JP7009766B2 - Carbon adhesion state analysis device, carbon adhesion state analysis method, computer program and computer-readable recording medium - Google Patents

Description

The present invention relates to a carbon adhesion state analysis device for identifying a carbon adhesion position having a large influence on an extrusion load on a furnace wall of a coke oven, a carbon adhesion state analysis method, a computer program, and a computer-readable recording medium.

In the coke oven, it is required to manage the operation so that the extrusion load of coke is minimized (that is, the operation is low load) from the viewpoint of stable operation and extension of the life of the furnace body. The extrusion load is affected by various factors such as coal composition, moisture content, and carbonization state, but in particular, the unevenness of the furnace wall surface greatly affects the extrusion load as the frictional effect between the coke and the furnace wall. The unevenness of the wall surface of the furnace fluctuates depending on the degree of carbon adhesion, and when the amount of carbon adhesion is large, convex carbon appears and leads to an increase in frictional resistance. It leads to an increase. Therefore, it is important to control the carbon adhering to the furnace wall in order to minimize the extrusion load.

It has been empirically known that carbon is attached to the furnace wall of a coke oven and the extrusion load changes depending on the amount of carbon attached. For example, in Patent Document 1, for the carbonization chamber of a coke oven, the state of a healthy brick surface, the state of uniform carbon adhesion, and the state of mottled carbon are determined from a furnace wall image (thermal image) using a concentration histogram to determine the surface state of the furnace wall. The device is disclosed. Further, in Patent Document 2, for the carbonization chamber of the coke oven, the sound brick surface state, the uniform carbon adhesion state, and the mottled carbon state are determined from the furnace wall image (thermal image) using the concentration co-occurrence matrix as a feature quantity. A furnace wall surface condition determination device is disclosed.

Japanese Unexamined Patent Publication No. 2016-60777 Japanese Unexamined Patent Publication No. 2016-65225 Japanese Unexamined Patent Publication No. 2008-146621

Roger Koenker, Gilbert Bassett Jr .: Regression Quantiles; Econometrica, vol.46, Jan.1978, pp.33-50.

However, the techniques described in Patent Documents 1 and 2 only quantitatively evaluate the carbon adhesion state, and do not specify the detailed relationship between the carbon adhesion state and the extrusion load. That is, the amount of carbon adhering to the furnace wall can be roughly predicted from the CO ₂ concentration measured at the time of carbon incineration by the lance performed after coke extrusion, or the amount of carbon adhering to the furnace wall can be estimated from the image of the furnace wall. The amount of carbon adhesion is determined by analysis, and the quantitative relationship between the amount of carbon adhesion and the extrusion load has not been clarified for each location of the furnace wall.

Therefore, the present invention has been made in view of the above problems, and an object of the present invention is to clarify the quantitative relationship between the carbon adhesion amount and the extrusion load for each location of the furnace wall. It is an object of the present invention to provide a new and improved carbon adhesion state analyzer, a carbon adhesion state analysis method, a computer program, and a computer-readable recording medium.

In order to solve the above problems, according to a certain viewpoint of the present invention, a plurality of furnace wall images acquired as actual data in the past operation of the coke oven are analyzed, and the carbon adhesion state in each furnace wall image is determined. The carbon determination unit and each furnace wall image are divided into a plurality of small areas, and each small area set at the same position in each furnace wall image is small based on the determination result of the carbon adhesion state by the carbon determination unit. A correlation analysis unit that analyzes the correlation between the carbon adhesion area of the region and the extrusion load when the furnace wall image including the small region is acquired, and the carbon of each small region obtained by the correlation analysis unit. Provided is a carbon adhesion state analysis apparatus including an influence region specifying portion for specifying an influence region affecting the coke extrusion load based on the correlation between the adhesion area and the coke extrusion load.

The affected area identification part is negative with the small area identified as having a positive correlation on the entire furnace wall surface based on the correlation between the carbon adhesion area of each small area obtained by the correlation analysis part and the extruded load of coke. The distribution with the small region identified as having a correlation may be obtained, and the region of influence that affects the extrusion load of coke may be specified by distinguishing between the positive correlation and the negative correlation. Further, the correlation analysis unit may calculate the carbon adhesion area for each small region and calculate the correlation with the extrusion load by using simple regression analysis or correlation analysis.

The affected area identification section is further limited to the specified affected area that has a greater effect on the coke extrusion load, based on an index indicating the certainty of the correlation coefficient obtained by the dispersion ratio or the uncorrelated test. The affected area may be specified.

The carbon adhesion state analysis device calculates the carbon adhesion area between the positively correlated influence region and the negatively correlated influence region specified by the influence region identification part, and at least one of the carbon adhesion areas of each influence region. An appropriate range calculation unit may be further provided to analyze the relationship between one and the coke extrusion load and calculate the appropriate range of the carbon adhesion area where the extrusion load is equal to or less than the control value.

The appropriate range calculation unit may calculate the relationship between the carbon adhesion area in the affected area and the coke extrusion load by using probability distribution correlation analysis or quantile regression.

Further, the carbon adhesion state analysis device calculates the carbon adhesion area in the influence area specified by the influence area identification part for the furnace wall image for determining whether the carbon adhesion state is appropriate, and the calculated carbon adhesion area and the calculated carbon adhesion area. A determination processing unit for determining whether or not the carbon is in an appropriate carbon adhesion state may be further provided by comparing with the appropriate range calculated by the appropriate range calculation unit.

When the calculated carbon adhesion area is out of the appropriate range, the determination processing unit adjusts the amount of carbon adhesion to the furnace wall according to the correlation between the carbon adhesion area in the affected area and the coke extrusion load. The amount of air in the lance may be adjusted.

The correlation analysis unit may divide the furnace wall image into a mesh and set a small area.

Further, in order to solve the above problems, according to another viewpoint of the present invention, a plurality of furnace wall images acquired as actual data in the past operation of the coke oven are analyzed, and the carbon adhesion state in each furnace wall image is analyzed. The carbon determination step for determining the above and each furnace wall image are divided into a plurality of small regions, and the small regions set at the same position in each furnace wall image based on the determination result of the carbon adhesion state in the carbon determination step. For each, a correlation analysis step for analyzing the correlation between the carbon adhesion area of a small region and the extrusion load when the furnace wall image including the small region is acquired, and the carbon adhesion area and coke of each small region. A carbon adhesion state analysis method is provided, which comprises an affected region specifying step for identifying an affected region that affects the extruded load of coke based on the correlation with the extrusion load of the coke.

