Sign In to Follow Application
View All Documents & Correspondence

A Process Of Early Detection Of Water Leakage Through Any Of The Tuyeres Of A Blast Furnace Based On Differential Water Flow Rate

Abstract: A process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential water flow rate, comprising the steps of: determining a differential flow rate through a tuyere by processing the data representing the differential flow of each tuyere collected from the inlet and outlet flow meters; assigning a non-dimensional number in the form of LDF (Leakage detection factor) which is calculated from a relative deviation of the current differential flow from a normal differential flow, wherein the current differential flow of a tuyere is considered as the average differential flow during the last two hours, and wherein the normal differential flow is the maximum of the differential flow in the last 24 hours and monitoring the LDF of each tuyere in real time and generating an alarm if the LDF value rises beyond a pre-determined threshold limit set for the specific tuyere. {FIGURE – 6}

Get Free WhatsApp Updates!
Notices, Deadlines & Correspondence

Patent Information

Application #
Filing Date
09 February 2016
Publication Number
45/2017
Publication Type
INA
Invention Field
MECHANICAL ENGINEERING
Status
Email
docketing@lsdavar.in
Parent Application
Patent Number
Legal Status
Grant Date
2020-09-02
Renewal Date

Applicants

TATA STEEL LIMITED
Research and Development and Scientific Services Division, Jamshedpur-831001, State - Jharkhand; India

Inventors

1. ASHISH AGRAWAL
c/o. TATA STEEL LIMITED Research and Development and Scientific Services Division, Jamshedpur-831001, State - Jharkhand India.
2. SWAPNIL CHINTAMAN KOR
c/o. TATA STEEL LIMITED Research and Development and Scientific Services Division, Jamshedpur-831001, State - Jharkhand India
3. MAHESH KUMAR AGARWAL
c/o. TATA STEEL LIMITED Research and Development and Scientific Services Division, Jamshedpur-831001, State - Jharkhand India
4. UTPAL NANDY
c/o. TATA STEEL LIMITED Research and Development and Scientific Services Division, Jamshedpur-831001, State - Jharkhand India
5. ABHIK ROY CHAUDHURY
c/o. TATA STEEL LIMITED Research and Development and Scientific Services Division, Jamshedpur-831001, State - Jharkhand India

Specification

FIELD OF THE INVENTION
The present invention relates to a process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential water flow rate.
BACKGROUND OF THE INVENTION
In different types of blast furnaces, varying number of tuyeres are present between 20 to 34 tuyeres, the number increasing with the furnace volume. Hence to monitor water leakage in the flow conduit associated with a large number of tuyeres, 40 to 64 flow meters are required, two for each conduit. The flow of cooling water at the inlet and outlet of the cooling circuit is monitored to determine if there is a leak in the cooling circuit, in the form of difference between the two flows. The flow measurement must be as accurate as possible to quickly identify a leak, and reliable to prevent unnecessary alarms and shutdowns.
Several cooling circuits are generally provided to control heat and maintain thermal balance in blast furnace. Cooling circuits are critical in blast furnace because its malfunctioning can change the process and influence the furnace performance. Monitoring the cooling circuit means checking the flow while ensuring no water leaks in the circuit. Tuyere cooling at the blast furnace is one such relevant circuit of major concern both of operational and safety concern [1]. Tuyere leakage is a problem due to the chilling effect of water in the hearth, and the leaks are generally not detectable until significant damage has occurred. Water leakage may also result in problems in taphole which may disrupt furnace operation. This results in erratic operation and difficulty to recover from the chilled conditions. Water leakage also affects the furnace campaign life if it damages the refractories [2]. A number of methods of detecting leakage have are known in the art. A significant amount of work is done in Japan implementing flow measurement to detect variations from inlet to outlet, however the accuracy and sensitivity of leakage detection is not achievable and the capital and

