JP2021139599A - Continuous firing furnace and continuous firing method - Google Patents

Abstract

To provide a continuous firing furnace and a continuous firing method capable of easily performing the control of firing conditions such as the temperature and atmosphere in a muffle.SOLUTION: A continuous firing furnace has a process where, while continuously moving a body to be fired in a muffle comprising an inlet part and an outlet part, firing is continuously performed. The continuous firing furnace comprises a detection section detecting the conditions in a continuous firing furnace during firing and a determination section determining firing conditions from information detected by the detection section using a learning-finished learner. The detection section detects at least one of the temperature in the muffle and the atmosphere in the muffle.SELECTED DRAWING: Figure 9

Description

The present invention relates to a continuous firing furnace and a continuous firing method.

In Patent Document 1, the degreased body, which is the body to be fired, is continuously carried into the furnace from the carry-in port, continuously fired while moving in the muffle, and the fired body is continuously carried out from the carry-out port. Describes the firing furnace.

International Publication No. 2006/013652

In the continuous firing furnace of Patent Document 1 and the like, since firing is performed while moving the degreased body in the muffle, the length of the muffle becomes relatively long. Therefore, there is a problem that it is difficult to control the firing conditions such as the temperature and atmosphere in the muffle to a state suitable for firing. Further, since it is difficult to judge the quality of the firing conditions without measuring the fired body after firing, it takes time until the determination result is obtained. If it becomes necessary to change the firing conditions according to the determination result, there is a risk that the degreased body moving in the furnace will be wasted. The present invention has been made in view of these circumstances, and an object of the present invention is to provide a continuous firing furnace and a continuous firing method that enable easy management of firing conditions.

The continuous firing furnace of the present invention for solving the above problems is a continuous firing furnace in which a body to be fired is continuously fired while being moved from the inlet portion to the outlet portion in a muffle having an inlet portion and an outlet portion. The gist is that it has a detection unit that detects the state in the continuous firing furnace during firing and a determination unit that determines firing conditions from the information detected by the detection unit using a learned learning device. And.

According to the above configuration, the determination unit can determine the firing conditions more quickly and more accurately by determining the information detected by the detection unit using the learned learner. Therefore, the firing conditions can be easily managed.

For the continuous firing furnace of the present invention, it is preferable that the detection unit detects at least one of the temperature in the muffle and the atmosphere in the muffle. According to this configuration, the firing conditions can be determined more accurately by detecting at least one of the temperature in the muffle and the atmosphere in the muffle, which have a great influence on the firing condition.

It is preferable that the continuous firing furnace of the present invention has an instruction unit for instructing a change in firing conditions based on the determination result of the determination unit. According to this configuration, the firing conditions can be easily managed by changing the firing conditions according to the instructions of the instruction unit.

It is preferable that the continuous firing furnace of the present invention has a learning unit for further learning the learning device based on the state after firing of the body to be fired. According to this configuration, the accuracy of the determination result in the determination unit can be improved.

The continuous firing method of the present invention for solving the above problems is a continuous firing method in which a body to be fired is continuously fired while being moved from the inlet portion to the outlet portion in a muffle having an inlet portion and an outlet portion. The gist is that the detection unit detects the state in the continuous firing furnace during firing, and the determination unit determines the firing conditions from the information detected by the detection unit using a learned learning device. And.

According to the above configuration, by determining the information detected by the detection unit using the learned learner, it is possible to determine the firing conditions more quickly and more accurately. Therefore, the firing conditions can be easily managed.

Regarding the continuous firing method of the present invention, it is preferable that the detection unit detects at least one of the temperature in the muffle and the atmosphere in the muffle. According to this configuration, the firing conditions can be determined more accurately by detecting at least one of the temperature in the muffle and the atmosphere in the muffle, which have a great influence on the firing condition.

Regarding the continuous firing method of the present invention, it is preferable that the indicating unit instructs the change of the firing conditions based on the determination result of the determination unit. According to this configuration, the firing conditions can be easily managed by changing the firing conditions according to the instructions of the instruction unit.

