Claims:1. A system for controlling strip temperature in a continuous galvanization annealing furnace, wherein the system comprising:
a continuous galvanization annealing furnace;
a processor, communicatively coupled with continuous galvanization annealing furnace, wherein the processor is configured to
receive a real-time data associated with a first strip in the continuous galvanization annealing furnace, and a second strip about to enter the continuous galvanization annealing furnace, wherein the data comprises one or more of strip temperature, the strip dimension, the strip properties associated with both the first strip and the second strip and furnace zone temperature, fuel flow rate and line running speed;
compute a time required to reach an optimal temperature associated with the second strip, and an optimal fuel flow; and
control the continuous galvanization annealing furnace and thereby the strip temperature based on the computation.
2. The system as claimed in claim 1, wherein one end of first strip is coupled to other end of the second strip.
3. The system as claimed in claim 1, wherein the computation comprises:
calculating a predicted strip temperature profile in immediate future under current operation condition and
comparing the predicted strip temperature profile with a target trajectory of annealing strip temperature.
4. The system as claimed in claim 1, wherein the optimal fuel flow minimize the sum of square of deviation between the predicted strip temperature and target strip temperature, and minimizes the incremental fuel variation.
5. The system as claimed in claim 1, wherein the processor is configured to:
obtain parameters adapted periodically by online recursive learning algorithm to encounter the effect of unknown disturbance and
calculate the optimal fuel flow with predicted strip temperature by new learned parameters.
, Description:FIELD OF THE INVENTION
[0001] The present disclosure generally relates to an annealing. More specifically the
invention relates to a system for controlling strip temperature in continuous annealing
furnace.
BACKGROUND OF THE INVENTION
[0002] In a steel industry, an annealing furnace is used to control the quality of material
and surface properties of a steel strip by means of heat treatment. Cold rolled steel strips
are subjected to a heat treatment called annealing to regain the ductility and forming
properties which are lost during other cold process working. In continuous annealing
process, head end of one coil which is welded with tail end of preceding coil is subjected
to indirect heating in continuous heating furnace which consist of the preheating,
heating and soaking furnace. Depending on the grade of the coils and desired
mechanical property of final product, quality department prescribes temperature profile
continuous annealing furnace and reference trajectory temperature of steel strip along
with its allowed maximum and minimum temperatures in the form of ‘annealing
recipes’.
[0003] A continuous galvanization line (CGL) requires annealing process to be carried
out prior to coating of zinc. In continuous galvanization line (CGL), strip temperature
is measured and controlled at the outlet of the furnace. The strip temperature at the
outlet of the furnace has difficulties in control because of slow strip temperature
dynamics, delay time, change of strip thickness, strip width, or reference temperature.
These complicated system makes it difficult to achieve satisfactory control
performances by conventional control system like proportional, integral, and
differential (PID) controllers. For the smooth and satisfactory control, this is a difficult
task to control temperature of strip by PID controllers, particularly in transient cases
when a welded joint traverses the furnace or when the strip velocity changes. Strip
temperature control has further difficulties by the fact that the strip temperature can
only be monitored by radiation pyrometers located at an exit of heating and socking
zone of furnace (at limited location).
3
SUMMARY OF THE INVENTION
[0004] Before the present system for strip temperature control in continuous annealing
furnace is described, it is to be understood that this application is not limited to a
particular continuous annealing furnace, as there may be multiple possible
embodiments, which are not expressly illustrated in the present disclosures. It is also to
be understood that the terminology used in the description is for the purpose of
describing the particular implementations, versions, or embodiments only, and is not
intended to limit the scope of the present application. This summary is provided to
introduce aspects related to a system for controlling strip temperature in continuous
annealing furnace. This summary is not intended to identify essential features of the
claimed subject matter nor is it intended for use in determining or limiting the scope of
the claimed subject matter.
[0005] In one embodiment, a system for controlling strip temperature in a continuous
galvanization annealing furnace is disclosed. The system comprises a continuous
galvanization annealing furnace and a processor, communicatively coupled with
continuous galvanization annealing furnace. Further the processor is configured to
receive a real-time data associated with a first strip in the continuous galvanization
annealing furnace, and a second strip about to enter the continuous galvanization
annealing furnace, wherein the data comprises one or more of strip temperature, the
strip dimension, the strip properties associated with both the first strip and the second
strip and furnace zone temperature, fuel flow rate and line running speed. Furthermore,
the processor is configured to compute a time required to reach an optimal temperature
associated with the second strip, and an optimal fuel flow and control the continuous
galvanization annealing furnace and thereby the strip temperature based on the
computation.