maintenance is also too high [3]. Hirano has invented a system for monitoring of a water flow conduit to detect water leakage in the blast furnace, where in the pulse output signals from the inlet and outlet flow meters are converted into the analog current signal proportional to the pulse frequency and this analog signal is applied to the water flow meters conduits to identify the water leakage [4].
Despite the availability of water flow meter at each tuyere it is still difficult to identify the leaking tuyere at the early stage unless the rupture size grows bigger and so the leakage. This is so because firstly the water flow meters measure the water inflow rate and outflow rate and does not determine the water differential flow rate, secondly the range of water inflow rate and water outflow rate are different for different tuyeres and it varies with the varying process performance. Lastly the water differential flow is very small and it becomes difficult to identify the water ingressing. In several cases the furnace is forced to shut down to visually inspect the leakage occurring without prior knowledge of the exact tuyere number being leaked.
Further, the differential flow of the tuyeres normally vary from around 0.1-4 liter/min. Thus, the lowest values of differential flow are almost forty times lesser than the highest values. As a result, it becomes quite challenging to set a common criterion for generating alarms. For example, a particular relative deviation which is alarming for one tuyere, could be a perfectly normal case for the other tuyeres. An obvious solution to overcome this problem, was to study each tuyere one by one and set alarm thresholds unique to each tuyere. However, this solution was found to be unfeasible, because the quantum of differential flow changes dynamically with the passage of time and upon every replacement of a burnt nose. Further, carrying-out such an exercise for each tuyere, when the total number of tuyere is high, would be quite tedious.
The tuyere nose is the inner most part of the tuyere which protrudes inside the furnace conveying the hot blast and is therefore subjected to stringent conditions

of temperature, dropping molten liquids and coke in the furnace raceway. The tuyere nose is made of copper and is water cooled by a cooling circuit embedded within its surface. Due to wear and tear and exposure to extreme conditions, the tuyere nose surface develops cracks which lead to leakage of water into the furnace from the cooling circuit. The crack slowly expands over time causing more water to enter into the furnace. Ingress of water implies furnace cooling. As a result, not only the coke demand and cost increases, there is also loss of production and deterioration in hot metal quality owing to colder spells. Despite the presence of flow measurements available in all tuyeres, it is often difficult to identify the leaking tuyere. If a leak starts, furnace is to be taken shutdown to visually inspect the leak.
It is known that a leak in the cooling water circuit changes the dynamics of the cooling system, impacting the reaction in the furnace, and decreasing the performance and efficiency of the furnace. Moreover, if the leak is not detected in a timely manner and water is injected into the furnace, hydrogen is produced resulting in dangerous safety conditions and causing erosion in the furnace walls and roof that can be expensive and time consuming to repair.
OBJECTS OF THE INVENTION
It is therefore an object of the invention to propose a process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential water flow rate.
Another object of the invention is to propose a process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential water flow rate, in which the differential flows are divided to three different zones and monitor the flow data of the zones to determine threshold values.
A further object of the invention is to propose a process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential

water flow rate, in which three alarms are set for the operates based on said threshold values which enable early detection of water leakage in all of the plurality of tuyeres, thereby eliminate the disadvantages of prior art to set individual alarm for each of the large number of tuyeres.
SUMMARY OF THE INVENTION
Accordingly, there is provided A process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential water flow rate, comprising the steps of: determining a differential flow rate through a tuyere by processing the data representing the differential flow of each tuyere collected from the inlet and outlet flow meters; assigning a non-dimensional number in the form of LDF (Leakage detection factor) which is calculated from a relative deviation of the current differential flow from a normal differential flow, wherein the current differential flow of a tuyere is considered as the average differential flow during the last two hours, and wherein the normal differential flow is the maximum of the differential flow in the last 24 hours and monitoring the LDF of each tuyere in real time and generating an alarm if the LDF value rises beyond a pre-determined threshold limit set for the specific tuyere.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
Figure 1 : schematic of a known tuyere Cooling Water Leakage Detection System.
Figure 2 : Water Tank Level {%) monitoring
Figure 3 : Water Tank Level Drop Rate monitoring

