TOTEM Co Ltd.

 

Effect of the charging pattern upon blast furnace performance
when a Bell-less rotary charging unit is used

 

Dr. (Tech.) B.M. Boranbaev, V.N.Vakulin, Y.M. Glazer
TOTEM Co.Ltd.
3-rd Mytischinskaya str., 16, block 60, Moscow, 129626, Russia
Tel.: +7(495)6316209
Fax: +7(495)6314777
E-mail:totem@totem-engineering.com

 

P.S. Rana, F.Sarcar, Manish K, Jha (JSPL)
Tel.: +91(7762)227001-5
Fax: +91(7762)227021-22
E-mail:psrana@jspl.com

 

Key words: Blast furnace, Performance, Rotary charging unit, Burden distribution,
Mathematical model, Interface, Criteria

 

INTRODUCTION

In August and October 2006 two new blast furnaces of 1681 m3 volume each were put into service at JSW and JSPL steel plants (Jindal Group) in India. For the first time world-over the Bell-less rotary charging units (BRCU), designed by TOTEM Co. Ltd, were installed on these furnaces. In this paper the operating results obtained at the initial stage of operation of blast furnace No.2 JSPL using BRCU are discussed.

DISIGHN AND OPERATING FEATURES OF THE BELL-LESS ROTARY CHARGING UNIT

The general view and main dimensions of BRCU are shown in Figure 1. BRCU consists of a receiving funnel, upper and lower banks of valves, transfer hopper, burden gate with compensator, angular gearbox, central gearbox, rotor and remote rotor drive. The central gearbox has one cylindrical gearing, wherein the drive gear is fastened on the outgoing shaft of the angular gearbox, and the inner ring of the driven gear mates the slewing ring of the support bearing. Inside the central gearbox enclosure there are two sealing rings, on the bottom part of it there is a quiescent cooling system. The upper and lower banks of valves are of similar design, they consist of burden gates and gas sealing valves. The gas sealing valve delivers a closure having a metal-to-metal contact. There is a provision of possibility to use a disc seat outfit with rubber-to-metal contact. During 16 months of BRCU operation on blast furnace No.2 JSPL there has been no replacement of the gas sealing valves discs. The valves are designed in such a way that a disc and seat can be replaced within 2-3 hours.
BRCU mechanisms performance, their condition and the preset charging patterns are controlled through ACS Rotor, which can be integrated into the existing instrumentation and control of the blast furnace concerned.
The salient features of charging through BRCU, that provide for better distribution of stock as compared with the other known charging apparatuses, are as follows:

    • A multi-layer charging of stock in wide flared flows which enables to average the batch under charging. As one batch of material is being charged, five rotor vanes would lay down up to 40 and more layers of stock.
    • High circular uniformity which is achieved thanks to splitting the stream of stock into five equal slewing flows.
    • A soft discharge of stock on the burden surface, which would not destroy the profile of the preceding batch of material. It is possible thanks to the fact that the stream of stock is split into 5 flows and discharged in wide thin bands. Such a soft discharge makes it possible to forecast better the distribution of material later-wise.
    • A flexible adjustment of radial distribution of material, which is delivered by varying the rotor speed.

     

    Figure 1    General view of BRCU

    Figure 1 General view of BRCU

    PERFORMANCE INDICES OF BLAST FURNACE No.2 JSPL

    In Table 1 there are indices of heats seen during the periods of comparatively steady runs of the furnace. The necessity to refer to these periods is justified by the fact that the quality of burden materials and other heat conditions, particularly in the first months after the start-up of the furnace. It is worthy to note that due to the excessive availability of hot metal in the shop, the furnace productivity was somehow curbed. For this reason also it was difficult to establish optimum parameter for heats. However, despite such constraints, the data as they are shown in Table 1, testify to a quite good indices (for conditions in India) of BF No.2 JSPL performance at the initial stage of operation.

    Table 1 Blast furnace No.2 JSPL performance indices

    Indices

    Periods of 2007

    1

    2

    3

    4

    5

    6

    7-13.06

    25-29.09

    8-14.10

    26-31.10

    24-30.11

    9-13.12

    7 days

    5 days

    7 days

    6 days

    6 days

    5 days

    1. Productivity, t/day

    3241.4

    3332.2

    3422.2

    3609.4

    3391.5

    3322.0

    2. Specific prod-ty (on work. vol.), t/m3day

    2.22

    2.28

    2.34

    2.47

    2.32

    2.27

    3. Coke rate (dry), kg/t

    485.5

    493.4

    480.0

    488.7

    435.0

    415.4

    4. Coke nut rate, kg/t

    24.1

    18.3

    22.6

    23.2

    23.8

    23.0

    5. PCI rate, kg/t

    35.54

    47.16

    43.65

    32.16

    67.85

    74.60

    6. Total fuel rate, kg/t

    545.1

    558.8

    546.2

    544.1

    526.7

    513.0

     