Further, in order to solve the above-mentioned problems, according to another aspect of the present invention, the computer analyzes a plurality of furnace wall images acquired as actual data in the past operation of the coke oven, and in each furnace wall image. The carbon determination unit for determining the carbon adhesion state and each furnace wall image were divided into a plurality of small regions, and were set at the same position on each furnace wall image based on the determination result of the carbon adhesion state by the carbon determination unit. For each small region, obtained by the correlation analysis unit and the correlation analysis unit that analyze the correlation between the carbon adhesion area of the small region and the extrusion load when the furnace wall image including the small region is acquired. It functions as a carbon adhesion state analysis device provided with an influence region specifying part that specifies an influence region that affects the coke extrusion load based on the correlation between the carbon adhesion area of each small region and the coke extrusion load. A computer program is provided.

In addition, a computer analyzes a plurality of furnace wall images acquired as actual data in the past operation of the coke oven, and determines a carbon adhesion state in each furnace wall image, and a plurality of each furnace wall image. The carbon adhesion area of the small region and the small region are included for each of the small regions set at the same position in each furnace wall image based on the determination result of the carbon adhesion state by the carbon determination unit. The correlation between the carbon adhesion area of each small region and the coke extrusion load obtained by the correlation analysis unit and the correlation analysis unit that analyzes the correlation with the extrusion load when the furnace wall image is acquired . Based on this, a computer-readable recording medium is provided, comprising a region of influence identifying an region of influence that affects the extrusion load of coke, and recording a program for functioning as a carbon adhesion state analyzer.

As described above, according to the present invention, it is possible to clarify the quantitative relationship between the carbon adhesion amount and the extrusion load for each location of the furnace wall. By quantitatively clarifying the relationship between the carbon adhesion amount and the extrusion load for each location of the furnace wall, for example, the optimum furnace wall management target location and carbon at that location for appropriately controlling the carbon adhesion amount. It is possible to determine the control standard for the adhesion area (hereinafter, also referred to as "carbon control value").

It is explanatory drawing which shows the outline of the carbon adhesion state analysis by the carbon adhesion state analysis apparatus which concerns on one Embodiment of this invention. It is a block diagram which shows the functional structure of the carbon adhesion state analysis apparatus which concerns on the same embodiment. It is a flowchart which shows the analysis processing by the analysis processing part which concerns on the same embodiment. It is explanatory drawing which shows an example of the carbon adhesion position specified from the furnace wall image, and the state which mesh-divided the furnace wall image. It is explanatory drawing which shows the outline about the analysis of the correlation between the carbon adhesion area and the extrusion load performed for each mesh. It is explanatory drawing which shows an example which obtained the correlation between the carbon adhesion area and the extrusion load of coke by simple regression analysis. It is explanatory drawing which shows an example which obtained the correlation between the carbon adhesion area and the extrusion load of coke by the correlation analysis. It is explanatory drawing which shows an example of the influence area (limited influence area) specified by the influence area identification part. It is explanatory drawing which shows the calculation example of the carbon adhesion area of the influence area (limited influence area) by the appropriate range calculation part. It is explanatory drawing which shows an example of the appropriate range determined by the probability distribution correlation analysis method in one dimension. It is explanatory drawing which shows an example of the appropriate range determined by the probability distribution correlation analysis method in two dimensions. It is a flowchart which shows the determination process of the carbon adhesion area by the determination process part. It is explanatory drawing which shows the other example which divides the furnace wall surface into a small area. It is a block diagram which shows the hardware configuration example of the carbon adhesion state analysis apparatus which concerns on the same embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the present specification and the drawings, components having substantially the same functional configuration are designated by the same reference numerals, so that duplicate description will be omitted.

<1. Overview>
First, with reference to FIG. 1, the outline of the carbon adhesion state analysis by the carbon adhesion state analysis device according to the embodiment of the present invention will be described. Note that FIG. 1 is an explanatory diagram showing an outline of carbon adhesion state analysis by the carbon adhesion state analysis device according to the present embodiment.

In the present embodiment, the following two-step analysis is performed in order to determine the control standard (carbon control value) for controlling the carbon adhesion state of the furnace wall of the coke oven.
[First stage] Identification of carbon adhesion position (influenced area) that affects extrusion load [Second stage] Determination of appropriate range of carbon adhesion area in affected area

That is, in the first stage, a region (referred to as an “affected region”) that affects the extrusion load due to the carbon adhesion state in the entire furnace wall surface is specified. In the second step, an appropriate range of carbon adhesion area is determined for the influence area identified in the first step in order to reduce the extrusion load of coke. Specifically, the processing as shown in FIG. 1 is performed at each stage. The "carbon adhesion area" refers to the area of the portion of the furnace wall where carbon is adhered.

First, before carrying out the first-stage processing, image data of a plurality of furnace walls (hereinafter referred to as "furnace wall image") acquired in each of a plurality of kilns constituting one furnace group and the furnace wall image. It is assumed that the coke extrusion load at the time of acquisition is acquired as actual data (S10). Further, as a result of image analysis of each furnace wall image, it is assumed that the carbon adhesion position on the furnace wall is specified (S20). These treatments may be specified by using known techniques such as the techniques described in Patent Documents 1 and 2.

When the carbon adhesion position is specified from the furnace wall image, in the first stage, the furnace wall surface is first divided into small areas in order to identify the affected area of the entire furnace wall surface that affects the extrusion load due to the carbon adhesion state. (S30). Then, based on the carbon adhesion position specified in step S20, the presence or absence of a correlation between the carbon adhesion area in the small region and the magnitude of the extrusion load is evaluated (S40), and similar small regions having a correlation are integrated. By doing so, the affected area is specified (S50).

In the second step, the relationship between the carbon adhesion area and the coke extrusion load is evaluated for the affected area identified in step S50 (S60), and based on the result, the coke extrusion load is reduced for each affected area. The appropriate range of carbon adhesion area for is determined (S70).

By finding the correlation between the extrusion load of coke and the carbon adhesion area by analyzing the carbon adhesion state according to the present embodiment, the carbon management standard in the target furnace group can be determined. Hereinafter, the configuration of the carbon adhesion state analysis device according to the present embodiment and the carbon adhesion state analysis method based on the configuration will be described in detail.