BRIEF DESCRIPTION OF DRAWINGS
[0006] The foregoing detailed description of embodiments is better understood when
read in conjunction with the appended drawings. For the purpose of illustrating the
present subject matter, an example of construction of the present subject matter is
provided as figures; however, the present subject matter is not limited to the specific
anchorage mechanism disclosed in the document and the figures.
4
[0007] The present subject matter is described in detail with reference to the
accompanying figures. In the figures, the left-most digit(s) of a reference number
identifies the figure in which the reference number first appears. The same numbers are
used throughout the drawings to refer various features of the present subject matter.
[0008] Figure 1 illustrates a schematic representation of Continuous Galvanization
Annealing Furnace, in accordance with the present subject matter.
[0009] Figure 2 illustrates actual operational data sample for model development, in
accordance with the present subject matter.
[0010] Figure 3 illustrates Parameter recursive learning with actual plant data, in
accordance with the present subject matter.
[0011] Figure 4 illustrates Temperature prediction result with adapting model
parameters, in accordance with the present subject matter.
[0012] Figure 5 illustrates a schematic representation of Continuous Galvanization
Annealing Furnace, in accordance with the present subject matter.
[0013] Figure 6 illustrates HMI Screen for CGL2 Model Predictive control, in
accordance with the present subject matter.
DETAILED DESCRIPTION
[0014] Some embodiments of this disclosure, illustrating all its features, will now be
discussed in detail. The words "comprising," "having," "containing," and "including,"
and other forms thereof, are intended to be open ended in that an item or items following
any one of these words is not meant to be an exhaustive listing of such item or items,
or meant to be limited to only the listed item or items. It must also be noted that as used
herein and in the appended claims, the singular forms "a," "an," and "the" include plural
references unless the context clearly dictates otherwise. Although any system for
controlling strip temperature in a continuous galvanization annealing furnace and,
similar or equivalent to those described herein may be used in the practice or testing of
embodiments of the present disclosure, the exemplary, anchorage mechanism is now
described.
5
[0015] Various modifications to the embodiment will be readily apparent to those
skilled in the art and the generic principles herein may be applied to other embodiments.
However, one of ordinary skill in the art will readily recognize that the present
disclosure is not intended to be limited to the embodiments described, but is to be
accorded the widest scope consist in this regard, in a generic sense.
[0016] As discussed above, before the continuous galvanization (CGL), annealing heat
treatment processed where cold rolled coils are heated in continuous pattern according
to a predetermined recipe. The process is required to recover the ductility and forming
behavior of steel which is lost during cold working. Steel comes out of the annealing
process as not only softer but also with preferred microstructure and textural properties.
Coils are welded by head end of one coil with tail end of preceding coil at welding
platform and make a continuous line before the annealing furnace. Coils heated directly
in preheating furnace by hot Hydrogen-Nitrogen mixture gas (HNX) and indirectly in
heating furnace as well as soaking furnace. Radiant tubes are located surrounding to
running strip in heating and soaking furnace. Fuel burns through burner placed inside
of each radiant tube and heated the strip steel indirectly from outside without direct
contact between combustion gas and coils. To facilitate the heat transfer within the
annealing chamber and making inert atmosphere, a circulating gaseous mixture (HNX)
is maintained in it. Fig.1 gives schematic of continuous annealing furnace.
[0017] Heating of strip temperature depends on strip thickness, width, line speed, and
fuel flow rates. The strip annealing temperature is controlled in such a way that the
temperature of annealing furnace of each zone is maintained as per the annealing recipe
provide to operator. At the time of coil setup change (transition of coil), operator must
manually adjust the furnace zone temperature set points. Conventional control system
with such as proportional, integral, and differential (PID) controllers send feedback to
fuel flow control valve to maintained furnace zone temperature. Using classical PID
control approaches it is difficult to achieve the desired control performance, especially
in case of transient operating conditions when a welded joint traverse the furnace or
when the strip velocity changes. In transient operating cases, some length of coil gets
under/over annealing and that is affecting to the quality of product, fuel consumption
and furnace throughput. Therefore, model based predictive control is more suitable tool
6
to obtain desired process behavior which will improve the quality of product and
maximum throughput.
[0018] A mathematical model predicts strip temperature in immediate future by using
current operational data, current and future coil information data given by recipe and
tracking data provide by level1. Compare predicted strip temperature with desired
target strip temperature trajectory and calculate optimal fuel flow set point in future
time step horizon by minimizing the sum of square of deviation between predicted strip
temperature and target temperature. Send the fuel flow set point for next time step to
level1 DCS, which is the command value of total fuel flow rate in annealing furnace.
