Air leak detection method and air leak detection device

Abstract

To provide a leak detection method and leak detection device capable of accurately detecting leaks in the exhaust gas duct of a Dwightroid-type sintering machine. [Solution] The leak detection method according to the present invention is a leak detection method for detecting leaks in the exhaust gas duct of a Dwightroid type sintering machine that fires sintering raw materials, and includes: a measurement step of measuring the oxygen concentration in the exhaust gas duct using an oxygen concentration meter installed on the inlet side of an exhaust fan provided in the exhaust gas duct; a prediction step of calculating predicted values ​​of oxygen concentration components during processing by inputting information during processing to a prediction model in which information indicating the operating state of the Dwightroid type sintering machine and the mixing state of the sintering raw materials is used as input variables and predicted values ​​of oxygen concentration components affected by the operating state of the Dwightroid type sintering machine and the mixing state of the sintering raw materials are used as output variables; and a detection step of detecting leaks in the exhaust gas duct based on the deviation between the oxygen concentration measured in the measurement step and the predicted values ​​of oxygen concentration components calculated in the prediction step.

Description

Technical Field

[0001] The present invention relates to a method and an apparatus for detecting air leakage in an exhaust gas duct of a Dwight Lloyd type sintering machine for firing raw materials for sintering.

Background Art

[0002] A Dwight Lloyd type sintering machine manufactures sintered ore by firing raw materials for sintering, which are a mixture of iron ore, auxiliary raw materials, and carbonaceous materials as a heat source. Specifically, the Dwight Lloyd type sintering machine deposits raw materials for sintering on a circulating pallet, ignites the surface of the raw materials, and then sucks air from the upper part to the lower part of the pallet by a wind box. As a result, sintered ore is manufactured by firing the raw materials for sintering. The air sucked by the wind box is discharged into an exhaust gas duct via a wind leg.

[0003] In a Dwight Lloyd type sintering machine having such a configuration, when air leakage occurs in the air discharge system, unnecessary air suction is induced from the air leakage location, and the amount of air required for firing the raw materials for sintering is reduced. As a result, the progress of firing the raw materials for sintering is adversely affected, and the productivity of sintered ore decreases. Therefore, in order to maintain the productivity of sintered ore, it is important to detect the air leakage location at an early stage and take measures such as repair. Against this background, techniques for detecting air leakage locations in the air discharge system have been proposed.

[0004] Specifically, Patent Document 1 describes a technique for measuring the oxygen concentration at a plurality of positions in an exhaust gas treatment system of a sintering machine and comparing the oxygen concentration between two measurement positions to detect the presence or absence of air leakage between the two measurement positions. Further, Patent Document 2 describes a technique for measuring the oxygen concentration and temperature on the upstream side and downstream side of an exhaust gas treatment system of a sintering machine and detecting the air leakage state of the exhaust gas treatment system based on the difference in oxygen concentration or temperature.

Prior Art Documents

Patent Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2022-134457 [Patent Document 2] Japanese Patent Application Publication No. 6-300459 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, the technologies described in Patent Documents 1 and 2 cannot distinguish between changes in oxygen concentration due to fluctuations in the operating state of the Dwightroid sintering machine and the blending state of the sintering raw materials, and changes in oxygen concentration due to air leakage. Therefore, according to the technologies described in Patent Documents 1 and 2, even though the change in oxygen concentration is due to fluctuations in the operating state of the Dwightroid sintering machine and the blending state of the sintering raw materials, it may be determined that the change in oxygen concentration is due to air leakage, potentially leading to false detection of air leakage.

[0007] The present invention was made to solve the above problems, and its objective is to provide a leak detection method and leak detection device that can accurately detect leaks in the exhaust gas duct of a Dwightroid type sintering machine. [Means for solving the problem]

[0008] The leak detection method according to the present invention is a leak detection method for detecting leaks in the exhaust gas duct of a Dwightroid type sintering machine that fires sintering raw materials, and includes: a measurement step of measuring the oxygen concentration in the exhaust gas duct using an oxygen concentration meter installed on the inlet side of an exhaust fan provided in the exhaust gas duct; a prediction step of calculating predicted values ​​of the oxygen concentration components during processing by inputting the information during processing to a prediction model in which information indicating the operating state of the Dwightroid type sintering machine and the mixing state of the sintering raw materials is used as input variables and predicted values ​​of oxygen concentration components affected by the operating state of the Dwightroid type sintering machine and the mixing state of the sintering raw materials is used as output variables; and a detection step of detecting leaks in the exhaust gas duct based on the deviation between the oxygen concentration measured in the measurement step and the predicted values ​​of the oxygen concentration components calculated in the prediction step.