<2. Configuration of carbon adhesion state analysis device>
Based on FIG. 2, the functional configuration of the carbon adhesion state analysis device 100 according to the present embodiment will be described. FIG. 2 is a block diagram showing a functional configuration of the carbon adhesion state analysis device 100 according to the present embodiment. As shown in FIG. 2, the carbon adhesion state analysis device 100 according to the present embodiment includes an analysis processing unit 110 and a determination processing unit 120. In the present embodiment, the carbon adhesion state analysis device 100 will be described as including the analysis processing unit 110 and the determination processing unit 120, but the present invention is not limited to this example. For example, the analysis processing unit 110 and the determination processing unit 120 may be configured as separate devices, and communication between the devices may be possible so that the determination processing unit 120 can acquire the analysis result by the analysis processing unit 110. good.

[2-1. Analysis processing unit]
The analysis processing unit 110 acquires the correlation between the carbon adhesion area and the coke extrusion load based on the actual data, and affects the extrusion load on the furnace wall, and further influences the influence region. The affected area is specified only in the affected area (hereinafter referred to as "limited influence area"). As shown in FIG. 2, the analysis processing unit 110 includes a carbon determination unit 111, a correlation analysis unit 113, an influence region specifying unit 115, and an appropriate range calculation unit 117.

The carbon determination unit 111 determines the carbon adhesion state at each position of the furnace wall image in the actual data obtained from the past operation results. The actual data is stored in the actual data storage unit 10 that stores the actual data of the past operation results, and the carbon determination unit 111 reads the actual data from the actual data storage unit 10 and describes the individual furnace wall images. Determine the carbon adhesion state at each position. The carbon determination unit 111 may specify the carbon adhesion state from the furnace wall image by using, for example, the techniques described in Patent Documents 1 and 2. In the present embodiment, using the technique described in Patent Document 2, the position where carbon is substantially uniformly adhered to the furnace wall surface so as to cover the refractory bricks constituting the furnace wall (in Patent Document 2, " The position called "uniform carbon adhesion state") is specified as the carbon adhesion position. The identification of the carbon adhesion position is the determination of the carbon adhesion state in the present embodiment. Further, the above-mentioned "carbon adhesion area" refers to the area of the portion of the furnace wall surface specified to be in the "uniform carbon adhesion state" by using the technique described in Patent Document 2.

The correlation analysis unit 113 divides the furnace wall surface into a plurality of small regions, and analyzes the correlation between the carbon adhesion area and the extrusion load for each small region. The correlation analysis unit 113 calculates the carbon adhesion area specified as the carbon adhesion position by the carbon determination unit 111 in the entire small region of each of the small regions set at the same position in the plurality of furnace wall images. Then, the correlation analysis unit 113 acquires the relationship between the carbon adhesion area and the coke extrusion load when the furnace wall image is acquired for each small region, and if there is a correlation or if there is a correlation. Calculate what kind of correlation there is. In the present embodiment, the case where the coke extrusion load increases when carbon adheres to the furnace wall is regarded as a positive correlation, and conversely, the case where the coke extrusion load decreases when carbon adheres to the furnace wall is regarded as a negative correlation. Identify the correlation for the region.

The affected region specifying unit 115 is identified as having a positive correlation in the entire furnace wall surface based on the correlation between the carbon adhesion area of each small region of the furnace wall surface obtained by the correlation analysis unit 113 and the coke extrusion load. Obtain the distribution of the small area and the small area identified as having a negative correlation, and identify the affected area that affects the coke extrusion load. In the present embodiment, the small region portion specified to have a positive correlation is specified as an influence region in which the extrusion load of coke increases when carbon adheres to the furnace wall, and the small region portion specified to have a negative correlation is specified. , It is identified as an area of influence where the extrusion load of coke is reduced when carbon adheres to the furnace wall. Further, the influence area specifying unit 115 may specify a limited influence area having a higher degree of influence in each of the specified influence areas. At this time, the affected region specifying unit 115 statistically evaluates the relationship between the carbon adhesion area and the coke extrusion load, and based on the result, identifies a statistically significant small region among the affected regions as a limited impact region. ..

The appropriate range calculation unit 117 evaluates the relationship between the carbon adhesion area and the coke extrusion load for each influence region or limited impact region specified by the influence region identification unit 115, and carbon for reducing the coke extrusion load. Determine the appropriate range of adhesion area. By setting the carbon adhesion area within the appropriate range determined by the appropriate range calculation unit 117, the coke extrusion load can be reduced.

[2-2. Judgment processing unit]
The determination processing unit 120 determines whether or not carbon is properly attached to the furnace wall surface based on the newly acquired furnace wall image. As shown in FIG. 2, the determination processing unit 120 includes a determination carbon determination unit 121, an affected region carbon adhesion area calculation unit 123, and an adhesion state determination unit 125.

The determination carbon determination unit 121 determines the carbon adhesion state at each position of the furnace wall image newly acquired as a determination target of the carbon adhesion state on the furnace wall surface of the coke oven. The determination carbon determination unit 121 functions in the same manner as the carbon determination unit 111 of the analysis processing unit 110, and the carbon adhesion state may be specified from the furnace wall image using a known analysis technique. In the present embodiment, the technique described in Patent Document 2 is used to specify a position where carbon is substantially uniformly adhered to the furnace wall surface so as to cover the refractory bricks constituting the furnace wall as a carbon adhesion position. do.

The affected area carbon adhesion area calculation unit 123 determines the carbon adhesion area in the portion specified as the affected area or the limited affected area by the affected area specifying unit 115 of the analysis processing unit 110 for the furnace wall image to be determined. Calculated based on the determination result of the determination unit 121.

The adhesion state determination unit 125 determines whether or not the carbon adhesion area of the influence region or the limited influence region calculated by the influence region carbon adhesion area calculation unit 123 is within an appropriate range. The adhesion state determination unit 125 compares the carbon adhesion area of the influence region or the limited influence region with the appropriate range of the carbon adhesion area in the influence region or the limited influence region calculated by the appropriate range calculation unit 117 of the analysis processing unit 110. , Determine whether the carbon adhesion to the furnace wall is normal. The adhesion state determination unit 125 determines that the carbon adhesion state of the furnace wall is normal if the carbon adhesion area is within the appropriate range, and adjusts the carbon adhesion state of the furnace wall if the carbon adhesion area is outside the appropriate range. Judge that it is necessary. If adjustment is required, countermeasure actions such as adjustment of the amount of air supplied by the lance are implemented. Further, the adhesion state determination unit 125 outputs the determination result of the carbon adhesion state to the furnace wall via an output device (not shown) such as a display device or a voice output device, and outputs the carbon of the current furnace wall to the operator. You may notify the adhesion state.