Figure 4: Identification of faulty tuyere state based on the relative water differential flow
Figure 5 : The Tuyere Detection System of Figure 1 showing the spike of LDF during water leakage
Figure 6 : Illustration of a water Leakage detection process of the invention.
Figure 7 : Current Tuyere Life and Tuyere Life History of All the Tuyeres
Figure 8 : Tuyere Nose Burn Failure in H and I Blast Furnaces. (Montly-wise and Year-wise tuyere failure trend in H and I blast furnaces.
DETAIL DESCRIPTION OF THE INVENTION
Accordingly, there is provided a process of early detection of water leakage through any of the tuyeres of a blast furnace based on differential water flow rate. Thus, the invention identified water leakage based on the study of the water flow patterns through the tuyeres; and a non-dimensional factor LDF (Leak Detection Factor) associated with each tuyere is established. LDF is derived from the relative deviation of the differential flow in the cooling circuit of the tuyere from the maximum permissible differential flow observed in the recent past. This LDF value rises sharply and instantly like a spike whenever water leaks from tuyere.
The innovative process basically provides an early warning or indication of the leaking tuyere, right at the time when water starts creeping into the furnace, by processing the data of the flow meters positioned at the inlet and outlet of the water cooling pipe in each tuyere, the cooling water tank level of the cooling circuit corresponding to the tuyeres and a set of predefined criteria which has been arrived at after extensive of process parameters analysis. The relative deviations between the current differential flow levels and the long term

differential flow levels have been used for the leakage detection. The benefit of this process is twofold. Firstly it detects the leakage during the very early stages and secondly it pinpoints the exact tuyere where the leakage is actually taking place.
The measurement of the process data must be repeatable and reliable to prevent false alarms and nuisance furnace shut downs, and the measurement devices should be self-checking to insure they are working correctly.

The process of the invention is based on the inflow and outflow rates. Water is used to cool the tuyeres carrying hot air and fuel into the furnace. The flow of cooling water at the inlet and outlet of the cooling circuit is monitored to determine if there is a leak in the cooling circuit, in the form of difference between the two flows. The flow measurement must be as accurate as possible to quickly identify a leak, and reliable to prevent unnecessary alarms and shutdowns. The cooling water used for cooling the tuyere comes from the water tank. This cooling

circuit is very critical for furnace stability and therefore the level of water tank is continuously being monitored, as sudden fall in the tank level indicates the chances of loss of water in the cooling circuit of the furnace. The tank level is monitored by the level sensors mounted at the top of the tank facing inwards. The tank level is allowed to vary over a fixed range as shown in Figure 2. The level sensors are programmed such that when the minimum threshold limit of the water tank is reached, the tank starts filling and when it reaches the maximum permissible limit the filling is stopped. The time taken by water tank level to reach the minimum threshold limit from the maximum permissible limit is called the 'make-up time' and is calculated as given in eqn. (1). Whenever there is a leak detected in the cooling circuit the make-up time reduces, it means there is a sudden fall in the tank level. The rate of fall in level of water tank level is called 'drop-rate' and is calculated as given in eqn. (2). Whenever there is a leak detected in the cooling circuit the drop-rate increases. These two measurements are the basic indicators of leak occurring in the cooling circuit. However, they give a prior indication of the leak without identifying the exact leaking tuyere number.
Where tm = make-up time/ hr, tc= current time of maximum tank level/ hr and tp = previous time of minimum tank level/ hr, dr= drop rate of tank/ % hr1, Tc= current tank level/ %, T(c-10) = tank level before 10 minutes/ %.
The method of identifying the leak is based on the study of the water flow patterns through the tuyeres; and a non-dimensional leak detection factor (LDF) associated with each tuyere is established. LDF is derived from the relative deviation of the water differential flow in the cooling circuit of the tuyere from the maximum permissible differential flow observed in the recent past. This LDF value rises sharply and instantly like a spike whenever water leaks from tuyere. It