Regarding the continuous firing method of the present invention, it is preferable that the learning unit further learns the learning device based on the state after firing of the body to be fired. According to this configuration, the accuracy of the determination result in the determination unit can be improved.

According to the continuous firing furnace and the continuous firing method of the present invention, the firing conditions can be easily managed.

(A) is a horizontal sectional view of a continuous firing furnace, and (b) is a vertical sectional view of a continuous firing furnace. A vertical sectional view of a heating chamber of a continuous firing furnace cut in the width direction. A vertical sectional view of a preheating chamber of a continuous firing furnace cut in the width direction. Perspective view of the honeycomb structure. The perspective view of the porous ceramic member used for the honeycomb structure. FIG. 5 is a sectional view taken along line 6-6 of FIG. A graph showing the relationship between average pore size and lightness. The schematic diagram which shows the relationship between the average pore diameter and the lightness. Correlation diagram of the predicted value of lightness using the trained learner and the measured value of lightness.

An embodiment of a continuous firing furnace will be described.
As shown in FIGS. 1A and 1B, the continuous firing furnace 10 includes a cooling furnace material 11 such as a cooling jacket constituting an outer wall and a heat insulating material 12 provided inside the cooling furnace material 11. It is provided with a tubular muffler 13 provided inside the heat insulating material 12, and has a shape in which the entire muffler 13 extends long along the axial direction. The body 14 to be fired (see FIG. 2) is configured to be continuously fired while moving in the muffle 13.

The continuous firing furnace 10 includes a sensor (not shown) as a detection unit for detecting a state in the continuous firing furnace 10 during firing, and a computer 10a for acquiring information detected by the sensor. The computer 10a has a learned learner, and is configured so that the firing condition can be determined from the information detected by the sensor using the learned learner. In other words, the computer 10a has a determination unit that determines the firing condition from the information detected by the sensor using the learned learner.

Further, the computer 10a has an instruction unit for instructing the change of the firing condition based on the determination result of the determination unit, and a learning unit for further learning the learning device based on the state after firing of the object to be fired. doing. The sensor, the determination unit, the instruction unit, and the learning unit as the detection unit will be described later.

The member composition of the continuous firing furnace 10 will be described.
As shown in FIGS. 1A and 1B, the inlet side degassing chamber (hereinafter, also referred to as "first degassing chamber") 20 is inside the cooling furnace material 11 of the continuous firing furnace 10. The preheating chamber 21, the heating chamber 22, the slow cooling chamber 23, the cooling chamber 24, and the outlet side degassing chamber (hereinafter, also referred to as “second degassing chamber”) 25 are provided in this order.

The first degassing chamber 20 is provided at one end of the continuous firing furnace 10 in the longitudinal direction. A carry-in port (not shown) of the body 14 to be fired is provided at one end of the first degassing chamber 20. At the other end of the first degassing chamber 20, a communication port 20a communicating with the preheating chamber 21 is provided. The communication port 20a is provided with a partition door 20b that separates the communication port 20a from the preheating chamber 21.

The second degassing chamber 25 is provided at the other end of the continuous firing furnace 10 in the longitudinal direction. At the other end of the second degassing chamber 25, a carry-out port (not shown) of the fired body to be fired (hereinafter, also simply referred to as “fired body”) is provided. A communication port 25a communicating with the cooling chamber 24 is provided at one end of the second degassing chamber 25. The communication port 25a is provided with a partition door 25b that separates the communication port 25a from the cooling chamber 24.

Inside the cooling furnace material 11, a heat insulating material 12 is arranged between the first degassing chamber 20 and the second degassing chamber 25 along the longitudinal direction of the continuous firing furnace 10. A cylindrical muffle 13 is arranged inside the heat insulating material 12.

As shown in FIGS. 1A and 1B, one end of the muffle 13 is located at the communication port 20a of the first degassing chamber 20, and the other end of the muffle 13 is located on the other end. It is located at the communication port 25a of the second degassing chamber 25. Therefore, the end portion on one end side of the muffle 13 functions as an inlet portion of the fired body 14, and the end portion on the other end side of the muffle 13 functions as an outlet portion of the fired body 14. Inside the muffle 13, a space constituting a preheating chamber 21, a heating chamber 22, a slow cooling chamber 23, and a cooling chamber 24 is provided. These spaces communicate with each other within the muffle 13.