<3. Carbon adhesion state analysis method>
Hereinafter, the carbon adhesion state analysis method by the carbon adhesion state analysis device 100 according to the present embodiment will be described with reference to FIGS. 3 to 12.

[3-1. Analysis processing]
First, the analysis processing by the analysis processing unit 110 will be described with reference to FIGS. 4 to 11 based on the flowchart showing the analysis processing by the analysis processing unit 110 shown in FIG. In the analysis process by the analysis processing unit 110, the processes of the first stage and the second stage shown in FIG. 1 are executed.

(S100: Actual data input)
First, the actual data of the past operation results stored in the actual data storage unit 10 is input to the carbon determination unit 111 of the analysis processing unit 110 (S100). The actual data includes at least the furnace wall image and the extrusion load in the coke extrusion step performed immediately before (or may be immediately after) the acquisition of the furnace wall image. In this embodiment, analysis processing is performed for each furnace group, and the carbon control value of each furnace group is determined. If a sufficient number of actual data can be obtained for each kiln constituting the furnace group, the actual data may be analyzed for each kiln to determine the carbon control value of each kiln.

(S110: determination of carbon adhesion state)
The carbon determination unit 111 determines the carbon adhesion state at each position of the furnace wall image based on the input actual data (S110). The input of the actual data may be performed, for example, by the carbon determination unit 111 reading the actual data from the actual data storage unit 10. Here, in step S110, as described above, the carbon adhesion state may be specified from the furnace wall image by using a known technique. In the present embodiment, the technique described in Patent Document 2 is used to specify a position where carbon is substantially uniformly adhered to the furnace wall surface so as to cover the refractory bricks constituting the furnace wall as a carbon adhesion position. do. For example, as shown on the upper side of FIG. 4, in the furnace wall surface W of the furnace wall, the region C specified as the carbon adhesion position can be represented.

(S120: Correlation analysis, S130: Influence area identification)
Next, the correlation analysis unit 113 calculates the correlation between the carbon adhesion area and the coke extrusion load in each small region set by partitioning the furnace wall (S120). In order to identify the region of influence that affects the coke extrusion load in the next step S130, the correlation analysis unit 113 finely partitions the furnace wall surface, and the carbon adhesion area and coke extrusion are made for each small region formed by the compartment. Calculate the correlation with the load. The partition of the furnace wall surface may be divided into meshes, for example. With mesh division, it is possible to easily set a small area. For example, when the furnace wall surface W is divided into meshes of equal size M as shown in the lower side of FIG. 4, for each mesh M, the carbon adhesion position in the mesh region and the area of the specified carbon adhesion region C (that is, carbon adhesion) Area) is calculated. The number of mesh divisions may be appropriately set according to the furnace wall area and the resolution of analysis. For example, the number of mesh divisions may be set to 200. Further, in the present embodiment, the carbon adhesion area is calculated by calculating the carbon adhesion area of each mesh for each of the left and right furnace wall images acquired from one kiln, and then obtaining the average value of the left and right furnace walls. It shall be. The carbon adhesion area may be expressed by the number of pixels.

When the correlation analysis unit 113 calculates the carbon adhesion area for each mesh of the furnace wall image to be analyzed, it analyzes the correlation between the carbon adhesion area and the extrusion load for each mesh and identifies the affected region (S130). Correlation analysis can be performed using, for example, simple regression analysis or correlation analysis.

Here, the analysis of the correlation between the carbon adhesion area and the extrusion load performed for each mesh will be described with reference to FIG. In the example of FIG. 5, based on three actual data (data 1: extrusion load 20 tonf, furnace wall image G1, data 2: extrusion load 30 tonf, furnace wall image G2, data 3: extrusion load 40 tonf, furnace wall image G3). The correlation between the carbon adhesion area and the extrusion load is sought. In the furnace wall images G1 to G3 shown in FIG. 5, the carbon adhesion region C specified as the carbon adhesion position in step S110 and the mesh set for the furnace wall images G1 to G3 are shown. The correlation analysis unit 113 acquires the relationship between the carbon adhesion area and the coke extrusion load in the meshes at the same positions in the furnace wall images G1 to G3.

For example, looking at the meshes a1 to a3 at the same positions in the furnace wall images G1 to G3 in FIG. 5, the carbon adhesion areas contained in the mesh regions are different between the meshes a1, a2, and a3. Further, the extrusion load of coke when the furnace wall images G1 to G3 are acquired is also different. When these relationships are graphed, it can be seen that the coke extrusion load increases as the carbon adhesion area increases. In the present embodiment, when there is a relationship that the extrusion load of coke increases as the carbon adhesion area increases, it is said that there is a positive correlation between the carbon adhesion area and the coke extrusion load.

On the other hand, looking at the meshes b1 to b3 at the same positions in the furnace wall images G1 to G3 in FIG. 5, the carbon adhesion areas contained in the mesh regions are different for the meshes b1, b2, and b3. When these relationships are graphed, it can be seen that the coke extrusion load decreases as the carbon adhesion area increases. In the present embodiment, when there is a relationship that the extrusion load of coke decreases as the carbon adhesion area increases, it is said that there is a negative correlation between the carbon adhesion area and the coke extrusion load.

In this way, the correlation between the carbon adhesion area and the coke extrusion load differs depending on the position of the furnace wall. The correlation analysis unit 113 analyzes the correlation at each position (each mesh) of the furnace wall, and obtains the distribution of the obtained positive and negative correlation relationships to identify the position that has a large effect on the extrusion load. It becomes possible to do.

The correlation between the carbon adhesion area and the coke extrusion load can be performed, for example, by using simple regression analysis or correlation analysis as described above.

For example, in simple regression analysis, when simple regression y = ax + b is taken as the carbon adhesion area x and the coke extrusion load y, the correlation between the carbon adhesion area and the coke extrusion load can be expressed by the regression coefficient a. .. If the regression coefficient is a positive value, there is a positive correlation between the carbon adhesion area and the coke extrusion load, and if the regression coefficient is a negative value, there is a negative correlation between the carbon adhesion area and the coke extrusion load. The upper side of FIG. 6 shows an example of the regression coefficient values of each mesh set by dividing the furnace wall surface. In the example of the distribution of the correlation by the regression coefficient shown on the upper side of FIG. 6, when focusing only on the positive and negative of the coefficient regardless of the magnitude of the regression coefficient, the region of the positive coefficient that spreads mainly on the extruder side and mainly the guide. It can be divided into areas with negative coefficients that extend to the vehicle side. From this, it can be seen that there is a positive correlation on the extruder side and a negative correlation on the guide vehicle side. In addition, although there is a mesh in which the numerical values "0.00" and "-0.00" are described in the regression coefficient shown on the upper side of FIG. 6, the positive and negative are divided by the hidden value of 3 digits after the decimal point. ..

