with the feedback of calculation results. Since, it is an indirect estimation method, the accuracy of collected data (such as off-gas content and flow rate) and the respond time of mathematical model greatly affects the prediction accuracy of the end-point carbon. Hence, the off-gas analysis system is normally used together with the sub-lance system to control the end-point carbon with required accuracy in several steel melting shops.
Spectral features of the flame at the basic oxygen converter mouth are related to the carbon content of liquid steel and hence, change during the basic steelmaking process. Based on the spectral characteristics of the flame radiation information, the flame spectrometric analysis system has been developed to predict end-point carbon content. The on-line prediction of carbon content of liquid steel can be completed, through analyzing the relation between flame spectrum of different blowing time and converter bath status.
Optical sensors have been used to dynamically predict the carbon content of low carbon heats (the target end-point carbon content lower than 0.06 %) in basic oxygen steelmaking in a steel melting shop, which has resulted in a considerable improvement.
Although the dynamic end-point carbon prediction can give a significant prediction improvement when compared with the static prediction, the collection of a real, full sized, and rich dataset which can represent the overall behaviour of the entire steelmaking process, self-learning, and self-adapting of the prediction model are particular challenges of dynamic end-point carbon prediction. Hence, the intelligent end-point carbon prediction is built based on dynamic end-point carbon prediction.
Intelligent prediction – With the development of data collection and intelligent models, intelligent end-point carbon prediction for basic oxygen steelmaking has now become available. It is established on the basis of full-sized and rich dataset with different features and has strong ability of self-learning to improve prediction accuracy. Besides the sub-lance system, automatic basic oxygen steelmaking system mainly adopt other techniques namely (i) on-line slag detection during oxygen blowing providing the guidance for slagging operation, (ii) off-gas analysis system dynamically estimating the carbon content and temperature of liquid steel during blowing process, and (iii) intelligent models with strong ability of self-learning and self-adapting. Fig 6 shows establishment of an intelligent model.
Fig 6 Establishment of an intelligent model
With application of above techniques, the intelligent endpoint carbon prediction for basic oxygen steelmaking can be automatically and efficiently implemented with computer rather than manual operation, and the prediction accuracy of end-point carbon content is greatly improved. With the practical application of intelligent end-point prediction at a steel melting shop has reduced the reblow ratio from 14 % to 1 %, and the tap-to-tap time has reduced from 37 minutes to 29 minutes, thus greatly improving the efficiency of the basic oxygen steelmaking process.
There is no doubt that intelligent prediction greatly improves the prediction accuracy of end-point carbon content. Recently, more and more auto-detection techniques are developed and applied to the end-point carbon control for basic oxygen steelmaking, such as robotic sampling and temperature measuring system, wireless composition measuring system, and so on. Meanwhile, Industrial Internet of Things (IIoT) is developed rapidly along with the fifth generation mobile communication technology (5G) and big data analytics, and it is gradually applied to smart steel manufacturing. Hence, the intelligent end-point carbon prediction for the basic oxygen steelmaking is gaining more attention in future.
From the industrial implementation point of view, the intelligent end-point carbon prediction model with high accuracy can be installed on existing process system, continuously predicting the process carbon content and providing the guidance for operators based on the actual events and planned events during the basic oxygen steelmaking process.
Expert system is the comprehensive group of process models which images and optimizes the process of steelmaking. Expert system monitors the metallurgical and thermal process, cyclically calculates the actual condition of steel bath and slag. This provides the analysis and temperature of steel bath and slag at any time and the set-point model calculations are always based on the actual situation.
The expert system process models optimize and control the steelmaking process during the entire treatment in the converter. The Level-2 system assists the operator according to the model calculations based on the stored production schemes per steel grade. Expert system continuously informs operating personnel about the overall status of the heat (i.e. weight, temperature, and analysis) during treatment.
The extensive group of set-point models in expert system determines the expert system set-point, which is responsible for determining the needed supplies of raw materials, gas volumes, and / or energy for different treatment steps. Some of the set-point models are described here.
Expert system first charge calculation is done by taking different scenarios of variable input data (such as variable scrap and variable hot metal, variable scrap and fixed hot metal, or fixed scrap and variable hot metal) can be applied. Additionally scrap cost optimization can be used. As model output the optimum charge mix to reach the targets following the planned steel grade from the production schedule is provided.
Expert system second charge calculation is done immediately after the actual data related to the charged hot metal and scrap, including the partial weights of the different scrap types has been received. The second charge calculation model calculates the necessary vessel additions and oxygen volume to achieve the target analysis and the target temperature of the steel at the end of blow.
Expert system in-blow correction calculation is done by sub-lance model. Depending on the availability of the data (temperature, carbon) the cyclic online model takes over the measured values and applies some corrective measures due to the fact that the sub-lance measurement is done close to the hot spot. The remaining required oxygen amount, heating, or cooling agents and additional slag formers are calculated.
Expert system re-blow correction calculation can be started if certain steel bath properties (e.g. temperature, carbon content, or phosphorus content) are not within the specified target limits at the end of blowing. The actual steel bath analysis and temperature are taken from a temperature measurement or from an actual steel sample. The required oxygen amount, heating or cooling agents and additional slag formers for re-blow are calculated.