[0009] The prediction model is preferably a machine learning model that has been trained using a set of actual values ​​of information indicating the operating status of the Dwightroid-type sintering machine and the blending status of the sintering raw materials, and actual values ​​of the oxygen concentration component, as training data.

[0010] The leak detection device according to the present invention is a leak detection device for detecting leaks in the exhaust gas duct of a Dwightroid type sintering machine that fires sintering raw materials, and comprises: an oxygen concentration meter installed on the inlet side of an exhaust fan provided in the exhaust gas duct for measuring the oxygen concentration inside the exhaust gas duct; and a control device that inputs the information during processing to a prediction model in which information indicating the operating state of the Dwightroid type sintering machine and the mixing state of the sintering raw materials are input variables, and predicted values ​​of oxygen concentration components affected by the operating state of the Dwightroid type sintering machine and the mixing state of the sintering raw materials are output variables, calculates predicted values ​​of the oxygen concentration components during processing, and detects leaks in the exhaust gas duct based on the deviation between the oxygen concentration measured by the oxygen concentration meter and the calculated predicted values ​​of the oxygen concentration components. [Effects of the Invention]

[0011] According to the leak detection method and leak detection device of the present invention, leaks in the exhaust gas duct of a Dwightroid type sintering machine can be detected with high accuracy. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a schematic diagram showing the configuration of a sintering machine to which a leak detection method and leak detection device, which are embodiments of one embodiment of the present invention, are applied. [Figure 2] Figure 2 shows the components that make up the oxygen concentration inside the exhaust gas duct. [Figure 3] Figure 3 is a flowchart showing the flow of a leak detection process, which is one embodiment of the present invention. [Figure 4] Figure 4 shows an example of the air leak detection method according to the present invention. [Modes for carrying out the invention]

[0013] Hereinafter, with reference to the drawings, an embodiment of one embodiment of a leak detection method and a leak detection device will be described.

[0014] 〔composition〕 First, with reference to Figure 1, the configuration of a sintering machine to which a leak detection method and leak detection device, which are part of one embodiment of the present invention, are applied will be described.

[0015] Figure 1 is a schematic diagram showing the configuration of a sintering machine to which a leak detection method and leak detection device, which are part of one embodiment of the present invention, are applied. As shown in Figure 1, the sintering machine to which the leak detection method and leak detection device, which are part of one embodiment of the present invention, are applied is a Dwightroid type sintering machine 1. As shown in Figure 1, the Dwightroid type sintering machine 1 is equipped with a plurality of pallets 2, charging devices 3 (3a, 3b), an ignition furnace 4, a plurality of wind boxes 5 and wind legs 6, a main exhaust fan 7, and a crusher 8.

[0016] Multiple pallets 2 are endlessly connected between two sprockets 11a and 11b and circulate in the direction of the arrow. The charging device 3 is a device that deposits sintering raw materials in layers on the pallets 2. The ignition furnace 4 is a device that ignites the sintering raw materials deposited on the pallets 2. Multiple wind boxes 5 are devices that are continuously arranged in the circumferential direction of the multiple pallets 2 and draw air from the top to the bottom of the pallets 2 through the deposited sintering raw materials. The main exhaust fan 7 is a device that discharges the air drawn in by the multiple wind boxes 5 as exhaust gas to the outside via wind legs 6 and exhaust gas ducts 12 connected to the bottom of the wind boxes 5. The crusher 8 is a device that crushes the sintered ore produced by firing the sintering raw materials on the pallets 2.

[0017] In addition, the Dwight Lloyd type sintering machine 1 includes oxygen concentration meters 21 and 22 and a control device 23 as a control system. The oxygen concentration meters 21 and 22 are composed of oxygen concentration meters such as zirconia type oxygen concentration meters. The oxygen concentration meter 21 is arranged at the lower part of the wind leg 6, and the oxygen concentration meter 22 is arranged in the exhaust gas duct 12 on the inlet side of the main exhaust fan 7. The oxygen concentration meter 21 measures the oxygen concentration in the air sucked from the pallet 2 that has passed directly above the wind box 5, and inputs an electric signal indicating the measured oxygen concentration to the control device 23. The oxygen concentration meter 22 measures the oxygen concentration in the air sucked by the main exhaust fan 7, and inputs an electric signal indicating the measured oxygen concentration to the control device 23. The control device 23 is composed of an information processing device such as a computer, and controls the operation of the Dwight Lloyd type sintering machine 1. Further, in the present embodiment, the control device 23 detects air leakage in the exhaust gas duct 12 based on the oxygen concentration measured by the oxygen concentration meter 22, and when air leakage is detected, notifies information indicating that air leakage has occurred, etc.