The positive coefficient region and the negative coefficient region of the regression coefficient are the regions of influence that affect the extrusion load according to the carbon adhesion state of the furnace wall. The method of affecting the extrusion load differs depending on the carbon adhesion state of each affected area. Specifically, it can be said that the negative correlation region is a region where carbon should be adhered, and the positive correlation region does not adhere carbon. It can be said that it is a better area.

Furthermore, by taking the variance ratio of the sum of squares of regression and the sum of squares of error, it is possible to identify the region with a higher degree of influence among the regions of influence identified by the simple regression analysis. For example, if the variance ratio of the sum of squares of regression and the sum of squares of error is obtained from the results of the simple regression analysis shown on the upper side of FIG. 6, the relationship shown on the lower side of FIG. 6 can be obtained. Here, when a mesh having a dispersion ratio of 2 or more is specified, there are limited influence regions having a high degree of influence on the extrusion load of coke in the height center portion on the guide vehicle side and the height center portion to the lower portion on the extruder side. You can see that. That is, since the limited influence region in the center of the height on the guide vehicle side has a negative correlation, it is highly possible that the extrusion load increases as the carbon adhesion area becomes smaller, so it is better to attach carbon to the furnace wall. It becomes an area. In addition, since the limited influence region from the center to the bottom of the height on the extruder side has a positive correlation, there is a high possibility that the extrusion load will increase as the carbon adhesion area increases, so it is better not to adhere carbon to the furnace wall. It will be a good area.

Further, when the correlation is acquired by using the correlation analysis, the influence region can be specified by, for example, obtaining the correlation coefficient and performing an uncorrelated test for evaluating whether or not the correlation coefficient is 0. The correlation coefficient r _xy is expressed by the ratio of the product of the standard deviation of the carbon adhesion area x and the standard deviation of the extrusion load y and the covariance, where the carbon adhesion area x and the coke extrusion load y are taken. The correlation coefficient r _xy takes a value from -1 to +1 and indicates that the larger the absolute value of the correlation coefficient, the stronger the relationship between the two data. Further, when the correlation coefficient r _xy is 0, it is determined that there is no correlation. The uncorrelated test tests the following hypothesis.

[Null hypothesis] The population correlation coefficient is 0 ⇔ [Alternative hypothesis] The population correlation coefficient is not 0

Assuming that the correlation coefficient of n data is r _xy , the statistic t ₀ used in the test is expressed by the following equation (1). When the statistic t ₀ is tested at a significance level of 5% according to the t distribution with n-2 degrees of freedom, the null hypothesis (that is, the population correlation coefficient is 0) under the condition of the following equation (2). Is rejected, that is, it is statistically judged that the correlation coefficient is not 0. Since it is a two-sided test, the t distribution is compared with a probability of 0.025. The statistic t ₀ is an index showing the certainty of the correlation coefficient obtained by the uncorrelated test, and among the correlations that can be determined based on the correlation coefficient by using the index, the statistically significant correlation is obtained. Area can be specified.

FIG. 7 shows an example of the relationship between the carbon adhesion area obtained by the correlation analysis and the coke extrusion load. The upper side of FIG. 7 shows the value of the correlation coefficient of each mesh. When this correlation coefficient is divided into regions according to the positive and negative values as in the case of the regression coefficient, as shown in the upper part of FIG. 7, the region of the positive coefficient that spreads mainly on the extruder side and the region of the positive coefficient mainly on the guide wheel side. It can be divided into a region of widening negative coefficients. That is, it can be seen that there is a positive correlation on the extruder side and a negative correlation on the guide wheel side, as in the case of the regression coefficient distribution shown on the upper side of FIG.

Next, when an uncorrelated test is performed based on the correlation coefficient on the upper side of FIG. 7, it is possible to identify a region having a higher degree of influence among the affected regions identified by the correlation analysis. The lower part of FIG. 7 shows the statistic shown in the above equation (1) obtained as a result of the uncorrelated test for each mesh. As shown in the lower part of FIG. 7, similar to the distribution of the dispersion ratio shown in the lower part of FIG. It can be seen that there is a high degree of limited influence area. On the other hand, the region specified as the limited influence region is smaller than that when the dispersion ratio shown in the lower part of FIG. 6 is used, and the result that the influence region is more limited can be obtained.

In this way, in step S130, it is possible to specify the management target location for appropriately managing the carbon adhesion amount of the furnace wall, and to grasp the influence of carbon adhesion on the extrusion load on the furnace wall from the positive and negative of the correlation. It will be possible. Further, by specifying the limited influence region, it is possible to specify the region affected by the extrusion load of coke. Although the calculation of the affected region and the limited impact region has been described with respect to step S130, it is not always necessary to specify the limited impact region. For example, the limited influence region may be specified when it is desired to specify a position where the carbon adhesion state significantly affects the extrusion load of coke.

(S140: Appropriate range calculation)
After that, the appropriate range calculation unit 117 analyzes the relationship between the carbon adhesion area of the specified affected area and the extrusion load of coke, and determines the appropriate range of the carbon adhesion area that can reduce the extrusion load (S140). That is, in step S140, in order to see the effect of carbon adhering to the statistically significant mesh region on the coke extrusion load, first, the positive obtained in step S130, for example, as shown in FIG. The carbon adhesion area of each region is calculated for the correlation influence region (limited influence region) PC and the negative correlation influence region (limited influence region) NC.

More specifically, as shown in FIG. 9, for example, the left and right furnace wall images LW and RW obtained by one-time coke oven wall diagnosis are affected by the positive correlation as shown in FIG. The total sum of the carbon adhesion areas in the limited influence region) PC and the negative correlation influence region (limited influence region) NC is calculated. As a result, the carbon adhesion area in the positive correlation influence region and the carbon adhesion area in the negative correlation influence region are acquired for the left furnace wall, and the carbon adhesion area and the negative correlation influence region in the positive correlation influence region for the right furnace wall. The carbon adhesion area and is obtained. Then, in the present embodiment, for the positive correlation influence region and the negative correlation influence region, the average value of the carbon adhesion areas of the left and right furnace walls is calculated to obtain the positive correlation influence region and the negative correlation influence region. The carbon adhesion area of is calculated.