provides an early warning or indication of the leaking tuyere, right at the time when water starts creeping into the furnace. It does so by processing the information of the flow meters positioned at the entry and exit of the water cooling pipe in each tuyere, the cooling water tank level of the cooling circuit corresponding to the tuyeres and a set of predefined criteria which has been arrived at after extensive data analysis. The relative deviations between the current differential flow levels and the long term differential flow levels have been used for the leak detection.
According to the invention, and regardless of the type of cooling circuit, the working principle of leak detection is as follows: the inlet and outlet flow rates of a cooling circuit are measured by means of suitable flow meters. If a leak occurs in the cooling circuit, a shift in the flow between the inlet and outlet streams will be detected and an alarm will be generated. Figure 1 shows the working principle of tuyere cooling WLDS. It can be seen schematically that based on the water inflow and outflow rates, (he difference, flow rate is calculated which is used as a Key parameter in monitoring the water leakage through tuyere. It does so by processing the information of the differential flow of each tuyere obtained from the inlet and outlet flow meters. A non-dimensional number LDF .calculated from the relative deviation of the current differential flow from the normal differentia, flow in the recent past. In a conventional monitoring system ,n which me instantaneous flow rates in the inlet and outlet lines are compared to produce an alarm signal, false alarms will be produced. The current differential flow of a tuyere is considered as the average differential flow of the same in last 2 hours, whereas the normal differential flow in the recent past is the peak differential flow observed in the last 24 hours. The empirical expression of LDF ,s shown in eqn.


The alarm is generated when LDF value rises beyond the pre-defined threshold limits set for that tuyere. The faulty tuyere in blast furnace is identified based on the water differential flow, where the differential flow is the difference of inlet and outlet water flow rate. The relative deviation of the water differential flow is evaluated based on the behavior of water flow of last 24 hours. Since the flow patterns are varying in every next cycle instead of taking the instantaneous flow, the average water differential flow of last 2 hours is considered.


SECONDARY EVIDENCE OF TECHNICAL ADVANCEMENT IN THE FORM OF ECONOMIC SIGNIFICANCE
The savings incurred by the system is around 87 lacs for each tuyere nose burn failure. Eg: In the current year 2014, this system has detected 29 numbers of tuyere nose burn failure in H blast furnace and 33 such cases in I blast furnace. With the implementation of this system the unscheduled shut down of inspecting the tuyere leakage identification is eliminated. Now the usual practice is to stop the water circuit of the particular tuyere ingressing and to continue run the furnace without hampering the operation. There is no such need for the operators to go close to the tuyere platform area in the cast house and physically inspect all the tuyeres and detect the water leakage. The benefits perceived after the implementation of this system is that the unscheduled shut down hours is reduced. Earlier unscheduled shut down of the furnace were taken place

frequently because of the tuyere nose burn failure resulting in water ingressing and thereby chilling the furnace and therefore consumption of higher coke rate. The loss of water into the atmosphere is avoided with resultant savings in water consumption. Consumption of higher coke rate is also reduced and thus C02 emission in the atmosphere.
Earlier (2009-2011) when water leakage through any tuyere occurs, the furnace was forced to shut down and manual inspection was to be done for identification of the exact tuyere ingressing. This manual inspection was done every time a tuyere undergoes water leakage. The total time for the inspection was about 2 hours. For H & I blast furnaces with an average hot metal production rate of 8800 tons/day, a production loss of 2 hours means a loss of almost 730 tons (365*2).
References
[1]http://www.es.endress.com/es/Grupo-Endress-Hauser/industry-automation-
expertise/primaries-metal/safety/converter
[2]. D. Jameson, H. Lungen, D. Lao, "Technical study into the means of
prolonging blast furnace campaign life", technical steel research, 1997.
[3]. D. Farrington, w. Stewart, P. Kitson, M. London, N. Paddy, "Dynamic
monitoring of blast furnace plant", technical steel research, 1998.
[4]. Hirano, "System For Monitoring Flow Rate Difference In Water Cooling
Conduit", Us Patent May 16,1978.