    7. Carbon rate, kg/t

    457.8

    479.8

    468.5

    464.5

    448.9

    433.7

     

    8. Carbon intensity rate, t/m3day

    0.883

    0.951

    0.954

    0.997

    0.906

    0.857

     

    9. Ore rate (dry), kg/t

    497.1

    524.0

    483.3

    475.0

    460.0

    472.0

     

    10. Sinter rate kg/t,

    1174.3

    1094.0

    1184.3

    1170.0

    1202.0

    1168.0

     

    11. Raw flux rate, kg/t

    1.7

    8.6

    0

    4.2

    1.7

    2.7

     

    12. Quartzite rate, kg/t

    23.0

    20.8

    24.0

    19.3

    19.1

    25.2

     

    13. Fe in ore part of burden, %

    56.66

    57.03

    56.67

    56.68

    56.36

    56.83

     

    14. Ore intensity, t/m3day

    3.27

    3.27

    3.44

    3.58

    3.40

    3.30

     

    15. Blast rate (calculated), m3/min

    2872

    3195

    3145

    3240

    2885

    2737

     

    16. Blast temperature, ?C

    1109

    1122

    1115

    1119

    1153

    1190

     

    17. Blast pressure, bar

    2.64

    2.61

    2.58

    2.58

    2.53

    2.36

     

    18. Top pressure, bar

    1.33

    1.17

    1.20

    1.20

    1.15

    1.07

     

    19. O2 in blast, %

    23.7

    24.0

    24.4

    24.1

    23.3

    23.2

     

    20. Blast moisture, g/m3

    42.9

    41.8

    42.8

    40.8

    44.6

    45.0

     

    21. Slag yield, kg/t

    337

    338

    333

    327

    323

    321

     

    22. Ore/coke ratio, t/t

    3.49

    3.39

    3.52

    3.41

    3.87

    4.01

     

    23. Top temperature, ?C

    ND

    140

    98

    93

    78

    91

    24. CO utilization rate, %

    45.01

    43.91

    44.74

    44.66

    45.48

    45.93

    25. Si in hot metal, %

    0.55

    0.61

    0.61

    0.60

    0.68

    0.75

    26. Hot metal temperature, ?C

    ND

    1471

    1476

    1482

    1485

    1489

    27. Slag composition, %:

    33.79

    34.49

    34.54

    34.43

    34.50

    34.86

    SiO2

    35.66

    35.98

    35.83

    35.42

    34.91

    35.29

    Al2O3

    18.90

    17.84

    17.91

    18.58

    18.72

    18.87

    MgO

    9.17

    8.97

    8.79

    8.69

    8.99

    8.16

    CaO/SiO2

    0.95

    0.96

    0.96

    0.97

    0.99

    0.99

    CaO+MgO/SiO2

    1.20

    1.21

    1.21

    1.22

    1.25

    1.23

    28. One tap weight, t

    368

    387

    390

    406

    456

    493

    29. Fe in ore, %

    64.31

    63.67

    64.07

    63.59

    63.85

    63.74

    30. Sinter composition, %: Fe

    54.62

    54.85

    54.8

    55.01

    54.47

    54.33

    CaO

    10.32

    9.95

    10.31

    9.86

    10.08

    10.38

    SiO2

    5.23

    5.17

    5.46

    5.25

    5.34

    5.34

    CaO/SiO2

    1.97

    1.92

    1.89

    1.88

    1.89

    1.94

    31. Coke, %: Moisture

    8.21

    9.08

    7.57

    6.76

    7.74

    6.95

    Ash

    12.31

    12.33

    12.46

    12.46

    12.10

    12.39

    Carbon

    87.06

    86.86

    87.08

    86.34

    87.29

    87.00

    CSR

    ND

    66.2

    67

    65.2

    62

    66.4

    CRI

    ND

    24.2

    23.0

    24.8

    28.0

    23.6

    M40

    ND

    86.6

    86.7

    86.6

    84.5

    85.4

    M10

    ND

    5.42

    5.23

    5.49

    5.43

    5.57

    32. PCI, %: Ash

    9.01

    9.96

    10.03

    10.78

    10.40

    11.53

    Carbon

    76.44

    75.78

    75.73

    73.33

    72.20

    71.12

    Volatiles

    14.56

    14.26

    14.24

    16.10

    17.40

    17.35


    MATHEMATICAL MODEL OF BURDEN RADIAL DISTRIBUTION

    The above-said features of the rotary charging practice were demonstrate during the filling of the furnace before blow-in 1, 2.Regarding these features, a mathematical simulation model has been worked out to visualize the formation of burden layers in the upper part of the blast furnace shaft, as BRCU is used. Figure 2 shows the main screen (interface) of the charging mathematical model.