When the appropriate range calculation unit 117 calculates the carbon adhesion area between the positive correlation influence region (limited influence region) and the negative correlation influence region (limited influence region), coke with at least one of these carbon adhesion areas. The relationship with the extrusion load is analyzed, and the appropriate carbon adhesion area (appropriate range) at which the extrusion load is below the control value is determined.

For example, when there is only a positive correlation limited influence region (that is, there is no negative correlation limited influence region), the total carbon adhesion area of the positive correlation limited influence region as shown in FIG. 10 (horizontal axis). ), The appropriate range of carbon adhesion area can be calculated from the representative value (vertical axis) of the extrusion load. The representative value of the extrusion load can be determined, for example, by using the probability distribution correlation analysis method described in Patent Document 3. The probability distribution correlation analysis method is a general-purpose method for analyzing the correlation between two data items (for example, the explanatory variable x and the objective variable y) from the operation data. Specifically, the range taken by the value of the explanatory variable x is divided into small regions, and the probability density function (for example, normal distribution or lognormal distribution) specified for each divided region is applied to obtain a predetermined cumulative probability. This is a method for obtaining a conditional probability density model p'(y | x) of the explanatory variable x and the objective variable y. This method can be applied even if the explanatory variables are two or more (see Patent Document 3).

In the present embodiment, the value of the extrusion load having the specified cumulative probability is determined as the representative value. In the example of FIG. 10, the value of the extrusion load having a cumulative probability of 80% is used as a representative value, and in this case, it means that the extrusion load generated in the future has a probability of 80% and is equal to or less than the representative value. Therefore, as shown in FIG. 10, when the range in which the control value of the extrusion load is 30 tonf or less is defined as the appropriate range of the representative value of the extrusion load, the sum of the carbon adhesion areas in the limited influence region of the positive correlation sets this appropriate range. If it is within the satisfying range, the extrusion load can be suppressed to the control value of 30 tonf or less with a probability of 80%.

Similarly, when both the positive correlation influence region and the negative correlation influence region exist, the sum of the carbon adhesion areas (horizontal axis) of the positive correlation limited influence region and the negative axis are similarly as shown in FIG. The representative value of the extrusion load corresponding to the total carbon adhesion area (vertical axis) in the limited effect region of the correlation is calculated from the data obtained by using the probability distribution correlation analysis method, etc., and the extrusion load is less than the control value. You just have to decide the appropriate range. Here, in FIG. 11, the representative value is the value of the extrusion load at which the cumulative probability is 80%, and the magnitude thereof is shown by shading. The dark area on the upper left side of the control value line is an appropriate range of 30 tonf or less, which is the control value. If the sum of the carbon adhesion areas in the positive correlation limited influence region and the total carbon adhesion area in the negative correlation limited influence region are within the range that satisfies this appropriate range, there is an 80% probability that the extrusion load will be the control value of 30 tonf. It can be suppressed to the following.

In addition to the above probability distribution correlation analysis method, the present invention may calculate a representative value of the coke extrusion load by using, for example, the quantile regression model shown in Non-Patent Document 1.

As described above, the carbon adhesion area was obtained for each of the positive correlation effect region where carbon should not be adhered and the negative correlation influence region where carbon should be adhered, and the relationship with the coke extrusion load was analyzed. By doing so, it is possible to determine an appropriate carbon adhesion area where the extrusion load is low.
[3-2. Determination process]
Next, the determination processing of the carbon adhesion area by the determination processing unit 120 will be described with reference to FIG. 12. Note that FIG. 12 is a flowchart showing the determination processing of the carbon adhesion area by the determination processing unit 120.

First, a furnace wall image for determining whether or not the carbon adhesion state of the furnace wall is appropriate is input to the determination carbon determination unit 121 of the determination processing unit 120 (S200). The input of the furnace wall image may be performed, for example, by the determination carbon determination unit 121 reading the furnace wall image from a predetermined storage device. When the furnace wall image is input, the determination carbon determination unit 121 determines the carbon adhesion state at each position of the input furnace wall image (S210). In step S210, the carbon adhesion state may be specified from the furnace wall image in the same manner as in step S110 of FIG. In the present embodiment, the technique described in Patent Document 2 is used to specify a position where carbon is substantially uniformly adhered to the furnace wall surface so as to cover the refractory bricks constituting the furnace wall as a carbon adhesion position. do.

Next, the affected region carbon adhesion area calculation unit 123 calculates the carbon adhesion area of the region specified as the affected region in the furnace wall image to be determined (S220). As the affected area, the affected area specified by the affected area specifying unit 115 of the analysis processing unit 110 in step S130 of FIG. 3 is used. The affected area carbon adhesion area calculation unit 123 calculates the carbon adhesion area in the positive correlation influence region and the carbon adhesion area in the negative correlation influence region.

Then, the adhesion state determination unit 125 determines whether or not the carbon adhesion area of the influence region calculated by the influence region carbon adhesion area calculation unit 123 is within an appropriate range (S230). The adhesion state determination unit 125 compares the carbon adhesion area in the affected area with the appropriate range of the carbon adhesion area in the influence area calculated by the appropriate range calculation unit 117 of the analysis processing unit 110 in step S140 of FIG. Determine whether the carbon adhesion to the wall is normal. At this time, the adhesion state determination unit 125 outputs the determination result of the carbon adhesion state to the furnace wall via an output device (not shown) such as a display device or a voice output device, and outputs the determination result of the carbon adhesion state to the operator to the operator of the current furnace wall. The carbon adhesion state may be notified.

If the carbon adhesion area is within an appropriate range, the adhesion state determination unit 125 determines that the carbon adhesion state of the furnace wall is normal, and ends the determination process of FIG. 12 (S240). On the other hand, if the carbon adhesion area is out of the appropriate range, the adhesion state determination unit 125 determines that it is necessary to adjust the carbon adhesion state of the furnace wall, and takes measures so that the carbon adhesion state becomes appropriate (). S250). Specifically, by adjusting the amount of air supplied from the lance that is inserted from multiple inlets at the top of the kiln and burns and removes carbon by blowing air against the carbon adhering to the furnace wall. The carbon adhesion area can be adjusted. The relationship between the amount of carbon adhering and the amount of air can be obtained from operational experience.

For example, if the carbon adhesion area is less than the appropriate range in the limited influence region at the center of the height on the guide vehicle side, the extrusion load is likely to increase. Therefore, in order to grow the carbon adhering to the furnace wall at such a position, the air supply is reduced or stopped for the lance that supplies air to the vicinity of the affected area. Further, when the carbon adhesion area is larger than the appropriate range in the limited influence region from the central portion to the lower portion of the height on the extruder side, the extrusion load is likely to increase. Therefore, in order to remove the carbon adhering to the furnace wall at such a position, the lance that supplies air to the vicinity of the affected area is increased more than usual by, for example, setting the maximum amount of air to be supplied. After the countermeasure is completed, the determination process shown in FIG. 12 is terminated.