As shown in FIGS. Is located in. In other words, the heat insulating material 12 is not arranged outside the portion of the muffle 13 that constitutes the cooling chamber 24.

A plurality of heaters 15 are arranged outside the portion of the muffle 13 that constitutes the heating chamber 22 and with the heat insulating material 12 at a predetermined interval. The plurality of heaters 15 are connected to an external power source (not shown) via terminals 15a.

As shown in FIG. 2, the cooling furnace material 11 located outside the heating chamber 22 is provided with an introduction pipe 16. An inert gas is introduced from the introduction pipe 16 between the cooling furnace material 11 and the heat insulating material 12. The inert gas passes through the gap of the heat insulating material 12 and the heat insulating material 12 and is introduced into the muffle 13.

As shown in FIG. 3, the cooling furnace material 11, the heat insulating material 12, and the muffler 13 located outside the preheating chamber 21 are provided with an exhaust pipe 17 in a state of penetrating them. The inert gas introduced into the muffler 13 is discharged from the exhaust pipe 17.

As shown in FIGS. 2 and 3, the fired body 14 is placed in a box-shaped firing jig 18 having a lid, and moves in the muffle 13 with a plurality of the firing jigs 18 stacked. It is configured in. Hereinafter, a product in which a plurality of firing jigs 18 are laminated is referred to as a laminating jig 19.

The firing procedure using the continuous firing furnace 10 will be described.
As shown in FIGS. 1A and 1B, the laminating jig 19 on which the body to be fired 14 is placed is carried into the first degassing chamber 20 from the carry-in port of the continuous firing furnace 10. After degassing the inside of the first degassing chamber 20 to evacuate the inside, an inert gas is introduced into the first degassing chamber 20. The inert gas is not particularly limited, and for example, argon gas can be used. By introducing the inert gas into the first degassing chamber 20, the inside of the laminating jig 19 is made into an inert gas atmosphere.

Similarly, the inside of the continuous firing furnace 10, specifically, the inside of the cooling furnace material 11 which is the outer wall of the continuous firing furnace 10 is also set to have an inert gas atmosphere. As a result, the preheating chamber 21, the heating chamber 22, the slow cooling chamber 23, and the cooling chamber 24 in the muffle 13 also have an inert gas atmosphere.

Next, the partition door 20b of the first degassing chamber 20 is opened, and the laminating jig 19 is made to enter the preheating chamber 21 through the communication port 20a. At this time, by pressing the rear end side of the laminating jig 19 with a transport jig (not shown), the laminating jig 19 moves into the preheating chamber 21. The laminating jig 19 that has entered the preheating chamber 21 is heated by the heat of the heater 15 provided outside the heating chamber 22.

Further, another laminating jig 19 is carried into the first degassing chamber 20 to make the first degassing chamber 20 an inert gas atmosphere in the same manner. The laminating jig 19 is brought into the preheating chamber 21 using a transport jig. When the laminating jig 19 enters the preheating chamber 21, the rear end side of the laminating jig 19 that first entered the preheating chamber 21 is pushed, so that the laminating jig 19 that first enters the preheating chamber 21 is heated. Move to the room 22 side. By repeating this operation and allowing the plurality of laminating jigs 19 to enter the muffle, the laminating jig 19 intermittently moves in the muffle 13 toward the second degassing chamber 25 side.

The laminating jig 19 that has entered the preheating chamber 21 is gradually heated by the heat of the heater 15 in the heating chamber 22. As the laminating jig 19 moves to the heating chamber 22, the temperature of the laminating jig 19 rises. When the laminating jig 19 reaches a certain point, the temperature inside the laminating jig 19 reaches the firing temperature of the body 14 to be fired. The body 14 to be fired is continuously fired while moving in the heating chamber 22 after reaching the firing temperature.

