A method for identifying blast furnace collapse conditions based on thermal image of material surface

Abstract

The present invention provides a method for identifying the condition of collapsed material in a blast furnace based on thermal images of the material surface. The thermal imager of the material surface and image recognition technology are used to establish judgment rules, judge and identify the condition of collapsed material in the blast furnace during the production process, and remind the blast furnace operator to take countermeasures to avoid serious impact on the stable and smooth operation of the blast furnace. It includes: installing a thermal imager on the top of the blast furnace to establish a throat material surface coordinate system, dividing the throat section of the blast furnace into n×m units according to the coordinates; starting with the closing of the furnace top material flow valve and ending with the opening of the furnace top material flow valve to lay a batch of furnace materials, the furnace top thermal imager scans the material surface once every scanning cycle, and reads the coordinates (x i ,θ j , k) unit temperature T(x i ,θ j , k) and material line height Z(x i ,θ j , k), starting from the second scan of the furnace top thermal imager, the material unloading speed and temperature change amplitude of each unit are calculated after each scan, and whether the material collapse occurs during the process of the furnace charge descending is determined based on the material unloading speed and temperature change amplitude.

Description

Technical Field

[0001] The invention relates to the technical field of metallurgy, and in particular to a method for identifying a blast furnace material collapse condition based on a material surface thermal image. Background Art

[0002] The goal of blast furnace is high efficiency, low consumption and long life. It is necessary to ensure the long-term stable operation of blast furnace and avoid abnormal furnace conditions, or to identify the signs of abnormal furnace conditions in time and take correct countermeasures in the early stage of abnormal furnace conditions to reduce the losses caused by abnormal furnace conditions. When the blast furnace has abnormal furnace conditions such as hanging material, collapsed material, pipeline travel, etc., it will destroy the stable operation of the blast furnace, which is easy to cause insufficient physical heat of molten iron, poor molten iron quality, reduced blast furnace production capacity, increased coke ratio, and even serious hazards such as slag filling at the tuyere and freezing of the furnace. Among the types of abnormal furnace conditions in blast furnaces, collapsed material is the most common one. Collapse of material is a phenomenon in which the material column at a high place collapses to a low place due to the uneven distribution of ore and coke in the material column and the uneven chemical reaction rate during the descent of the material, causing the material column at a high place to collapse to a low place, resulting in a local collapse of the material layer. If the blast furnace has continuous collapsed material, it will seriously affect the effect of ore and high-temperature preheating and indirect reduction in the furnace, induce the furnace to cool sharply, and have a serious impact on the stable operation of the blast furnace.

[0003] At present, blast furnace operators still rely mainly on years of experience, combined with a few parameters such as material size, wind pressure, and air volume, to judge abnormal furnace conditions such as material collapse that have occurred or may occur during the blast furnace production process. This judgment method based mainly on operator experience is affected by the operator's experience and knowledge. If a misjudgment or delay occurs, it will cause unnecessary losses to the blast furnace.

[0004] In addition, some blast furnaces use blast furnace expert systems or artificial intelligence systems, which use neural networks and other algorithms, combined with the expert knowledge of blast furnace operators to judge abnormal furnace conditions such as material collapse during the blast furnace production process. This method is an indirect method and cannot accurately and directly judge abnormal furnace conditions during the blast furnace production process. It is easy to miss judgments and cause unnecessary losses to the blast furnace.

[0005] With the application of thermal imaging and radar detection technology in blast furnaces, a thermal imager is used on the top of the blast furnace to detect the image of the material surface in the blast furnace online, and image recognition technology is used to establish the material surface temperature field. A mathematical model is established based on the material surface image and temperature field, and the distribution shape of the material surface is simulated and calculated. The overall shape of the entire material surface in the field of view of the thermal imager can be easily and accurately obtained, assisting the blast furnace operator in judging the rationality of the operation of the material distribution system. So far, there has been no report on the application of thermal image detection technology to judge the abnormal furnace condition of material collapse in the blast furnace production process. Summary of the invention

[0006] In order to solve the technical problems raised by the background technology, the present invention provides a method for identifying the material collapse condition of a blast furnace based on thermal images of the material surface. A material surface thermal imager and image recognition technology are used to establish judgment rules, judge and identify the material collapse condition of the blast furnace during the production process, and remind the blast furnace operator to take countermeasures to avoid serious impact on the stable and smooth operation of the blast furnace.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A method for identifying a blast furnace collapse condition based on a material surface thermal image comprises the following steps:

[0009] Step 1: Install a thermal imager on the top of the blast furnace;