Total fuel flow then distributed to each zone of the furnace to achieve a desired profile
of the furnace temperature. Control calculation done at every interval time cycle. Model
will ensure the smooth change in fuel flow rate during control action. It will prevent the
violation of input-output constraint. Model parameters adapted periodically with
learning conditions for encounter the effect of unknown disturbance in the process. In
transient cases, model will take a smooth control action at optimal time before next coil
coming to annealing furnace. This prevents coils from getting under annealed or over
annealed with corresponding benefits in quality/productivity. Hence, model based
control is a very relevant solution for efficient operation of Continuous Annealing
Furnaces.
[0019] Figure 1 illustrates a schematic representation of Continuous Galvanization
Annealing Furnace. The system comprise a preheating furnace, a heating furnace and a
soaking furnace. In the pre-heating furnace, strip is heated directly by hot Hydrogen-
Nitrogen mixture gas (HNX). Further, the strip passes through the preheating furnace
into the heating furnace and the soaking furnace. In the heating furnace and the soaking
furnace, radiant tubes are located surrounding to the running strip. Thus, the running
strip steel is heated indirectly, by in heating and soaking furnace. Each radiant tube
comprises a burner, where the burner is placed inside the radiant tube. The burner burns
a fuel inside the radiant tube and heats the strip from outside without direct contact
between combustion gas and strip. To facilitate heat transfer within the annealing
chamber and making an inert atmosphere, a circulating gaseous mixture (HNX) is
maintained in it.
7
[0020] Further, Figure 2 illustrates actual operational data sample for model
development, in accordance with the present subject matter and Figure 3 illustrates
Parameter recursive learning with actual plant data, in accordance with the present
subject matter. Furthermore, Figure 4 illustrates Temperature prediction result with
adapting model parameters, in accordance with the present subject matter. Figure 5
illustrates a system architecture of Continuous Galvanization Annealing Furnace. The
system comprises a Continuous Annealing furnace, a programmable logic controller
(PLC), an Open Platform Communication (OPC) interface, a model server and a
Human Machine Interface (HMI). In the subsequent section, the present subject matter
is described with reference to Figures 1-5. Further the nomenclature in the equations
presented in the description have been listed below.
Nomenclature
TS strip temperature.
Q fuel flow rate to furnace zone
TSavg average strip temperature in normal operation
Qavg average fuel flow rate in normal operation
????, b1…bn, c1 model parameters
n model order
D disturbances
d delay or dead time
w coil width
t coil thickness
V line speed
?????????? estimated model parameter vector at time instant k
8
e(k) prediction error of strip temperature
K(k) updated gain vector at time instant k
???????? process parameter vector at time instant k
P(k) updated covariance matrix of model parameters at time instant k
?????????? forgetting factor of ith parameter
???? parameter variability index for the ith parameter (constant)
???????? diagonal matrix of forgetting factor
J cost function of MPC
???? weightage for strip temperature prediction error
w?u weightage for fuel flow variation (input move) in control action
p prediction horizon
q control horizon
yr(k) target temperature of coil at time instant k
y(k) measured strip temperature at instant k
?????????? predicted strip temperature at instant k
j future time instant
Uf(k) future input fuel flows vector over control horizon
uL input lower limit
uH input upper limit
?u input move in consecutive instances (fuel flow variation)
?uL input move lower limit
9
?uH input move upper limit
[0021] Referring to real time data capturing aspect of the present subject matter, a realtime
data acquisition module has been deployed in the system. The real-time data
acquisition module captures data from the Continuous Galvanization Annealing
Furnace. The PLC captures data such as furnace operating data, setup coil information
data and tracking information data from the Continuous Galvanization Annealing
Furnace. The data captured by the PLC is passed to the OPC interface, and is ultimately
stored in a database. Further, the data stored in the database is fed as an input to a
processor. The processor predicts the strip temperature based on the input data.
Furthermore, the processor, in accordance with the predicted strip temperature changes
the gas flow rate. The duration for capturing data is 3 sec and the model calculation
control cycle time is 1 minute.
[0022] Referring to mathematical model aspect of the present subject matter,
temperature of heating strip depends on parameters such as strip thickness, width, line
speed and fuel flow rates to heating furnace. Strip temperature predicted by time series
of data driven model that is Auto-Regressive exogenous form (ARX model). First two
terms in right hand side of following equation showing the effect of past strip
temperature and past fuel flow series on present strip temperature. Third term showing
the effect of disturbance on strip temperature. Disturbance term D is due to change in
thickness, width and line speed hence this is known disturbances.
?????????? ?? ?????????? ?? 1?? ?? ???? ??
?????? ?? ?? ?? ?? ?? 1??
??
??????
