7/25/2019 Stelmor

1/5

Development and application of online Stelmor Controlled Cooling System

Yu Wan-Hua a, * , Chen Shao-Hui b , Kuang Yong-Hai b , Cao Kai-Chao ba School of Materials Science and Engineering, University of Science and Technology Beijing, 30 XueYuan Road, Beijing 100083, PR Chinab Jiang-Su Sha-Steel Group Co., Ltd., Zhang-Jia-Gang City 215625, PR China

a r t i c l e i n f o

Article history:Received 3 March 2008Accepted 3 March 2009Available online 25 March 2009

Keywords:Controlled coolingModelingStelmorRod productionQuality prediction

a b s t r a c t

An online Stelmor Controlled Cooling System (SCCS) has beendeveloped successfully for the Stelmor pro-duction line, which can communicate with the material ow management system and Program LogicControl System (PLCs) automatically through local network. This online model adopts Implicit Finite Dif-ference Time Domain (FDTD) method to calculate temperature evolution and phase transformation dur-ing the production process and predicts nal properties. As Continuous Cooling Temperature(CCT) curvesof various steels can be coupled in the model, it can predict the latent heat rise and range of phase trans-formation for various steels, which can provide direct guidance for new steel development and optimi-zation of present Stelmor cooling process. This unique online system has been installed in threeStelmor production lines at present with good results.

Crown Copyright 2009 Published by Elsevier Ltd. All rights reserved.

1. Introduction

With increasing tough competition in the steel industry, how

to develop new steel products and stabilize quality of presentproducts becomes the major concern for steel producers. Pushingthe mechanical properties of rod wire closer to its technical lim-its, the demand on more reliable predictive control technique forthe cooling process in Stelmor production line increasescontinuously.

Stelmor is the most popular controlled cooling process to pro-duce the steel wire due to its fast production speed and homoge-neous mechanical properties along the length of wire coil. InStelmor process as shown in Fig. 1 , a rod wire with temperatureabove 1000 C coming from the nishing mill quickly passesthrough several water tanks to the laying head at a specic tem-perature to form into loops, depositingon to a conveyor in an over-lapping pattern, the specic cooling rate is achieved by opening of a series of fans below.

The nal mechanical properties depend mainly on the chemicalcomposition and the cooling rate before the phase transformationfor high carbon steel [1,2,6] . As the cooling rate and phase transfor-mation cannot be observed directly during production, it is ur-gently required to develop an online model to predict the nalmechanical properties and phase transformation. Although thereare several research reports in this eld [36] , the online qualityprediction model for Stelmor process has not found reported yet.

Adopting Implicit Finite Difference Time Domain (FDTD) method,an online controlled cooling model was developed and installedto monitor the real production process. This paper focuses on

introduction of its basic theory and control method of the onlinesystem, which is helpful for stabilization of the product quality.At present, this online model SCCS has been installed in three Stel-mor production lines with satisfactory performance.

2. Mathematical model

2.1. Thermal model

In the Stelmor production line, axial heat conduction can be ig-nored because of an innitely long steel rod moving at high speed,the model can be formulated to solve 1D heat conduction based onfollowing assumptions: (1) radial symmetry; (2) uniform circularcross-section; and (3) uniform initial temperature, which is brieyclose to reality.

Basic equation to solve the heat ow within the rod isfollowing:

@ @ r

k @ T

@ r kr @ T @ r g T q C p @ T @ t 1Note that g (T ) is the volumetric rate of heat generation within

the rod due to phase transformation, q material density, C p theheat capacity and k the thermal conductivity.

In order to reduce loss of material message during calculation,real experimental data q , k, and C p can be input directly in theinterface of model as shown in Figs. 2 and 3, compared with tradi-tional regression method.

1359-4311/$ - see front matter Crown Copyright 2009 Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.applthermaleng.2009.03.012

* Corresponding author. Tel.: +86 010 62332572, +86 13381186768; fax: +86 1082381466.

E-mail addresses: Wanhua.Yu@mater.ustb.edu.cn , ustbywh@sina.com (W.-H.Yu).