[0010] Step 2: Establish the throat material surface coordinate system, take the throat cross section center as the origin, divide the throat cross section area n into multiple circles, and determine the radial coordinate x of each circle. i , i = 1, 2, 3 ... n, specify a radius of the furnace throat section as the starting point, divide the center angle of the furnace throat section into m sectors in a clockwise direction, and determine the center angle θ of each sector j , j = 1, 2, 3 ... m;

[0011] Step 3: Draw the cross section of the blast furnace throat (x i ,θ j ) The coordinates are divided into n×m units;

[0012] Step 3: Draw the cross section of the blast furnace throat (x i ,θ j ) The coordinates are divided into n×m units;

[0013] Step 4: Starting with the closing of the furnace top material flow valve and ending with the opening of the furnace top material flow valve to place the next batch of furnace materials, the furnace top thermal imager scans the material surface once every scanning cycle, and the number of scanning cycles k = 1, 2, 3..., reads the coordinates (x i ,θ j ) The unit temperature T(x) of each scan i ,θ j , k)(℃) and material line height Z(x i ,θ j , k)(cm);

[0014] Step 5: Starting from the second scan of the furnace top thermal imager, calculate the material unloading speed v(x i ,θ j , k)(cm / s) and the temperature variation amplitude ΔT(x i ,θ j , k)(℃ / s);

[0015] Step 6: According to the material feeding speed v(x i ,θ j , k) and the temperature variation amplitude ΔT(x i ,θ j , k) Determine whether material collapse occurs during the charge descent process.

[0016] Furthermore, in step 2, the radial coordinate x of each ring i The calculation method is as follows:

[0017]

[0018] In the formula: x i , radial coordinate of each ring, cm;

[0019] r0 is the furnace throat radius, cm.

[0020] Furthermore, in step 5, the material feeding speed v(x i ,θ j , k) is calculated as follows:

[0021]

[0022] i=1,2,3…n;j=1,2,3…m;k=2,3…,t is the sampling period, unit is s.

[0023] Furthermore, in step 5, the temperature variation range ΔT(x i ,θ j , k) is calculated as follows:

[0024]

[0025] i=1,2,3…n;j=1,2,3…m;k=2,3…,t is the sampling period, unit is s.

[0026] Furthermore, in step 6, the method for judging whether the charge collapses during the charge descent process is as follows:

[0027] If A1>v(x i ,θ j , k) ≥ A2, but ΔT(x i ,θ j , k)>B1, it is determined that abnormal furnace conditions such as material collapse will occur; if v(x i ,θ j , k) ≥ A1, it is directly determined that the abnormal furnace condition of material collapse occurs;

[0028] A1, A2 are the material feeding speed v(x i ,θ j, k) is the maximum and minimum value of the set threshold, B1 is the temperature change amplitude ΔT(x i ,θ j , k) is the maximum value of the set threshold.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] The present invention uses a material surface thermal imager to extract material surface images of the material surface falling process in the furnace in real time, utilizes image recognition technology to divide the entire material surface into multiple areas of equal area, extracts the image coordinates of each area, the material surface temperature and characteristics of each area respectively, establishes judgment rules, judges and identifies the material collapse condition of the blast furnace during the production process, and reminds the blast furnace operator to take countermeasures to avoid serious impact on the stable and smooth operation of the blast furnace, ensures stable and smooth furnace condition, stable production of the blast furnace, and long life of the blast furnace, and is also conducive to realizing dynamic distribution control of the shape of blast furnace charge. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A diagram of establishing a throat material surface coordinate system for a method for identifying blast furnace collapse conditions based on material surface thermal images according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The specific implementation modes provided by the present invention are described in detail below.

[0033] A method for identifying a blast furnace collapse condition based on a material surface thermal image comprises the following steps:

[0034] Step 1: Install a thermal imager on the top of the blast furnace;

[0035] Step 2: Establish the throat material surface coordinate system, take the throat cross section center as the origin, divide the throat cross section area n into multiple circles, and determine the radial coordinate x of each circle. i , i = 1, 2, 3 ... n, specify a radius of the furnace throat section as the starting point, divide the center angle of the furnace throat section into m sectors in a clockwise direction, and determine the center angle θ of each sector j , j = 1, 2, 3 ... m;

[0036] Step 3: Draw the cross section of the blast furnace throat (x i ,θ j ) The coordinates are divided into n×m units;

[0037] The radial coordinate x of each ring i The calculation method is as follows:

[0038]

[0039] In the formula: x i , radial coordinate of each ring, cm;

[0040] r0 is the furnace throat radius, cm.