WE CLAIM
1. A process of early detection of water leakage through any of the tuyeres of
a blast furnace based on differential water flow rate, comprising the steps
of:
- determining a differential flow rate through a tuyere by processing the data representing the differential flow of each tuyere collected from the inlet and outlet flow meters;
- assigning a non-dimensional number in the form of LDF (Leakage detection factor) which is calculated from a relative deviation of the current differential flow from a normal differential flow,
wherein the current differential flow of a tuyere is considered as the average differential flow during the last two hours, and wherein the normal differential flow is the maximum of the differential flow in the last 24 hours;
and
- monitoring the LDF of each tuyere in real time and generating an alarm
if the LDF value rises beyond a pre-determined threshold limit set for
the specific tuyere.
2. The process as claimed in claim 1, wherein the level of water tank
supplying cooling water to the tuyere nose is continuously monitored to
identify a sudden fall or otherwise of the tank level which indicates a
possibility of water ingress into the furnace.

Documents

Application Documents

# Name Date
1 Power of Attorney [09-02-2016(online)].pdf 2016-02-09
2 Form 5 [09-02-2016(online)].pdf 2016-02-09
3 Form 3 [09-02-2016(online)].pdf 2016-02-09
4 Form 20 [09-02-2016(online)].pdf 2016-02-09
5 Drawing [09-02-2016(online)].pdf 2016-02-09
6 Description(Complete) [09-02-2016(online)].pdf 2016-02-09
7 201631004530-12-05-2016)-FORM-1.pdf 2016-05-12
8 201631004530-12-05-2016)-CORRESPONDENCE.pdf 2016-05-12
9 201631004530-FER.pdf 2019-05-02
10 201631004530-RELEVANT DOCUMENTS [31-10-2019(online)].pdf 2019-10-31
11 201631004530-OTHERS [31-10-2019(online)].pdf 2019-10-31
12 201631004530-FORM 3 [31-10-2019(online)].pdf 2019-10-31
13 201631004530-FORM 13 [31-10-2019(online)].pdf 2019-10-31
14 201631004530-FER_SER_REPLY [31-10-2019(online)].pdf 2019-10-31
15 201631004530-DRAWING [31-10-2019(online)].pdf 2019-10-31
16 201631004530-ABSTRACT [31-10-2019(online)].pdf 2019-10-31
17 201631004530-US(14)-HearingNotice-(HearingDate-27-07-2020).pdf 2020-06-25
18 201631004530-Correspondence to notify the Controller [20-07-2020(online)].pdf 2020-07-20
19 201631004530-Written submissions and relevant documents [11-08-2020(online)].pdf 2020-08-11
20 201631004530-PatentCertificate02-09-2020.pdf 2020-09-02
21 201631004530-IntimationOfGrant02-09-2020.pdf 2020-09-02
22 201631004530-RELEVANT DOCUMENTS [30-09-2022(online)].pdf 2022-09-30
23 201631004530-PROOF OF ALTERATION [25-02-2023(online)].pdf 2023-02-25
24 201631004530-Response to office action [23-05-2023(online)].pdf 2023-05-23
25 201631004530-26-09-2023-FORM-27.pdf 2023-09-26
26 201631004530-26-09-2023-CORRESPONDENCE.pdf 2023-09-26

Search Strategy

1 201631004530A_03-10-2018.pdf

ERegister / Renewals

3rd: 28 Nov 2020

From 09/02/2018 - To 09/02/2019

4th: 28 Nov 2020

From 09/02/2019 - To 09/02/2020

5th: 28 Nov 2020

From 09/02/2020 - To 09/02/2021

6th: 05 Mar 2021

From 09/02/2021 - To 09/02/2022

7th: 28 Feb 2022

From 09/02/2022 - To 09/02/2023

8th: 31 Jan 2023

From 09/02/2023 - To 09/02/2024

9th: 03 Feb 2024

From 09/02/2024 - To 09/02/2025

10th: 30 Jan 2025

From 09/02/2025 - To 09/02/2026