[0018] In this Dwight Lloyd type sintering machine 1, first, sintering raw materials composed of iron ore powder, recovered powder in the steelworks, undersize sinter ore powder, CaO-containing raw materials (CaO-based auxiliary raw materials) such as limestone and dolomite, granulation aids such as quicklime, and coagulants such as coke powder and anthracite are stored in the charging device 3. Next, the sintering raw materials are charged from the charging device 3 onto each pallet 2 to form a raw material layer. Next, the ignition furnace 4 ignites the coagulant on the upper surface of the raw material layer, and by sucking air from each wind box 5, the coagulant is sequentially burned from the upper layer to the lower layer of the raw material layer, and the sintering raw materials are heated and melted by the combustion heat generated at this time. Thereby, a sintering layer (sinter cake) is generated. And finally, the sintering layer generated on each pallet 2 is crushed by the crusher 8 and recovered as sinter ore.

[0019] 〔Air leakage detection process〕 Next, referring to FIGS. 2 and 3, a method for detecting air leakage in the exhaust gas duct 12 of the Dwight Lloyd type sintering machine 1 shown in FIG. 1 will be described.

[0020] FIG. 2 is a diagram showing components constituting the oxygen concentration in the exhaust gas duct 12 measured by the oxygen concentration meter 22. As shown in FIG. 2, the oxygen concentration in the exhaust gas duct 12 measured by the oxygen concentration meter 22 includes components affected by fluctuations in the operating state of the Dwightroid type sintering machine 1 and the blending state of the raw materials for sintering (operation influence), components affected by air leakage in the pallet 2 and the wind box 5 (pallet - wind box air leakage influence), and components affected by air leakage in the exhaust gas duct 12 (duct air leakage influence). Therefore, in order to accurately detect the air leakage in the exhaust gas duct 12, it is necessary to remove the components of the operation influence and the pallet - wind box air leakage influence from the actual value of the oxygen concentration.

[0021] Therefore, in the present invention, by performing machine learning using a set of past operation performance data, blending state performance data of the raw materials for sintering, and the oxygen concentration measured by the oxygen concentration meter 21 as learning data, a machine learning model with the operation data of the Dwightroid type sintering machine 1 and the blending state data of the raw materials for sintering as input variables and the oxygen concentration corresponding to the sum of the components of the operation influence and the pallet - wind box air leakage influence as the output variable is created in advance as a prediction model. Then, when detecting the air leakage in the exhaust gas duct 12, the deviation between the oxygen concentration (actual value) measured by the oxygen concentration meter 22 and the oxygen concentration (predicted value) obtained by inputting the operation data of the Dwightroid type sintering machine 1 and the blending state data of the raw materials for sintering into the prediction model is calculated as the component of the duct air leakage influence.

[0022] Examples of the machine learning method include, for example, commonly used neural networks (including deep learning, convolutional neural networks, etc.), decision tree learning, random forest, support vector regression, etc. Also, a prediction model may be created using a method such as multiple regression analysis instead of machine learning.

[0023] Hereinafter, referring to FIG. 3, the flow of the air leakage detection process, which is an embodiment of the present invention based on the above idea, will be described.

[0024] Figure 3 is a flowchart showing the flow of the air leak detection process, which is one embodiment of the present invention. The flowchart shown in Figure 3 starts when the operation of the Dwightroid sintering machine 1 begins, and the air leak detection process proceeds to step S1.

[0025] In step S1, the control device 23 uses the oxygen concentration meter 22 to measure the oxygen concentration inside the exhaust gas duct 12. This completes step S1, and the leak detection process proceeds to step S2.

[0026] In step S2, the control device 23 inputs the operating data of the Dwightroid sintering machine 1 and the blending state data of the sintering raw materials into a prediction model to calculate a predicted value of oxygen concentration corresponding to the sum of the components of the operational influence and the components of the pallet / windbox air leakage influence. With this, the process of step S2 is completed, and the air leakage detection process proceeds to step S3.

[0027] In step S3, the control device 23 detects air leaks in the exhaust gas duct 12 based on the difference between the measured oxygen concentration obtained in step S1 and the predicted oxygen concentration obtained in step S2. If an air leak is detected, the control device 23 notifies the operator of information indicating that an air leak has occurred. Upon receiving notification from the control device 23, the operator performs repair work on the exhaust gas duct 12 where the air leak is occurring. This completes step S3, and the series of air leak detection processes is finished.