Applied Thermal Engineering 29 (2009) 29492953

Contents lists available at ScienceDirect

Applied Thermal Engineering

j o u rn a l home page : www.e l s ev i e r. c om/ loca t e / ap the rmeng

mailto:Wanhua.Yu@mater.ustb.edu.cnmailto:ustbywh@sina.comhttp://www.sciencedirect.com/science/journal/13594311http://www.elsevier.com/locate/apthermenghttp://www.elsevier.com/locate/apthermenghttp://www.sciencedirect.com/science/journal/13594311mailto:ustbywh@sina.commailto:Wanhua.Yu@mater.ustb.edu.cn

7/25/2019 Stelmor

2/5

The FDTD CrankNicolson method has been adopted to solveabove equation as it is fast and unconditionally stable, which canmeet the requirement of online modeling.

In order to maintain balance between speed and accuracy of themodel, 20 nodes has been selected along radial direction after trial

and error. In general, large number of nodes is helpful to guaranteeaccuracy of the model, but it will cause the slow speed of the mod-el, which is unacceptable for the online model, as in real produc-tion, one rod wire can pass the laying head in no more than 2 s.

The following boundary conditions have been applied:

At the centerline

t > 0; r 0; k@ T @ r

0 2

At the rod surface

t > 0; r r 0 ; k@ T @ r

hT r 0 T a 3

where the initial condition is:

t 0; 0 6 r 6 r 0 ; T T in

where t is time in second, r 0 the radius of rod wire, h the heat trans-fer coefcient. T r 0 rod surface temperature, T a air or water temper-ature surround.

When the online model is running, T in is the measured tem-perature coming from the pyrometer installed after the nishingmill.

The model divides the Stelmor line into several stages, eachwith its constant heat transfer coefcient h value, which can beself-adapted to match real measurements fromseveral pyrometers

installed in the production line. In one Stelmor production line inSha-Steel company, there are 14 fan machines put below in se-quence to control the cooling rate of steel rod by opening its fanvolume. A constant h is assigned to each fan machine controlledstage, which can be self-adapted according to the following self-developed equation:

hnew hold T C T air =T M T air 4

where hold is the original heat transfer coefcient, T C the model pre-dicted temperature at one pyrometer position on the fan coolingproduction, T M the measured temperature from the speciedpyrometer, and T air the air temperature.

The computer model was checked by comparison between theFDTD model and commercial FEM model FEMLAB under the exact

same conditions, Fig. 4 shows the exact same results from bothmodels at different condition and proves the accuracy of the FDTDmodel.

As an online model, the SCCS model needs to communicate withthe material ow management system and PLC automaticallythrough local network continuously. The model needs to inputtwo groups of data: one is the real temperature from onlinepyrometers, another is the basic wire rod information. Eightpyrometers have been installed in the production line to providereal temperature data for the model to calculate and compare.

Fig. 4. Comparison of FDTD and FEM modeling result.

Fig. 1. Layout of Stelmor machine.

0

10

20

30

40

50

60

70

0 200 400 600 800 1000 1200

Temperature/ oC

K / J / m / s e c

/ o C

Fig. 3. The steels thermal conductivity change with temperature.

Temperature/ oC

C p

/ J / k g

o C

Fig. 2. The steels specic heat change with temperature.

2950 W.-H. Yu et al. / Applied Thermal Engineering 29 (2009) 29492953

7/25/2019 Stelmor

3/5

These temperature data is rst detected by online pyrometers, di-rectly passed to specic PLC, then saved to Oracle database in thedatabase sever by the model designed by Intouch software, nallythe model can read these data from the database at every 300 ms.another group of data that the model also needs to know is basicinformationof slab, such as its composition and steel serial numberand rod diameter. These data are delivered to the specic databaseonce slab comes out from the reheating furnace.

2.2. Phase transformation model

How to model the phase transformationprocess is critical to thesuccess of the model as the various cooling rates during phasetransformation can determine the nal microstructure. The modelsolved this problem by coupling of heat transfer analysis withphase transformation and microstructure changes, all unied inwhat is already called Microstructural Engineering.

Phase transformation process is described by following Avramiequation [7] :

X 1 exp bT t nT 5where X is the transformation fraction, b(T ), n(T ) are parametersthat vary with temperature, steel composition and austenite grainsize. User can input these parameters in the model directly basedon their experimental data.

One of the important parameters that need to determine isstarting temperature of austenite to pearlite transformation. Themodel can provide user the interface to input CCT experimentaldata of specic steels, then model can internally calculate the tem-perature according to the following relation [8] :

T AV T A1 a dT

dt m

6

where a and m can be regressively obtained from CCT gure. T A1 is

theholding temperature during CCT experiment which canbe inputin the model directly.

