Patent Grant
8494679
The present application is a 371 of International application PCT/EP2010/069599, filed Dec. 14, 2010, which claims priority of DE 10 2009 060 258.5, filed Dec. 23, 2009, the priority of these applications is hereby claimed and these applications are incorporated herein by reference.
The invention concerns a method for the dynamic control of the converter process, especially during the blowing of oxygen in steelmaking by means of exhaust as analyses.
Modern steelmaking processes require exact knowledge of the current composition and temperature of the liquid metal being processed. Especially during steelmaking in a converter, high match rates for the final carbon content and the bath temperature are necessary. Quantitatively precise blowing of oxygen corresponding to the final target carbon content and the metal temperature are decisive factors affecting the economy of the process and the quality of the steel produced.
In order to realize relatively precise process monitoring, various procedures and processes are known, whose application is based in most cases on mass flow balances and the measurement of exhaust gas composition. All models work with a precision that depends on the exactness of the input data, especially with respect to the weight data of the raw materials and the chemical properties of the metal that is being used.
DE 28 39 316 A1 discloses a method for controlling a steelmaking process, in which the decarburization of the molten steel takes place under atmospheric pressure and with the formation of an exhaust gas containing CO, CO2, and N2. This method has the following steps: the formation of an intimate gas mixture from the exhaust gas and a measured quantity of a reference gas that is inert relative to the exhaust gas, mass spectrometric monitoring of a sample of the intimate mixture for the ionization currents for selected peaks related to the CO, CO2, N2, and reference gas in the sample, determination of the rate or amount of decarburization of the molten steel at the time of the monitoring from the measured value of the quantity of reference gas in the mixture and the measured values of the ionization currents for the selected peaks, and control of the steelmaking process according to the determined value of the rate or amount of decarburization of the molten steel.
WO 2008/049673 A1 describes a method for automatically controlling the CO output of the steelmaking process, in which oxygen (O2) is introduced into the melt to remove the carbon (C) present in the steel, the actual value of the carbon stream escaping from the melt is determined, a desired value for the escaping carbon stream is calculated on the basis of the amount of oxygen supplied and the carbon content of the melt, taking in account any other reactions, the desired value and the actual value are compared, and if the actual value is lagging behind the desired value, measures are undertaken to prevent the abrupt rising of gas bubbles. The following are specified as suitable measures:
WO 2009/030192 A1 discloses a method for indirect determination of the exhaust gas rate in metallurgical processes. This method is characterized in that first a reference gas, such as helium, is added to the exhaust gas, specifically, at a point in time which precedes the taking of a sample to a sufficient extent, with respect to flow, that the reference gas and the exhaust gas have become thoroughly mixed, i.e., a virtually homogeneous distribution has been achieved, and that then a quantitative helium and nitrogen analysis is undertaken, measured with a mass spectrometer, taking into account the amount of helium added, with the following individual determinations: O2, CO, CO2, N2, Ar, He, H2.
As explained in a paper in “Stahl and Eisen 113 (1993), No, 6, p. 56,” more than 20 years ago engineers began using the information about the decarburization process in the LD converter that is contained in the exhaust gas. For 9 of 17 basic oxygen steelmaking plants named in a review through 1980, an exhaust gas measurement was specified for process control of oxygen converter processes and an increased setpoint setting accuracy of 85% to 95% was given for a range of carbon content of ±0.020%. This corresponds to a standard deviation of the error of about 0.014% to 0.010%. In the meantime, exhaust gas measurement has become, as will be explained further—partly with mass spectrometer, partly with individual instruments—a permanent part of most process control systems for LD converters and the oxygen steelmaking processes derived from them. In supplementation of the so-called “static model” for precalculating the process, the exhaust gas measurement, together with the sublance, makes possible the continuous observation and control of the process in a so-called “dynamic model.” The decreasing decarburization rate before the end of treatment indicates when a sublance measurement is to be performed in order to reliably hit not only the target carbon content but also the target temperature with this position determination. The calculation of the slagging of iron, manganese, phosphorus and sulfur on the basis of the oxygen balance can determine the bath composition so exactly that most melts can be directly tapped and alloyed without taking a control sample and awaiting its analysis.