[0028] As is clear from the above explanation, in the leak detection process, which is one embodiment of the present invention, the control device 23 measures the oxygen concentration in the exhaust gas duct 12 using the oxygen concentration meter 22, inputs the operating data of the Dwightroid sintering machine 1 and the mixing state data of the sintering raw materials into a prediction model, calculates a predicted value of oxygen concentration corresponding to the sum of the components of the operational influence and the pallet / windbox leak influence, and detects leaks in the exhaust gas duct 12 based on the deviation between the measured value and the predicted value of the oxygen concentration. This makes it possible to accurately detect leaks in the exhaust gas duct 1, taking into account fluctuations in the operating state of the Dwightroid sintering machine 1 and the mixing state of the sintering raw materials. As a result, it is possible to suppress the decrease in the productivity of sintered ore due to leaks. [Examples]

[0029] Figure 4 shows an example of the air leak detection method according to the present invention. The vertical axis of Figure 4 shows the actual value of oxygen concentration and the deviation of oxygen concentration (Δoxygen concentration), where Δoxygen concentration = 0 indicates that there is no change in oxygen concentration due to air leak in the exhaust gas duct 12. Furthermore, Δoxygen concentration > 0 means that the amount of air leak has increased compared to the amount of air leak during the learning period, and Δoxygen concentration < 0 means that the amount of air leak has decreased compared to the amount of air leak during the learning period. The explanatory variables used to train the prediction model are shown in Table 1. In Table 1, WB# (number) is the identification number of the windbox 5, and combined damper opening degree indicates the opening degree of the damper provided in the exhaust gas duct 12.

[0030] As shown in Figure 4, after the repair work performed on day 50, as indicated in the graph, the Δ-oxygen concentration becomes negative, indicating that the amount of air leakage in the exhaust gas duct 12 has decreased due to the repair. However, the Δ-oxygen concentration increases sharply on day 60. This indicates that a new leak occurred on day 60. Thus, it has been confirmed that the leak detection method according to the present invention can accurately detect air leakage in the exhaust gas duct 12.

[0031] [Table 1]

[0032] Although embodiments applying the invention made by the present inventors have been described above, the present invention is not limited by the descriptions and drawings that constitute part of the disclosure of the present invention in this embodiment. That is, all other embodiments, examples, and operational techniques made by those skilled in the art based on this embodiment are included in the scope of the present invention. [Explanation of symbols]

[0033] 1. Dwightroid type sintering machine 2 pallets 3,3a,3b Charging device 4 Ignition furnace 5 Window Box 6 Wind Legs 7 Main exhaust fan 8. Crusher 11a, 11b sprocket 12 Exhaust duct 21,22 Oxygen concentration meter 23 Control device

Claims

1. A leak detection method for detecting leaks in the exhaust gas duct of a Dwightroid type sintering machine used for firing sintering raw materials, A measurement step of measuring the oxygen concentration inside the exhaust gas duct using an oxygen concentration meter installed on the inlet side of the exhaust fan installed in the exhaust gas duct, A prediction step involves inputting the information during processing into a prediction model, in which information indicating the operating state of the Dwightroid sintering machine and the blending state of the sintering raw materials is used as input variables, and predicted values ​​of the oxygen concentration components affected by the operating state of the Dwightroid sintering machine and the blending state of the sintering raw materials are used as output variables, thereby calculating predicted values ​​of the oxygen concentration components during processing. A detection step for detecting air leakage in an exhaust gas duct based on the deviation between the oxygen concentration measured in the measurement step and the predicted value of the oxygen concentration component calculated in the prediction step, A method for detecting air leaks, including the detection of air leaks.

2. The leak detection method according to claim 1, wherein the prediction model is a machine learning model trained using a set of actual values ​​of information indicating the operating state of the Dwightroid type sintering machine and the blending state of the sintering raw materials and actual values ​​of the oxygen concentration component as training data.

3. A leak detection device for detecting leaks in the exhaust gas duct of a Dwightroid-type sintering machine used for firing sintering raw materials, An oxygen concentration meter installed on the inlet side of the exhaust fan provided in the exhaust gas duct for measuring the oxygen concentration inside the exhaust gas duct, A control device that inputs information indicating the operating state of the Dwightroid-type sintering machine and the blending state of the sintering raw materials as input variables, and outputs predicted values ​​of oxygen concentration components affected by the operating state of the Dwightroid-type sintering machine and the blending state of the sintering raw materials as output variables, calculates predicted values ​​of the oxygen concentration components during processing by inputting the information during processing, and detects air leakage in the exhaust gas duct based on the deviation between the oxygen concentration measured by an oxygen concentration meter and the calculated predicted values ​​of the oxygen concentration components, A leak detection device equipped with the following features.

Images