Expert system alloying model calculates cost optimized the necessary alloying and deoxidation materials to be added into the tapping ladle. The analysis of the alloying agents and their specific losses are taken into account.
Expert system prediction model performs a simulation of the complete production process by using the results of supervision and set-point models. It provides a forecast of the progress and the final condition of the heat. It also predicts all required additions and actions and serves to optimize the production process. In the typical HMI screen for the prediction model, the different sections of the screen display the target and input data, model results, calculated analyses for steel, and slag and specific consumptions.
The expert system pre-calculation model simulates the complete steelmaking process before / after scrap and hot metal are charged into the converter. Expert system prediction model determines the optimum blowing and stirring strategy, as well as the exact time and portioning of vessel additions. The pre-calculation model is based on a pre-defined list of process steps (e.g. charging, main blowing, stirring, and tapping etc.) and target values from the standard operating practice (SOP) as defined by the process engineer.
The pre-calculation model consists of five different parts namely (i) calculation of hot metal and scrap input, (ii) calculation and distribution of heating and cooling agents, alloys, scraps and fluxes in order to reach the target weight, analysis, and basicity, (iii) calculation of blowing set-points in order to reach the target carbon content and temperature, (iv) calculation of ongoing reactions to predict the weight and analysis of steel, slag, and off-gas after each process step, and (v) information and warnings for the operator if target values for a process phase not reached.
Expert system supervision model which is an on-line model calculates cyclically the ongoing reactions in the steel bath and in the slag during the blowing process. This includes oxidation and reduction reactions, pick-up of oxygen, nitrogen and hydrogen, sulphur and phosphorus distribution between steel and slag and the post combustion from carbon di-oxide and hydrogen. In this way the effect of different blowing, stirring or material addition patterns as well as the dissolution of charged materials is taken into account for the course of the process.
Expert system dynamic control (a part of the expert system supervision model) is the dynamic blow end prediction for carbon based on actual off-gas data. From the actual off-gas data (such as off-gas flow, off-gas analysis (carbon mono-oxide, carbon di-oxide, oxygen, and nitrogen) as well as from actual process data expert system dynamic control predicts the carbon content at the end of the blowing process from the typical profile of the off-gas data close to blowing end. The result is a predicted carbon content at the end of the blowing process (typically for carbon contents below 0.3 %) and a blowing end request to reach the target carbon content at blowing end. In combination with the cyclic online model (expert system supervision model) a complete prediction of steel and slag (temperature, analysis and weight) can be made where the carbon content is taken from the expert system dynamic control and all other data are calculated by the expert system supervision model.
In the expert system, the carbon content calculation for the in-blow measurement is based on the raw data (i.e. the liquidus temperature Tliq) from the sub-lance measurement device instead of using the carbon content calculated by the measurement device. The in-blow carbon content is calculated using the equation Cin-bolw = a0 + a1xTliq + a2x square of Tliq. The tuning parameters a0, a1, a2 are maintained in the Level-2 database and are fitted by employing pairs of liquidus temperature and the carbon content from the in-blow sample.
The calculated carbon content from the in-blow measurement is taken over by the on-line model and thus corrects the carbon prediction model. To complete the existing automatic blow-stop functionality for basic oxygen converters based on dynamic off-gas measurement, this functionality is adopted for sub-lance systems as well. The automatic blow-stop functionality prolongs or shortens the final blowing phase in order to reach the temperature and carbon aims at end-of-blow.
The cyclic process model also known as saturation model considers the saturation concentrations of complex steelmaking slags by CaO (lime) and MgO (magnesia). Lime and dolomite dissolution is suspended when the corresponding saturation concentration is reached and continues when the slag composition allows further dissolution of slag-forming additions. Thus the process model keeps track of the liquid slag amount and analysis as well as the undissolved flux additions. The calculation of the equilibrium phosphorous distribution ratio is based on the optical basicity model. For determining the optical basicity only the composition of the liquid slag phase is employed, while the portion of undissolved fluxes is to be considered in the calculation of the mass transfer coefficients. Normally, the saturation model allows optimizing basicity (CaO / SiO2) and MgO and aims in order to avoid too much undissolved flux materials at the end-of-blow.
The expert system process model accounts for the thermal cracking of slag-forming additions which have been charged prior to hot metal charging. For these additions the portion of carbon di-oxide and water vapour is removed completely. This prevents to overestimate the cooling effect of pre-charge fluxes like limestone or raw dolomite and thereby improves the temperature calculation. Furthermore, the remaining slag in the converter from the previous heat is partially reduced by silicon after hot metal charging as per the reactions 2(FeO) + [Si] = 2[Fe] + (SiO2], 2(Fe2O3) + 3[Si] = 4[Fe] + 3(SiO2), 2(MnO) + [Si] = 2[Mn] + (SiO2) and to a small extend also by carbon. In case of considerable amounts of remaining slag, the reduction of FeO, Fe2O3, and MnO affects the temperature profile.
Whereas the models are adjusted specifically to the special requirements of the different sub-systems, the principle of expert system of combining the features of prediction, supervision, and set-point models for perfect quality is applied throughout the steelmaking automation.