Proceeding from the prior art described above, the objective of the invention is to specify a process which allows dynamic control of the converter process for steelmaking, especially during the blowing of oxygen, with said dynamic control being independent of the specified process model.
This objective is achieved by a method with the aid of a submodel that is independent of the process model and that operates as a process monitor based on exhaust gas analysis, a suitable combination of the constituent signals obtained is used to compute determined values for the critical decarburization time (t′crit) and for the end of O2 blowing (t′EoB), by which the oxygen content predicted by the process model is corrected at the beginning of the process and adjusted to the actual conditions towards the end of the decarburization period, where the critical decarburization time (t′crit) is calculated by:
t=t′crit if CP(t)≧CPP and CP(t)=(N2)^n/(CO)^m
wherein
CCP=limit parameter,
CP=(N2)^n/(CO)^m for converters with a gap and low CO post-combustion,
CP=(N2/CO2)^s for converters without a gap and high CO post-combustion, and the end of O2 blowing (t′EoB) is computed by:
t=t′EoB if EoB(t)≧EoBP1 and EoB=(CO+CO2)−rPC
wherein
PC=CO2/(CO+CO2)100=post-combustion in %
with
CO: carbon monoxide content of the exhaust gas in %
CO2: carbon dioxide content of the exhaust gas in %
and
EoBP1=limit parameter
r=adjustment parameter.
The end of oxygen blowing corresponds to the time at which the gradient of the function “EoB” exceeds a well-defined valence. The constant “r” amplifies the signal and increases the signal sensitivity. The physical configuration of this amplified signal means an end of the CO post-combustion in the converter vessel.
The parameters in the equations for the critical decarburization time and for the end of O2 blowing depend on the converter design and especially on the availability of gap adjustment and need to be adjusted at regular intervals. Expected adjustment ranges are:
n=6-7
m=3-6
s=3-4
r=0-5.
The critical decarburization time corresponds to that time at which the gradient of the function “CP” exceeds a well-defined valence, i.e., the decarburization kinetics change from the state of an oxygen deficit for carbon oxidation to a state of oxygen excess. The physical configuration of this amplified signal means the beginning of reduced decarburization, expressed by a continuous rise in the nitrogen concentration and a drop in the carbon monoxide concentration of the exhaust gas. The state before this time characterizes the virtually constant decarburization rate of the main decarburization phase and is directly proportional to the quantity of blown oxygen. After the critical decarburization time, the decarburization rate is controlled by a reaction potential, which is expressed by a difference between the current carbon concentration and its thermodynamic equilibrium. This decarburization rate exhibits an exponential trend. The reduced decarburization rate leads to a lower reaction potential as a function of the difference between the current carbon concentration and its thermodynamic equilibrium. This involves a function of the current chemical composition of the metal and its temperature. Exponents of the function amplify the signal and increase the signal sensitivity in all signal changes.
For converters with a gap and low CO post-combustion, the gradient of the function “CP” is:
CP=(N2)^n/(CO)^m
with
N2=the nitrogen concentration of the exhaust gas in %.
in the case of nonavailability, N2 can be calculated as
N2=100−O2—CO—CO2
and for converters without a gap and high CO post-combustion:
CP=(N2/CO2)^s
A supporting calculation makes it possible to determine, as a function of the different signal levels of the exhaust gas analysis, the captured carbon content in the metal bath, in which a certain desired final carbon concentration can be achieved by stopping the oxygen blowing. In addition, by combining the critical decarburization time with the end of O2 blowing, a definition is obtained of the time interval assigned to a carbon concentration in the metal bath after the critical decarburization time.
The method can be assigned the following procedure:
If t=t′crit
t:t′crit+Δt→Cf(t′crit+Δt)
t:t′crit+2Δt→Cf(t′crit+2Δt)
t:t′crit+nΔt→Cf(t′crit+nΔt)
wherein
Δt=(tf−t′crit)/n=time interval to be statistically analyzed.