[0041] Step 3: Draw the cross section of the blast furnace throat (x i ,θ j ) The coordinates are divided into n×m units;

[0042] Step 4: Starting with the closing of the furnace top material flow valve and ending with the opening of the furnace top material flow valve to place the next batch of furnace materials, the furnace top thermal imager scans the material surface once every scanning cycle, and the number of scanning cycles k = 1, 2, 3..., reads the coordinates (x i ,θ j ) The unit temperature T(x) of each scan i ,θ j , k)(℃) and material line height Z(x i ,θ j , k)(cm);

[0043] Step 5: Starting from the second scan of the furnace top thermal imager, calculate the material unloading speed v(x i ,θ j , k)(cm / s) and the temperature variation amplitude ΔT(x i ,θ j , k)(℃ / s);

[0044] Feeding speed v(x i ,θ j , k) is calculated as follows:

[0045]

[0046] i=1,2,3…n;j=1,2,3…m;k=2,3…,t is the sampling period.

[0047] Temperature change range ΔT(x i ,θ j , k) is calculated as follows:

[0048]

[0049] i=1,2,3…n;j=1,2,3…m;k=2,3…,t is the sampling period.

[0050] Step 6: According to the material feeding speed v(x i ,θ j , k) and the temperature variation amplitude ΔT(x i ,θ j , k) Determine whether the charge collapses during the charge descent process. The method for determining whether the charge collapses during the charge descent process is as follows:

[0051] If A1>v(xi ,θ j , k) ≥ A2, but ΔT(x i ,θ j , k)>B1, it is determined that abnormal furnace conditions such as material collapse will occur; if v(x i ,θ j , k) ≥ A1, it is directly determined that the abnormal furnace condition of material collapse occurs;

[0052] A1, A2 are the material feeding speed v(x i ,θ j , k) is the maximum and minimum value of the set threshold, B1 is the temperature change amplitude ΔT(x i ,θ j , k) is the maximum value of the set threshold.

[0053] The specific setting can be: judge whether the material collapse occurs during the charge descent process, if 10cm / s>v(x i+1 ,θ j+1 ,k+1)≥6.7cm / s, but ΔT(x i+1 ,θ j+1 ,k+1)>3.3℃ / s, it can be determined that abnormal furnace conditions such as material collapse will occur; if v(x i+1 ,θ j+1 ,k+1)≥10cm / s, it can be directly determined that abnormal furnace condition of material collapse will occur.

[0054] Best implementation method 1:

[0055] Suitable for blast furnaces with a throat diameter of 840 cm.

[0056] (1) Establish the throat material surface coordinate system, such as Figure 1 As shown, the throat cross-sectional area is divided into 11 equal parts, and the radial coordinate (x i ) is calculated as follows:

[0057]

[0058] Each ring x1,x2,x3,...x 11 The radial coordinates are 420cm, 400cm, 380cm, 358cm, 335cm, 310cm, 283cm, 253cm, 219cm, 179cm, and 127cm respectively.

[0059] (2) Taking the center of the furnace throat as the starting point, divide the central angle of the furnace throat into 8 equal sectors (θ j ),

[0060] The central angles of each sector θ1, θ2, θ3, θ4, θ5, θ6, θ7, and θ8 are 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315° respectively.

[0061] (3) Draw the cross section of the blast furnace throat according to (x i ,θ j ) coordinates are divided into 88 units.

[0062] (4) Starting with the closing of the furnace top material flow valve and ending with the opening of the furnace top material flow valve to place the next batch of furnace materials, the furnace top thermal imager scans the material surface for the second and third time, k = 2, 3, and reads the (400, 45, k) unit material line heights of 150 cm and 300 cm respectively.

[0063] (5) Calculate the unloading speed v(400, 45, 3) (cm / s) of the (400, 45, 3) unit

[0064]

[0065] (6) Determine whether there will be material collapse during the charge descent process. If v(400, 45, 3) = 10 cm / s, it can be determined that the (400, 45, 3) unit will experience an abnormal condition of material collapse.

[0066] Best practice 2:

[0067] Suitable for blast furnaces with a throat diameter of 900 cm.

[0068] (1) Establish the throat material surface coordinate system, such as Figure 1 As shown, the throat cross-sectional area is divided into 11 equal parts, and the radial coordinate (x i ) is calculated as follows:

[0069]

[0070] Each ring x1,x2,x3,...x 11 The radial coordinates are 450cm, 429cm, 407cm, 384cm, 359cm, 332cm, 303cm, 271cm, 235cm, 192cm, and 136cm respectively.

[0071] (2) Taking the center of the furnace throat as the starting point, divide the central angle of the furnace throat into 8 equal sectors (θ j ),

[0072] The central angles of each sector θ1, θ2, θ3, θ4, θ5, θ6, θ7, and θ8 are 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315° respectively.

[0073] (3) Draw the cross section of the blast furnace throat according to (x i ,θ j ) coordinates are divided into 88 units.