The additivity principle was adopted to calculate the heat andamount of phase transformation. During continuous cooling stage,at every timetemperature stage, previous phase change amountcan be transferred to the corresponding virtual time [7] :

t 0 ln 1

1 X i j 1

bT i j24

35

1

n T i j 7

At step j, amount of transformation can by described as:

X i j 1 exp bT i jt

0 D t nT i jh i 8

During this time period, amount of transformation can be de-scribed as:

D X i X i j X i j 1 9

Phase change heat ( g (T )) can be calculated based on followingequation [6] :

g T q H T D X D t

where H (T ) is the volumetric rate of heat generation within the roddue to phase transformation. For high carbon steel, it can calculatedbased on following formulas:

D H c -P 70621 225 :23 T 0:3469 T 2 6:755 10 5 T 3 ; T > 500 C

D H c -b 92 :0 kJ=kgD H c -M 82 :6 kJ=kg

10

2.3. Property prediction model

With the development of microstructural evolution models, it isnow becoming possible to predict nal rolled product microstruc-tures with increasing condence [711] . These predicted micro-structures can then be combined with the structurepropertyrelationship models to calculate mechanical properties. There area large number of factors that contribute to the strength of a steel,such as high carbon steel, according to the equation by Mclvor andcoworker [1] , there are two major factors that affect the nal ulti-mate tensile strength (UTS) value: rst is the cooling rate ( C R) be-fore phase transformation and second is the chemical composition,which can be described as:

UTS 267log C R 293 1029 % C 152 % Si 210 % Mn

232 % Cr 5244 % N f 442 % P 0:5 11

It needs to mention that at the paper put forward by Mclvor andco-worker [1] , C R means the cooling rate at 700 C, which brieyrefer to the cooling rate before phase transformation, as for highcarbon steel, phase transformation occurs just after 700 C. In theCSSC model, C R is modied to refer to the cooling rate in the region

from 700 C to start temperature from austenite to pearlite trans-formation which can be determined from the above equation [4] .It can be seen that high cooling rate before the transformation

can increase UTS correspondingly and strong inuence of nitrogenand phosphorus on UTS. Although the presence of elements likephosphorus, manganese, chromium and silicon are quite usefulto increase UTS, they also bring about low ductility of steel, more-over, increase the risk of precipitation of metastable phases likebainite and/or martensite [1] . As C R can be calculated directly bythe thermal model, chemical composition of specic slab can beknown immediately from the local material management system,it is possible to predict the nal mechanical properties duringproduction.

Yield tensile strength (YTS) is calculated based on followingrelation:

YTS A UTS 12

where A is constant and selected to be 0.7 in the model.

3. Results and discussion

3.1. Stability of the online model

It is required to make the model online one, which is a challengefor us as no mature experience can be learnt. There are two basicrequirements for the online model: stability and accuracy. It is ex-pected that the online model canrun continuously without stop, atthe same time, the output result is accurate enough to provide pro-duction guidance. First question is how to get the related parame-ters directly from other sources such as several online pyrometersand related slab parameters. After trial and error, a suitable solu-tion has been found as shown in Fig. 5 .

At present, three production lines have adopted this method topass data to SCCS model without any problem, SCCS model can runin cycles of 300 ms. In each cycle, the SCCS model calculates againthe temperature model (including the phase transformation andmechanical properties) for every rod point along the length of the cooling section, so that at any time the temporal temperature

Pyrometer PLC Database sever SCCS Station

Fig. 5. Temperature data pass process.

W.-H. Yu et al. / Applied Thermal Engineering 29 (2009) 29492953 2951

7/25/2019 Stelmor

4/5

course of each rod is known and can be controlled. All predicteddata is saved in Oracle database and can be checked later.