Finally, a supporting calculation of the remaining decarburization time on the basis of the critical carbon concentration is used to obtain an estimate of the remaining oxygen to be blown.
The measurement system used for the method of the invention should be based on at least the three fundamental exhaust gas constituents; oxygen, carbon monoxide and carbon dioxide, with nitrogen and residual gas components completing the gas balance. In a conventional measurement, nitrogen is calculated as the remainder up to 100%; hydrogen and all other components are neglected. When a mass spectrometer is used, all significant components of the exhaust gas become available directly from the measurement. Empirical quantities, which are expressed in analytic functions as specific combinations of the components, track as well-defined deciding factors the behavior of the different trends that are significant for the determination of the critical decarburization time and the end of blowing.
The roughly measured values of quantity and composition of the exhaust gas, which are received from the corresponding instruments of the measurement system that is used, must be partially processed before they are used in balancing calculations to smooth the signal fluctuations, to make sure of the correct relation of the various data points to the corresponding point in time, and to apply some quantitative correction parameters.
The smoothing of the signal fluctuations involves a customary mathematical method that is used with several quantities. For completeness of the parameter relations, the smoothing is given in the following formulas. A smoothing operator X=smooth(X, τ) is introduced as abbreviated notation by definition of the expression as:
wherein
u(k)=an unfiltered series of measured values of a signal in time t with time intervals of Δt, typically a sequence of measured values,
x(k+1)=the corresponding smoothed signal,
τ=an average time interval parameter which determines the degree of smoothing.
In the case of a double smoothing, in which the smoothing procedure is performed twice, the notation is X=smooth2(X, τ).
The exhaust gas measurement system usually yields signals with a scanning time of 0.3-0.5 seconds. These signals serve to form different specific support signals and control signals. The following filtering method can be used to smooth them:
x(k+1)=a x(k)+(1−a)u(k),
where k=0, 1, 2
wherein
x(k)=smoothed signal “x” at time k
x(k+1)=smoothed signal “x” at time k+1
u(k)=measured signal “x” at time k
and with the parameters
a=N/(N+1)aε*[0,1]
1−a=1/(N+1)
wherein
N=the number of signal values scanned during the smoothing procedure.
Due to the delay on the order of 15-20 seconds of the exhaust gas values determined at an interval of one second at a scanning time of 0.3-0.5 seconds, the exhaust gas values must be statistically analyzed for the determination of the critical decarburization time and the time of the end of blowing and their difference.
The method of the invention is characterized by the following coordinated operating procedure:
(A) Process model with statistical process computation and definition of:
(B) Submodel, independent of the process model, with dynamic process monitoring of
The design of an exhaust gas measurement system (
The exhaust gas samples 4 of the converter 1 are analyzed in the measurement system 10 (see
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2009 060 258 | Dec 2009 | DE | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2010/069599 | 12/14/2010 | WO | 00 | 9/21/2012 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2011/076615 | 6/30/2011 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4148629 | Kammerhofer | Apr 1979 | A |
| 4251270 | Hoshi et al. | Feb 1981 | A |
| 4416691 | Narita et al. | Nov 1983 | A |
| 6093235 | Reichel | Jul 2000 | A |
| 6190435 | Miyamoto et al. | Feb 2001 | B1 |
| 6468467 | Miyamoto et al. | Oct 2002 | B1 |
| 6805724 | Shver | Oct 2004 | B2 |
| 8092572 | Matschullat | Jan 2012 | B2 |
| 20100192672 | Reichel | Aug 2010 | A1 |
| Number | Date | Country |
|---|---|---|
| 2836694 | Mar 1979 | DE |
| 2839316 | Mar 1979 | DE |
| 0572848 X | Dec 1993 | EP |
| 2003929 | Mar 1979 | GB |
| 1079314 | Mar 1989 | JP |
| 2008049673 | May 2008 | WO |
| 2009030192 | Mar 2009 | WO |
| Entry |
|---|
| STAHL—UND—EISEN—Jun. 14, 1993 |
| Number | Date | Country | |
|---|---|---|---|
| 20130018508 A1 | Jan 2013 | US |