Since the heater 15 is not arranged outside the slow cooling chamber 23 in the muffle 13, when the laminating jig 19 moves to the slow cooling chamber 23, the temperature of the laminating jig 19 gradually decreases. The laminating jig 19 is further cooled as it moves from the slow cooling chamber 23 to the cooling chamber 24. Then, using a transfer jig (not shown), the stacking jig 19 is ejected from the communication port 25a to the second degassing chamber 25.

After returning the atmosphere in the second degassing chamber 25 from the inert gas atmosphere to the atmospheric atmosphere, the laminating jig 19 is carried out from the carry-out port of the continuous firing furnace 10. The body 14 to be fired can be fired by the above procedure.

The fired body 14 will be described.
The body 14 to be fired in the continuous firing furnace 10 is not particularly limited, and examples thereof include a degreasing body that becomes a porous ceramic member containing silicon carbide as a main component by firing. Here, the degreased body means a body obtained by molding a raw material composition of silicon carbide and heating a molded body to eliminate organic substances such as a binder contained in the molded body.

As shown in FIGS. 4 and 5, the porous ceramic 30 is produced by performing an assembly step of binding a plurality of porous ceramic members 31 produced by firing a degreased body via a sealing material layer.

As shown in FIG. 6, the porous ceramic member 31 is formed with a plurality of through holes 32 extending in the longitudinal direction, and the partition wall 33 separating the through holes 32 is formed to be porous. The average pore diameter of the partition wall 33 is set to, for example, 5 to 25 μm. One of the plurality of through holes 32 at both ends in the longitudinal direction is sealed with a sealing material 34. Therefore, the gas that has flowed into one through hole 32 is configured to always pass through the partition wall 33 that separates the through holes 32 and then flow out from the other through holes 32.

When the exhaust gas flows into one through hole 32, the exhaust gas passes through the partition wall 33 separating the through holes 32 and then flows out from the other through holes 32. At that time, the exhaust gas is purified by capturing the fine particles in the exhaust gas by the partition wall 33. Therefore, the porous ceramic 30 can be used as a particle collecting filter.

A sensor as a detection unit included in the continuous firing furnace 10 will be described.
A plurality of sensors as detection units are arranged inside the continuous firing furnace 10.
The type of sensor is not particularly limited, and examples thereof include a temperature sensor, a gas sensor, a voltage sensor, a current sensor, a power sensor, and a gas flow sensor. Examples of the gas sensor include a sensor for measuring carbon monoxide concentration and a sensor for measuring oxygen concentration.

The place where each sensor is attached is not particularly limited, but the temperature sensor may be attached, for example, in the first degassing chamber 20 or the second degassing chamber 25 to detect the temperature in each degassing chamber. .. Further, it may be attached to the preheating chamber 21, the heating chamber 22, the slow cooling chamber 23, and the cooling chamber 24 in the muffle 13 to detect the temperature of each chamber. The temperature of the cooling furnace material 11 or the heat insulating material 12 may be detected by being attached to the cooling furnace material 11 or the heat insulating material 12.

The place where the gas sensor is attached is not particularly limited, but may be attached to the inert gas introduction pipe 16 or the exhaust pipe 17 to detect the gas concentration, for example. The gas sensor can detect the atmosphere in the continuous firing furnace 10, particularly the atmosphere in the muffle 13.

The place where the voltage sensor, the current sensor, and the power sensor are attached is not particularly limited, but may be attached to the terminals 15a of the plurality of heaters 15 to detect the voltage, current, and power of the plurality of heaters 15.

The place where the gas flow rate sensor is attached is not particularly limited, but for example, it is attached to the inert gas introduction pipe 16 or the exhaust pipe 17 of the continuous firing furnace 10, and the amount of gas flowing into the continuous firing furnace 10 and the continuous firing furnace The amount of gas discharged from within 10 may be detected.

The information detected by each of the above sensors is configured to be transmitted to the computer 10a included in the continuous firing furnace 10 by wired communication or wireless communication. Since the plurality of sensors are arranged inside the continuous firing furnace 10, it is possible to detect the state inside the continuous firing furnace 10 during firing.