3.2. Accuracy of the thermal model

After 3 years of hard work and repeated trials, the online modelhas been installed in Stelmor production lines in Sha-Steel com-pany at Jiang-Su Province. After several version modication, theonline model can run normally. Eight online pyrometers have beeninstalled to feed the model the real temperature. One is installedafter the nishing mill, which provides the starting temperaturefor the online model. Next one is installed at the laying head, whichchecks the effect of several water boxes. Other six are installed atthe fan cooling stage with each section interval, which monitortemperature evolution in the production line and compare withmodeling results. The online model will calculate the temperaturesat every point, then automatically adapted heat transfer coefcient(HTC) at every section according to comparison between the realmeasurement and modeling result, all these can be nished in cy-cles of 300 ms, which can meet speed requirement for the onlinemodel. It is observed that calculated values of HTC and actual onesmeasured in operating rod mills can be considerably different, theonline model can easily study its difference and modify its HTC va-lue correspondingly, which is helpful for accurate prediction of phase transformation and nal mechanical property. Fig. 6 showsthe comparison between the real measurement and modelingresult.

Fig. 6 shows the close match between the real measurementand the modeling result with no more than 20 C difference. Theeffect of phase transformation on the temperature evolution canbe seen clearly, whichcauses slight rise during phase change stage,then drops gradually. The model can predict duration of phasetransformation, which provides guidance for technicians to adjustthe opening of fans. In general, it is expected that phase transfor-mation can be carried out at the same temperature within theshort time (such as no more than 10 s for 8 mm steel wire), whichcan guarantee the uniform microstructure, just like patentingtreatment.

3.3. Mechanical property

Mechanical property prediction is a major concern for the com-pany, if the model is accurate enough, the company will reduce thesample checking number which can save time and money. In Sha-Steel, SWRH82B steel (C 0.81/Si 0.25/Mn 0.76/Cr 0.15) is selectedto verify the model as this steel has very strict production require-ment. At rst, Eq. (11) has been coupled in the SCCS model to pre-dict the nal tensile strength, after checking with more than 400samples, it was found that model prediction values were relativelyhigher than the real measurement on the whole as shown in Fig. 7 .

Statistics analysis show that average value gap between both was30 MPa, with standard deviation 37 MPa. Although both show thesimilar trend on the whole, it is expected that accuracy of the mod-

el could be improved further.In order to improve accuracy of the model further, the multipleregression method was adopted to modify the coefcient amongchemical composition, tensile strength and cooling rate, a newequation is put forward as shown:

UTS 764 28log C R 93 % C 32 % Si 48 % Mn 1723 % Cr

21 % N 83 % P 0:5

13

It can be seen that frameworks of Eq. (11) and Eq. (13) are thesame, UTS is related to cooling rate and chemical composition, onlydifference is coefcient of various parameters, which can be ob-tained from real production data. Based on Eq. (13) , accuracy of the model has been improved considerably, Statistics analysis

show that average value gap between both is 0.39 MPa, with stan-dard deviation 16.8 MPa. It can be seen from Fig. 7 that both are inclose agreement, which proves reliance of the online model to pro-vide the production guidance. Comparing with Eq. (11) , it can beseen that except Cr element, all other factors such as cooling rateand C element have less signicant effect on the nal UTS. Thisdoes not mean that Eq. (11) is not accurate, as Eq. (13) comes fromthe one steel sample, which has not too much uctuation in chem-ical composition, this will cause regression deviation.

As for other high carbon steels with different composition,treatment is similar as SWRH82B. At rst, Eq. (11) is adopted topredict its UTS, compare with real measurement, then modify itsrelated coefcients based on the multi-regression result (seeFig. 8 ).

Fig. 6. Comparison between the real measurement and modeling result.

1000

1050

1100

1150

1200

1250

1300

1350

1400

1 51 101 151 201 251 301 351 401

Number of SWRH82B sample

U T S / M P a

Real Measurement Model Prediction

Fig. 7. Comparison between the real measurement and modeling result based onEq. (11) .

1080

1100

1120

1140

1160

1180

1200

1220

1240

1260

1 51 101 151 201 251 301 351 401

Number of SWRH82B sample

U T S / M P a

Real measurement Model prediction

Fig. 8. Comparison between the real measurement and modeling result based onEq. (13) .