[0074] (4) Starting with the closing of the furnace top material flow valve and ending with the opening of the furnace top material flow valve to place the next batch of furnace materials, the furnace top thermal imager scans the material surface for the third and fourth times, k=3, 4, and reads the (450, 135, k) unit material line heights of 150 cm and 250 cm respectively; through image recognition, the unit material surface temperatures are 150°C and 215°C respectively.

[0075] (5) Calculate the unloading speed v(450, 135, 4) (cm / s) and the temperature rise amplitude ΔT(450, 135, 4) (℃ / s) of the (450, 135, 4) unit

[0076]

[0077]

[0078] (6) Determine whether the charge will collapse during the charge descent process. v(450, 135, 4) = 6.7 cm / s, ΔT(450, 135, 4) = 4.3 (°C / s). It can be determined that the (450, 135, 4) unit will experience an abnormal charge collapse condition.

[0079] The present invention uses a material surface thermal imager to extract material surface images of the material surface falling process in the furnace in real time, utilizes image recognition technology to divide the entire material surface into multiple areas of equal area, extracts the image coordinates of each area, the material surface temperature and characteristics of each area respectively, establishes judgment rules, judges and identifies the material collapse condition of the blast furnace during the production process, and reminds the blast furnace operator to take countermeasures to avoid serious impact on the stable and smooth operation of the blast furnace, ensures stable and smooth furnace condition, stable production of the blast furnace, and long life of the blast furnace, and is also conducive to realizing dynamic distribution control of the shape of blast furnace charge.

[0080] The above embodiments are implemented based on the technical solution of the present invention, and detailed implementation methods and specific operation processes are given, but the protection scope of the present invention is not limited to the above embodiments. The methods used in the above embodiments are conventional methods unless otherwise specified.

Claims

1. A method for identifying blast furnace collapse conditions based on material surface thermal images, characterized in that: The steps include: Step 1: Install a thermal imager on the top of the blast furnace; Step 2: Establish the throat material surface coordinate system, take the center of the throat section as the zero point, divide the throat cross-sectional area n into multiple circles, and determine the radial coordinate x of each circle. i , i = 1, 2, 3 ... n, specify a radius of the furnace throat section as the starting point, divide the center angle of the furnace throat section into m sectors in a clockwise direction, and determine the center angle θ of each sector j , j = 1, 2, 3 ... m; Step 3: Draw the cross section of the blast furnace throat (x i ,θ j ) The coordinates are divided into n×m units; Step 4: Starting with the closing of the furnace top material flow valve and ending with the opening of the furnace top material flow valve to place the next batch of furnace materials, the furnace top thermal imager scans the material surface once every scanning cycle, and the number of scanning cycles k = 1, 2, 3..., reads the coordinates (x i ,θ j ) The unit temperature T(x) of each scan i ,θ j , k), unit ℃, and material line height Z (x i ,θ j , k), unit: cm; Step 5: Starting from the second scan of the furnace top thermal imager, calculate the material unloading speed v(x i ,θ j , k), in cm / s, and the temperature change amplitude ΔT(x i ,θ j , k), unit: °C / s; Step 6: According to the material feeding speed v(x i ,θ j , k) and the temperature variation amplitude ΔT(x i ,θ j , k) determine whether there is material collapse during the charge descent process; The method for judging whether the charge collapses during the charge descent process is as follows: If A1>v(x i ,θ j , k) ≥ A2, but ΔT(x i ,θ j , k)>B1, it is determined that abnormal furnace conditions such as material collapse will occur; if v(x i ,θ j , k) ≥ A1, it is directly determined that the abnormal furnace condition of material collapse occurs; A1, A2 are the material feeding speed v(x i ,θ j , k) is the maximum and minimum value of the set threshold, B1 is the temperature change amplitude ΔT(x i ,θ j , k) is the maximum value of the set threshold.

2. The method for identifying blast furnace collapse conditions based on material surface thermal images according to claim 1 is characterized in that: In step 2, the radial coordinate x of each ring is i The calculation method is as follows: In the formula: x i , radial coordinate of each ring, cm; r0 is the furnace throat radius, cm.

3. The method for identifying blast furnace collapse conditions based on material surface thermal images according to claim 1, characterized in that: In step 5, the material feeding speed v(x i ,θ j , k) is calculated as follows: i=1,2,3…n;j=1,2,3…m;k=2,3…,t is the sampling period, unit is s.

4. The method for identifying blast furnace collapse conditions based on material surface thermal images according to claim 1, characterized in that: In step 5, the temperature variation range ΔT(x i ,θ j , k) is calculated as follows: i=1,2,3…n;j=1,2,3…m;k=2,3…,t is the sampling period, unit is s.

Images