    Figure 2Charging mathematical model interface

    Figure 2 Charging mathematical model interface

    The model consists of two main parts.

    Part 1 (upper diagram on the right) simulates the distribution of individual batches of material as they are discharged onto a horizontal surface. It takes account of the impact exerted by the rotor operation mode, type of material, material flow rate per time unit and stock line level. Part 1 is designed on the basis of physical tests using the method of mathematical experiment design. The model computes the profile ordinate for each batch (material layer thickness) at ten points on the flat serfice.

    Part 2 (upper diagram on the left) describes the behavior of bulk material in the blast furnace top as charging goes on in certain cycles according to the preset program. In this Part, along with the distribution parameters for each batch, that are computed in Part 1, such effects upon the formation of burden layers are taken into account as burden descent rate along the furnace radius, angles of repose of bulk material, overspill of material during discharge and descent.
    Thus, it is a physical experiment which is used as a base of the mathematical model, which excludes the necessity to refer to many allowances that may affect the accuracy of the model.

    In the middle sector of the main screen there are graphs showing the variations of ore/coke ratio in charged material along the furnace radius, as they are computed for one charging cycle.

    To construct the mathematical model, the top span was conventionally split into 6 equal-in-area annular zones. To make the model more practical, three nondimensional criteria of radial distribution in the given cycle of batches have been formulated. These criteria, as we believe, enable to assess the impact of the burden charging parameters upon the furnace performance and to optimize the performance of the charging apparatus. Below are there the ratios, used to compute the above-said criteria.

    CRD1 = ---------------------------------------------

     mean ore/coke ratio in annular zones 4?6
    mean ore/coke ratio in annular zones 1?3

    CRD2 = -----------------------------------------------

     mean ore/coke ratio in annular zones 5 and 6
    mean ore/coke ratio in annular zone 1 (center)

    CRD3 = -----------------------------------------------

     mean ore/coke ratio in annular zones 2?4
    mean ore/coke ratio in annular zone 1 (center)

    To assess the degree of influence by the charging system, which is characterized by the radial distribution criterion CRD1, upon the furnace performance, it was decided to make use of such indicators as CO utilization ratio and specific fuel rate that were observed in the periods of the furnace steady runs. The visualization of furnace charging and CRDI determination with the help of the mathematical model were carried out on the basis of one day data from each period when the most salient system of charging took place. The computed values of CRD1, actual values of CO utilization ratio and specific fuel rate in the given 24 hours are shown in Table 2.

    Table 2 Impact of criterion CRD1 upon the blast furnace performance indices


    Date

    08.06.2007

    26.09.2007

    12.10.2007

    28.10.2007

    28.11.2007

    11.12.2007

    CRD1

    0.972

    0.797

    1.116

    1.124

    1.609

    N.D.

    CO utiliz.,%

    45.01

    43,91

    44.74

    45.66

    45.48

    45.93

    Fuel rate, kg/t

    545.1

    558.8

    546.2

    544.1

    526.7

    513.0

    As one could expect, there was quite a close correlation between the specific fuel rate and CO utilization rate (see graph in Figure 3). It means that the impact of the charging system upon the specific fuel rate one can successfully assess and predict by variations on CO utilization rate, which is determined by top gas analysis.

    Figure3Correlation between the specific fuel rate and CO utilization rate

    Figure 3 Correlation between the specific fuel rate and CO utilization rate

     

    Fig 4 shows the correlation graph between CO utilization rate and CRD1 criterion value. As can be seen from it, there is definitely a conspicuous trend between these parameters, which can be expressed by polynomial 2-d power. However, at this stage this presented correlation can be considered only as a trend and cannot be used for quantitative assessments.

    It should be noted that to plot this correlation the mean daily data were taken without having them filtrated and synchronized in time. Obviously, while working out specific procedures of monitoring and synchronization of data it will be possible to obtain more clear-cut correlations, useful for automatic system to control the blast furnace charging process.

    A possible development of such a system in future is our main goal of our research work. The preliminary results covered in this paper prove that further efforts are worth making. In the presented analysis we have not used so far two other criteria for the radial distribution of stock i.e. CRD2 and CRD3, but they might become useful in future for expanding the research and development work.