2952 W.-H. Yu et al. / Applied Thermal Engineering 29 (2009) 29492953

7/25/2019 Stelmor

5/5

3.4. Advantage of the online model in the production

It needs to point out that variation of UTS during mass pro-duction is inevitable as many factors such as steel compositionand cooling rate cannot maintain unchanged during whole pro-cession. Reasonable online prediction can help to alert workersto pay attention to some key factors such as temperatures atvarious point and cooling rates. At present, the online SCCSmodel has installed in one Stelmor production for more thanone year, which shows great advantage of model during pro-duction. Before installation of CSSC model, pre-xed policywas adopted for every steel during production, workers couldntdirectly observe the phase transformation range and tempera-tures at various point, sometime, phase transformation couldshift outside the controlled area, which could cause seriousquality problem, such as nal microstructure to be martensiteor coarsen pearlite. Workers can do nothing only after nalchecking. After installation of the model, workers can directlyobserve the production process from user-friendly interface of the model, adjust the opening volume of various fans accordingto requirement, control the temperatures at various points inthe reasonable range, and make phase transformation occur inthe exact area. According to statistics from this production line,rate of nal steel product with ne pearlite with grade 2.5 orbelow has been increased from 87.93% to 93.54% due to instal-lation of the CSSC model, at the same time, cases of qualityargument has greatly reduced to give users enough condencein its nal product.

4. Conclusions

An unique online quality predictionsystem for the Stelmor con-trolled cooling line was developed, following conclusions can beobtained based on the working experience:

1. The comprehensive model the for Stelmor controlled coolingline consists three parts: the thermal model is obtained byFDTD method, phase change model is obtained by solvingAvrami equation based on the experimental CCT data, andmechanical property model is physical model based onregressing the production data, three parts are coupledinternally.

2. This online model is open to get the production informationduring production, which can communicate with the materialow management system and Program Logic Control System(PLCs) automatically through local network.

3. No model is perfect and denitely accurate, so self-adaptedfunction was designed for this online model to adjust the pre-dicted temperature along the production line based on online

pyrometers andnal properties based on nal checking. Presentresults prove that it is possible to predict the nal mechanicalproperties with the help of this online model with standarddeviation 16.8 MPa, which meet the requirement.

After repeated modication, this online model has been put intoproduction for more than 2 years, which can greatly reduce the la-bor work and improve the stability of product quality.

References

[1] S. Jaiswal, I.D. Mclvor, Microalloyed high carbon steel rod, Ironmaking andSteelmaking 16 (1) (1989) 4954.

[2] I.D. Mclvor, Microalloyed lowcarbon steel rod, Ironmaking andSteelmaking 16(1) (1989) 5562.

[3] Prakash K. Agarwal, J.K. Brimacombe, Mathematical model of heat ow and

austenitepearlite transformation in eutectoid carbon steel rods for wire,Metallurgical Transactions B 12B (1981) 121133.

[4] R.D. Morales, A.G. Lopz, I.M. Olivares, Mathematical simulation of Stelmorprocess, Ironmaking and Steelmaking 18 (2) (1991) 128138.

[5] R.D. Morales, A.G. Lopz, I.M. Olivares, Heat transfer analysis during waterspray cooling of steel rods, ISIJ International 30 (1) (1990) 4857.

[6] Ettore Anelli, Application of mathematical modelling to hot rolling andcontrolled cooling of wire rods and bars, ISIJ International 32 (3) (1992)440449.

[7] E.B. Hawbolt, B. Chau, J.K. Brimacombe, Kinetics of austeniteferrite andaustenitepearlite transformations in a 1025 carbon steel, MetallurgicalTransactions A 16A (1985) 565578.

[8] S. Denis, S. Jstrm, A. Simon, Coupled temperature, stress, phasetransformation calculation model numerical illustration of the internalstresses evolution during cooling of a eutectoid carbon steel cylinder,Metallurgical Transactions A 18A (1987) 12031212.

[9] J. Iyer, J.K. Brimacombe, E.B. Hawbolt, Prediction of the structure andmechanical properties of control-cooled eutectoid steel rods, in: Conf. onMechanical Working and Steel Processing, Chicago 1984, vol. XXII, ISS/AIME,Penn, 1985, p. 47.

[10] A. Kumar, C. McCulloch, E.B. Hawbolt, I.V. Samarasekera, Modelling thermaland microstructural evolution on runout table of hot strip mill, MaterialsScience and Technology 7 (1991) 12111223.

[11] M. Militzer, E.B. Hawbolt, T.R. Meadowcroft, Microstructural model for hotstrip rolling of high-strength low-alloy steels, Metallurgical and MaterialsTransactions A 31A (2000) 12471259.

W.-H. Yu et al. / Applied Thermal Engineering 29 (2009) 29492953 2953