<4. Summary>
The configuration of the carbon adhesion state analysis device 100 according to the present embodiment and the carbon adhesion state analysis method based on the configuration have been described above. According to the present embodiment, the region of influence that affects the extrusion load of coke on the wall surface of the furnace is specified, and the appropriate range of the carbon adhesion area in the region of influence is determined. By adjusting the amount of carbon adhesion so that the carbon adhesion area is within the determined appropriate range, the coke extrusion load can be reduced.

By using the method according to the present embodiment, even if the knowledge about the carbon adhesion position of the furnace wall and the relationship between the carbon adhesion state and the coke extrusion load is not obtained in advance, the region of influence affecting the extrusion load is specified. can do. In addition, the carbon adhesion area at any place on the furnace wall can be managed independently. Furthermore, by determining the appropriate range of the carbon adhesion area in the affected area, it is possible to control whether the carbon adhesion amount in the affected area is insufficient or excessive and is out of the appropriate range, and the extrusion load can be extruded. Can be taken before the increase.

In the above embodiment, the furnace wall surface is mesh-divided as in step S120 of FIG. 3, and the correlation between the carbon adhesion area and the coke extrusion load is calculated for each mesh, but the furnace wall surface is calculated by a method other than mesh division. May be divided into small areas. For example, as shown in FIG. 13, a plurality of analysis target positions (for example, analysis target positions P1, P2, P3) are set on the furnace wall surface of the furnace wall W, and each has a preset radius centered on each analysis target position. A defined circle (eg, circles C1, C2, C3) may be set as a small area. Even when the small region is set in this way, the carbon adhesion area contained in each circle can be calculated and the correlation between the carbon adhesion area and the coke extrusion load can be obtained as in the above embodiment. Further, the small area set on the wall surface of the furnace may have overlapping portions as shown in FIG.

<5. Hardware configuration example>
Hereinafter, the hardware configuration of the carbon adhesion state analysis device 100 according to the present embodiment will be described in detail with reference to FIG. 14. FIG. 14 is a block diagram showing a hardware configuration example of the carbon adhesion state analysis device 100 according to the embodiment of the present invention.

The carbon adhesion state analysis device 100 mainly includes a CPU 901, a ROM 903, and a RAM 905. Further, the carbon adhesion state analysis device 100 further includes a bus 907, an input device 909, an output device 911, a storage device 913, a drive 915, a connection port 917, and a communication device 919.

The CPU 901 functions as an arithmetic processing device and a control device, and controls all or a part of the operation in the carbon adhesion state analysis device 100 according to various programs recorded in the ROM 903, the RAM 905, the storage device 913, or the removable recording medium 921. do. The ROM 903 stores programs, calculation parameters, and the like used by the CPU 901. The RAM 905 primaryly stores a program used by the CPU 901, parameters that are appropriately changed in the execution of the program, and the like. These are connected to each other by a bus 907 composed of an internal bus such as a CPU bus.

The bus 907 is connected to an external bus such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge.

The input device 909 is an operating means operated by the user, such as a mouse, a keyboard, a touch panel, buttons, switches, and levers. Further, the input device 909 may be, for example, a remote control means (so-called remote controller) using infrared rays or other radio waves, or an external connection device 923 such as a PDA corresponding to the operation of the carbon adhesion state analysis device 100. May be. Further, the input device 909 is composed of, for example, an input control circuit that generates an input signal based on the information input by the user using the above-mentioned operating means and outputs the input signal to the CPU 901. By operating the input device 909, the user of the carbon adhesion state analysis device 100 can input various data to the carbon adhesion state analysis device 100 and instruct the processing operation.

The output device 911 is composed of a device capable of visually or audibly notifying the user of the acquired information. Such devices include display devices such as CRT display devices, liquid crystal display devices, plasma display devices, EL display devices and lamps, audio output devices such as speakers and headphones, printer devices, mobile phones, facsimiles and the like. The output device 911 outputs, for example, the results obtained by various processes performed by the carbon adhesion state analysis device 100. Specifically, the display device displays the results obtained by various processes performed by the carbon adhesion state analysis device 100 as text or an image. On the other hand, the audio output device converts an audio signal composed of reproduced audio data, acoustic data, and the like into an analog signal and outputs the signal.

The storage device 913 is a data storage device configured as an example of the storage unit of the carbon adhesion state analysis device 100. The storage device 913 is composed of, for example, a magnetic storage device such as an HDD (Hard Disk Drive), a semiconductor storage device, an optical storage device, an optical magnetic storage device, or the like. The storage device 913 stores programs executed by the CPU 901, various data, various data acquired from the outside, and the like.

The drive 915 is a reader / writer for a recording medium, and is built in or externally attached to the carbon adhesion state analysis device 100. The drive 915 reads the information recorded on the removable recording medium 921 such as a mounted magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, and outputs the information to the RAM 905. The drive 915 can also write a record to a removable recording medium 921 such as a mounted magnetic disk, optical disk, magneto-optical disk, or semiconductor memory. The removable recording medium 921 is, for example, a CD media, a DVD media, a Blu-ray (registered trademark) media, or the like. Further, the removable recording medium 921 may be a compact flash (registered trademark) (CompactFlash: CF), a flash memory, an SD memory card (Secure Digital memory card), or the like. Further, the removable recording medium 921 may be, for example, an IC card (Integrated Circuit card) or an electronic device equipped with a non-contact type IC chip.

The connection port 917 is a port for directly connecting the device to the carbon adhesion state analysis device 100. As an example of the connection port 917, there are a USB (Universal Serial Bus) port, an IEEE1394 port, a SCSI (Small Computer System Interface) port, an RS-232C port, and the like. By connecting the external connection device 923 to the connection port 917, the carbon adhesion state analysis device 100 directly acquires various data from the external connection device 923 and provides various data to the external connection device 923. ..

The communication device 919 is, for example, a communication interface composed of a communication device or the like for connecting to the communication network 925. The communication device 919 is, for example, a communication card for a wired or wireless LAN (Local Area Network), Bluetooth (registered trademark), WUSB (Wireless USB), or the like. Further, the communication device 919 may be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), a modem for various communications, or the like. The communication device 919 can transmit and receive signals and the like to and from the Internet and other communication devices in accordance with a predetermined protocol such as TCP / IP. Further, the communication network 925 connected to the communication device 919 is configured by a network connected by wire or wirelessly, and may be, for example, the Internet, a home LAN, infrared communication, radio wave communication, satellite communication, or the like. ..