The determination mechanism by the determination unit will be described.
The determination unit determines the firing condition from the information detected by the sensor using the learned learner.

First, the trained learner predicts the brightness of the fired body when it is fired under the current firing conditions based on the information detected by the sensor.
Here, the brightness of the fired body means the brightness of the color of the surface of the fired body. The brightness of the fired body changes depending on the reflectance of light on the surface of the fired body, and can be measured by photographing the surface of the fired body with a camera.

As shown in FIG. 7, there is a correlation between the average pore diameter of the partition wall of the fired body and the measured value of lightness. Specifically, as shown in the linear shape in FIG. 7, the brightness of the fired body tends to decrease as the average pore diameter of the partition wall of the fired body increases.

As shown in FIG. 8, a comparison between the fired body having an average pore diameter of 8 μm (electron micrograph on the left side in the figure) and the fired body having an average pore diameter of 12 μm (electron micrograph on the right side in the figure) is compared. A fired body having an average pore diameter of 8 μm has many smaller pores and therefore easily reflects light. Therefore, the lightness value of the fired body having an average pore diameter of 8 μm is larger than that of the fired body having an average pore diameter of 12 μm.

If the firing temperature is too low or the firing time is too short, firing is insufficient and the average pore diameter tends to be small. Therefore, the brightness of the fired body tends to increase. On the contrary, if the firing temperature is too high or the firing time is too long, the firing proceeds and the average pore diameter tends to increase. Therefore, the brightness of the fired body tends to be small.

This makes it possible to evaluate the firing conditions based on the average pore size and brightness of the fired body. That is, if the average pore diameter and the brightness of the fired body are within a certain numerical range, it can be judged that the firing conditions are good.

In particular, the brightness of the fired body can be obtained relatively easily and in a short time. For example, before performing the assembly process after firing, the surfaces of all fired bodies can be imaged with a camera to measure the brightness of all fired bodies. Since it becomes easy to prepare teacher data for machine learning the learner, it becomes easy to improve the prediction accuracy of the learner.

The trained learner is constructed by using a neural network, a decision tree, discriminant analysis, and the like. Among these, it is preferable to use a decision tree, and it is more preferable to construct using a random forest which is a kind of decision tree.

The learner performs machine learning by using the information detected by the sensor and the brightness of the actually fired body as the teacher data.
As shown in FIG. 9, the predicted value of the brightness of the fired body (the value indicated by the circle in the figure) predicted by using the trained learner is the measured value of the brightness of the fired body so as to show the linearity. The values are close to each other, and the brightness of the fired body can be predicted with high accuracy.

Next, the determination unit of the computer 10a determines the firing conditions from the brightness of the fired body predicted by the learned learner. When the predicted brightness of the fired body is within a predetermined numerical range, the firing conditions are judged to be good. Further, when the predicted brightness of the fired body is not within the predetermined numerical range, the firing condition is determined to be negative.

If it is determined that the firing conditions are good, the display indicating that the firing conditions are good is displayed on the display of the computer 10a, and the firing is continued. Instead of displaying on the display of the computer 10a, a separate display unit may be provided near the continuous firing furnace 10 and displayed on this display unit. If it is determined to be good, it may not be displayed on the display, and it may be configured to be displayed on the display only when it is determined to be negative, which will be described later.

If the firing condition is determined to be negative, the fact that the firing condition is negative is displayed on the display of the computer 10a. Instead of the display of the computer 10a, a separate display unit may be provided near the continuous firing furnace 10 and displayed on this display unit. Further, it may be configured to be known by light or sound.

Furthermore, if it is determined that the firing conditions are negative, a trained learner is used to display how the firing conditions should be changed in order to keep the predicted brightness of the fired body within a good numerical range. Then, the worker may be instructed. Therefore, the computer 10a may have an instruction unit for instructing the change of the firing conditions. It may be configured to automatically change the firing conditions without instructing the change of the firing conditions.

The computer 10a may be configured to further learn the learner by using the measured value of the brightness of the fired body. Therefore, the computer 10a may have a learning unit for further learning the learning device.