?? ????????????
Wherein:
???????? ?? ?????????? ?? ??????????
???????? ?? ???????? ?? ????????
???????? ?? ?????????? ?? ???????? ?? ???????? ?? ???? ?? ?? ?? ????????????
10
The dynamic model of strip temperature in equation (1) has two parameters whose
value are fixed in control, that is dead time d and the model order n. These values
identified from actual operation data as shown in Figure 2 under various operating
condition were collected and investigated. Checking all possible combination of d and
n, most appropriate values determine by considering the values of Akaike’s information
criteria (AIC). Choosing delay value can easily found from plant sample data. Once
model order and delay values are fixed, model parameters values ????,b1…bn and c1 are
calculated by using good perturbation data sample of process plant as shown in Figure
2. This is operation batch data having rich variety in line speed, strip thickness, width
and fuel flow that affecting the strip and furnace temperature. By using least square
method, model parameter values are determined but it is better to use online adapted
model parameter instead of constant parameter for model calculations.
[0023] Referring to Online recursive parameter learning aspect of the present subject
matter for encountering the effect of known and unknown disturbance in the process,
model parameter online learning is very important for accurate control system.
Adaptive learning of strip temperature model is represented by the following algorithm
using recursive least square with online forgetting factor. Estimation is performed at
every sampling instance except at the point of transient of coil. Model parameters
?????????? updated at time k in the following way,
?????????? ?? ???????? ?? 1?? ?? ????????????????
Wherein:
???????? ?? ???????? ?? 1?? ?????? ?? ???? ? ?????? ?? ?? ?? ?? ?? 1?? ????????????
?????????? ?? ???????? ?? ??
?? … . … … … .?? ??
?? ??^
??
????
Prediction error of strip temperature y (t),
???????? ?? ???????? ?? ???????? ?? 1????????????
Updated gain vector for parameter learning,
11
???????? ??
?????? ?? 1????????????
??1 ?? ???????????????? ?? 1????????????
Forgetting factor corresponding each model parameters,
?????????? ?? ??1 ??
??????????????
??1 ?? ???????????????? ?? 1????????????
???????? ??
?
? ? ? ?
1
??????????
? 0
? ? ?
0 ?
1
???????????
?
?
?
?
Adaptive covariance matrix of model parameters,
?? ??
?????? ?? ??????????? ?? ?????????????????? ???????? ?? 1?????????
[0024] Prediction Model and model parameter estimation algorithm were applied to
real plant and estimated model parameters values plotted for some data in Figure 3.
After initial period of high fluctuation, model parameter estimates have converged to
reasonable values. Once parameters values converged to reasonable values, each value
of parameter online estimates is limited within a reasonable range for prevention of an
abnormal values.
[0025] Now, referring to Calculation of optimal gas flow rate references, Fuel flow set
point will be calculated based on online optimization by minimize the cost function J
of MPC. Cost function consisting of predicted control errors and incremental fuel flow
in finite future. Predicted control error is a sum of square of deviation between predicted
strip temperature and target strip temperature in finite future. Given the future target
strip temperature trajectory, the model predictive control problem at the sampling
instant k is defined as a constrained optimization problem whereby the future
manipulated input moves Uf(k) are determined by minimizing an objective function J
defined as follows
?????????? ?? ??????????, ?????? ?? 1?? . … … . , ?????? ?? ?? ?? 1????
12
?????????? ??
?????? ?? ?? ????
??
??????
?????????? ?? ???? ?? ???????? ?? ???????? ?? ????? ?????????? ?? ???? ?? ?????? ?? ?? ?? 1??????
??
??????
? Manipulated input constraints
???? ?? ?????? ?? ???? ?? ????
????? ?? ??????? ?? ???? ?? ?????
?? ?? 1,2, … … … ?? ?? 1
? Output quality constraints
???? ?? ?????? ?? ???? ?? ????
?? ?? 1,2, … … … ?? ?? 1
[0026] Model predictive control avoid the violation of input/output variable constraints
and large fuel flow fluctuation in control action. It will ensure smooth changes in input
variable during control action.
[0027] Now referring to Figure 6, the HMI screens will display details of current and
next coil id, actual fuel flow and furnace temperature of each zone, target and actual
strip temperature, predicted strip temperature and model calculated fuel flow set points.
Model predicted strip temperature response and actual plant response graph are
displaying to the operator through an overview screen of the HMI.
[0028] Although implementations for controlling strip temperature in a continuous
galvanization annealing furnace have been described in language specific to structural
features and/or system, it is to be understood that the appended claims are not
necessarily limited to the specific features or described. Rather, the specific features are
disclosed as examples of implementations.