The above is an example of a hardware configuration capable of realizing the function of the carbon adhesion state analysis device 100 according to the embodiment of the present invention. Each of the above-mentioned components may be configured by using general-purpose members, or may be configured by hardware specialized for the function of each component. Therefore, it is possible to appropriately change the hardware configuration to be used according to the technical level at the time of implementing the present embodiment.

Although the preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, the present invention is not limited to these examples. It is clear that a person having ordinary knowledge in the field of technology to which the present invention belongs can come up with various modifications or modifications within the scope of the technical ideas described in the claims. , These are also naturally understood to belong to the technical scope of the present invention.

10 Actual data storage unit 100 Carbon adhesion state analysis device 110 Analysis processing unit 111 Carbon judgment unit 113 Correlation analysis unit 115 Impact area identification unit 117 Appropriate range calculation unit 120 Judgment processing unit 121 Judgment carbon judgment unit 123 Impact area Carbon adhesion area Calculation unit 125 Adhesion status determination unit

Claims (12)

A carbon determination unit that analyzes multiple furnace wall images acquired as actual data in the past operation of the coke oven and determines the carbon adhesion state in each furnace wall image.
Each of the furnace wall images is divided into a plurality of small areas, and based on the determination result of the carbon adhesion state by the carbon determination unit , the small areas are set at the same position in each of the furnace wall images. Correlation analysis unit that analyzes the correlation between the carbon adhesion area and the extrusion load when the furnace wall image including the small area is acquired .
Based on the correlation between the carbon adhesion area of each of the small regions and the coke extrusion load obtained by the correlation analysis unit, an influence region identification unit that specifies an influence region that affects the coke extrusion load, and an influence region identification unit.
A carbon adhesion state analysis device. The affected region specifying portion is a small region identified as having a positive correlation on the entire furnace wall surface based on the correlation between the carbon adhesion area of each small region obtained by the correlation analysis unit and the extruded load of coke. 1. Carbon adhesion state analyzer. The carbon adhesion state according to claim 1 or 2, wherein the correlation analysis unit calculates the carbon adhesion area for each of the small regions and calculates the correlation with the extrusion load by using simple regression analysis or correlation analysis. Analyst. The affected region specifying portion is a region among the identified affected regions that has a greater effect on the coke extrusion load based on an index indicating the certainty of the correlation coefficient obtained by the dispersion ratio or the uncorrelated test. The carbon adhesion state analysis apparatus according to any one of claims 1 to 3, further limiting and specifying an affected area. The carbon adhesion area between the positively correlated area of influence and the negatively correlated area of influence identified by the affected area identification portion is calculated, and coke with at least one of the carbon adhesion areas of each of the affected areas. The carbon adhesion according to any one of claims 1 to 4, further comprising an appropriate range calculation unit for analyzing the relationship with the extrusion load and calculating the appropriate range of the carbon adhesion area where the extrusion load is equal to or less than the control value. State analyzer. The carbon adhesion state analysis device according to claim 5, wherein the appropriate range calculation unit calculates the relationship between the carbon adhesion area in the affected region and the coke extrusion load by using probability distribution correlation analysis or quantile regression. .. For the furnace wall image for determining whether the carbon adhesion state is appropriate, the carbon adhesion area in the influence region specified by the influence region identification portion is calculated.
5. The claim 5 or the present invention further comprises a determination processing unit for comparing the calculated carbon adhesion area with the appropriate range calculated by the appropriate range calculation unit to determine whether or not the carbon adhesion state is appropriate. 6. The carbon adhesion state analysis device according to 6. When the calculated carbon adhesion area is out of the appropriate range, the determination processing unit determines the carbon adhesion to the furnace wall according to the correlation between the carbon adhesion area in the affected area and the coke extrusion load. The carbon adhesion state analysis device according to claim 7, wherein the amount of air in the lance for adjusting the amount is adjusted. The carbon adhesion state analysis device according to any one of claims 1 to 8, wherein the correlation analysis unit divides the furnace wall image into a mesh and sets the small area. A carbon determination step for determining the carbon adhesion state in each furnace wall image by analyzing multiple furnace wall images acquired as actual data in the past operation of the coke oven, and
Each of the furnace wall images is divided into a plurality of small areas, and based on the determination result of the carbon adhesion state in the carbon determination step, the small areas set at the same position of each furnace wall image are said to be small. A correlation analysis step for analyzing the correlation between the carbon adhesion area of the region and the extrusion load when the furnace wall image including the small region is acquired ,
An influence region identification step for specifying an influence region that affects the coke extrusion load based on the correlation between the carbon adhesion area of each of the small regions and the coke extrusion load.
Carbon adhesion state analysis method including. Computer,
A carbon determination unit that analyzes multiple furnace wall images acquired as actual data in the past operation of the coke oven and determines the carbon adhesion state in each furnace wall image.
Each of the furnace wall images is divided into a plurality of small areas, and based on the determination result of the carbon adhesion state by the carbon determination unit , the small areas are set at the same position in each of the furnace wall images. Correlation analysis unit that analyzes the correlation between the carbon adhesion area and the extrusion load when the furnace wall image including the small area is acquired .
Based on the correlation between the carbon adhesion area of each of the small regions and the coke extrusion load obtained by the correlation analysis unit, an influence region identification unit that specifies an influence region that affects the coke extrusion load, and an influence region identification unit.
A computer program that functions as a carbon adhesion state analysis device. On the computer
A carbon determination unit that analyzes multiple furnace wall images acquired as actual data in the past operation of the coke oven and determines the carbon adhesion state in each furnace wall image.
Each of the furnace wall images is divided into a plurality of small areas, and based on the determination result of the carbon adhesion state by the carbon determination unit , the small areas are set at the same position in each of the furnace wall images. Correlation analysis unit that analyzes the correlation between the carbon adhesion area and the extrusion load when the furnace wall image including the small area is acquired .
Based on the correlation between the carbon adhesion area of each of the small regions and the coke extrusion load obtained by the correlation analysis unit, an influence region identification unit that specifies an influence region that affects the coke extrusion load, and an influence region identification unit.
A computer-readable recording medium on which a program for functioning as a carbon adhesion state analyzer is recorded.

Images