    Figure 4Relationship between CRD1 criterion and CO utilization rate

    Figure 4 Relationship between CRD1 criterion and CO utilization rate

     

    ANALYSIS OF THE CHARGING PATTERNS

    As the analysis of the charging systems (patterns) already realized in BF No.2 shows, in the majority of them there were cycles to charge 8 and more batches of material. It is quite a big volume of burden and in the course of it being charged, to change the radial distribution and correspondingly to vary CRD criteria may take a lot of time. In case of many batches in the cycle the whole system would become sluggish and hence poorly controlled. Besides, in big charging cycles there is a possibility of fluctuations in the thermal conditions of the furnace caused by the cyclic and non-uniform nature itself in the burden column structure within the volume of material in one cycle.

    Therefore, we have developed several charging systems, that consist of two batches only, that is , coke and ore. In this charging systems a batch of coke is loaded into the furnace center (ring 1), and a batch of all ore-bearing components is loaded upon one or two rings within the range from ring 1 to ring 4. To optimize the charging system in such a cycle means to keep shifting the ore bearing part of burden along the furnace radius until the best radial distribution of ore/coke ratio has been achieved. Such an approach will substantially simplify the task of optimization of the radial distribution of burden in the blast furnace top.

    Table 3 shows the parameters of two-batch charging cycles tested in the mathematical model.

    Table 3 Two-batches charging patterns


    No.No.
    of charging patterns

    Coke, %

    Ore, %
    (iron bearing materials)

    Ring 1
    (center)

    Ring 2

    Ring 3

    Ring 4

    1

    100

    100

     

     

    2

    100

    50

    50

     

    3

    100

     

    100

     

    4

    100

     

    50

    50

    5

    100

     

     

    100

    Fig.5 contains curves 1 ? 5, that show how ore/coke ratio would change along the furnace radius when the ore bearing components of burden are being charged by methods indicated in Table 3 (the curve numbers correspond to the numbers of charging patterns of each cycle). As can be seen in the graphs, while a batch of ore bearing component of burden is shifting to periphery, one can see adequate changes in the distribution of ore/coke ratio along the furnace radius.

    Figure 5 Distribution of ore/coke ratios along the furnace radius at various two-batch charging patterns

    Figure 5 Distribution of ore/coke ratios along the furnace radius at various two-batch charging patterns

     

    The visualization of the outward versions of the radial distribution of burden that are practiced under the charging system No.1 (100% of ore bearing component is dumped onto ring 2) and No.5 (100% of ore bearing component is dumped onto ring4) are shown in Fig.6.

    Charging pattern No.1Charging pattern No.2

    Figure 6Visualization of two-batch charging cycles

    Figure 6 Visualization of two-batch charging cycles

     

    Thus, as the investigation into the radial distribution of burden through a rotary charging apparatus with the help of the mathematical model shows, the practice of using charging systems (cycles), consisting of two batches of stock only, i.e. coke and ore, enables to promtly affect the radial distribution of ore/coke ratio with the aim of optimizing it and improving the heat indices. At the same time any radial distribution of ore/coke ratio can be formally characterized with the help of the above-said criteria CRD1, CRD2, CRD3.

     

    As an example, a curve is given in Fig.7, showing variations in CRD1 criterion as ore bearing components are shifting towards periphery under the two-batch charging systems, mentioned in Table 3. The graph demonstrates a close link between this criterion and variations in the burden column structure, which is shaped up in the upper part of the blast furnace shaft.

    Figure 7Variations in CRD1 criterion as ore bearing components of burden are shifting from the center to the periphery

    Figure 7 Variations in CRD1 criterion as ore bearing components of burden are shifting from the center to the periphery

     

    SUMMARY

    1. It should be noted that the possibility to create the simulating mathematical model of rotary charging was preconditioned by BRCU design features, that enabled to practice a multi-flow multi-layer distribution of burden material throughout the top span with high circular uniformity, soft descend of each batch into the furnace and flexible control of the radial distribution of ore/coke ratios.
    2. As the analysis of the production indices achieved in several periods when blast furnace No2 at JSPL happened to run steadily under BRCU, shows, there is a possibility to create ACS for the furnace rotary charging technology on the basis of the suggested criteria of the burden radial distribution.
    3. It is felt that addition research is required to assess the efficiency of chaging with minium number of batches in the cycle.

    REFERENSES

    1. B.M. Boranbaev, Yu.M.Glazet, V.N.Vakulin, Dr.B.N.Singh, T.K.Naha, A.Singh,
    "Distribution of burden with the help of rotary charging apparatus",
    6th Steelmaking Conference IAS, Argentina, 6-8.11.2007
    2. T.K. Naha, B.M.Boranbaev, G.J. Gravemaker,
    High performance operation of the new Jindal South West blast furnace with a bell-less rotary charging unit BRCU XXXVII Ironmaking and Raw Marerials Seminar / VIII Brazilian Symposium on Iron Ore,
    September 18-21, Salvador-Ba-Brazil

     

    Up

 

TOTEM Co Ltd.

""