The operation and effect of this embodiment will be described.
(1) The brightness of the fired body is predicted using a learned learner, and the firing conditions are determined based on the predicted value of the brightness. Therefore, since the firing conditions can be determined more quickly and more accurately, the firing conditions can be easily managed.

(2) When determining the firing conditions after measuring the brightness using the fired body after firing, it takes time for the judgment result to be obtained, so that the degreased body moving in the continuous firing furnace is wasted. There is a risk. On the other hand, in the present embodiment, since the lightness of the fired body is predicted by the learned learner and the firing conditions are determined, the determination result can be obtained at an earlier timing. By changing the firing conditions at the stage when the determination result is obtained, it is possible to reduce the waste of the degreased body moving in the continuous firing furnace.

(3) By detecting the temperature inside the muffle and the atmosphere inside the muffle, which have a great influence on the firing condition, the firing conditions can be determined more accurately.
(4) The trained learner predicts the brightness of the fired body. Since the measured value of the lightness of the fired body can be obtained relatively easily and in a short time, the lightness can be measured for the entire number of the fired bodies. Therefore, since it is easy to prepare the teacher data for machine learning the learner, it becomes easy to improve the prediction accuracy of the learner.

(5) It has an instruction unit for instructing a change in firing conditions. Therefore, the firing conditions can be easily managed by changing the firing conditions according to the instructions of the indicating unit.
(6) It has a learning unit for further learning the learning device based on the state of the fired body. Therefore, the accuracy of the determination result in the determination unit can be improved.

This embodiment can be modified and implemented as follows. The present embodiment and the following modified examples can be implemented in combination with each other within a technically consistent range.
-In the present embodiment, the fired body is a degreased body which becomes a porous ceramic member containing silicon carbide as a main component by firing, but is not limited to this embodiment. For example, in a continuous firing furnace, the body to be fired may be a molded body containing silicon carbide as a main component, as long as the molded body can be continuously degreased and fired.

-In the present embodiment, the sensor detects both the temperature and the atmosphere in the muffle, but is not limited to this embodiment. The sensor may be configured to detect either the temperature or the atmosphere in the muffle and not the other.

-In the present embodiment, the trained learner predicts the brightness of the fired body, but the learning device is not limited to this embodiment. The trained learner may predict other factors that can determine the firing conditions. Other factors include, for example, the average pore size of the fired body and the strength of the fired body.

-In the present embodiment, the instruction unit of the computer has instructed how to change the firing conditions, but the present invention is not limited to this embodiment. The instruction unit of the computer may only instruct that the firing conditions should be changed, and may not instruct how the firing conditions should be changed.

10 ... continuous firing furnace, 13 ... muffle, 14 ... fired body.

Claims (8)

A continuous firing furnace in which a body to be fired is continuously fired while being moved from the inlet portion to the outlet portion in a muffle having an inlet portion and an outlet portion.
A detection unit that detects the state in the continuous firing furnace during firing, and
A continuous firing furnace characterized by having a determination unit that determines firing conditions from information detected by the detection unit using a learned learning device. The continuous firing furnace according to claim 1, wherein the detection unit detects at least one of the temperature in the muffle and the atmosphere in the muffle. The continuous firing furnace according to claim 1 or 2, which has an instruction unit for instructing a change in firing conditions based on the determination result of the determination unit. The continuous firing furnace according to any one of claims 1 to 3, further comprising a learning unit for further learning the learner based on the state after firing of the body to be fired. A continuous firing method in which a body to be fired is continuously fired while moving from the inlet portion to the outlet portion in a muffle having an inlet portion and an outlet portion.
The detection unit detects the state inside the continuous firing furnace during firing,
A continuous firing method characterized in that a determination unit determines firing conditions from information detected by the detection unit using a learned learning device. The continuous firing method according to claim 5, wherein the detection unit detects at least one of the temperature in the muffle and the atmosphere in the muffle. The continuous firing method according to claim 5 or 6, wherein the instruction unit instructs the change of the firing conditions based on the determination result of the determination unit. The continuous firing method according to any one of claims 5 to 7, wherein the learning unit further learns the learning device based on the state after firing of the body to be fired.

Images