JPH11256166A - Diagnosis of coke oven body - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for diagnosing a coke oven body capable of constructing database for quantifying damage to carbonizing chamber oven wall of coke oven, quantifying diagnosis and carrying out repair having high objectivity and data for control. SOLUTION: In this method, damage area is extracted from wall image in a carbonizing chamber and damage part data comprising the area image, damage characteristic amount value and damage name are calculated every damage area and the damage part data are successively accumulated in a storage device together with a carbonizing chamber No. and date and time. The accumulated damage part data are searched by using a carbonizing chamber No., a damage name, damage characteristic amount, or the like., as indices. Deterioration extent of the coke oven is quantified as characteristic amount deteriorating index and each of damage part data is compared with characteristic amount-deteriorating amount and evaluated, and ranking data to be repaired of damage part data (in the carbonizing chamber) are calculated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates generally to a method for extending the life of a furnace body of a coke oven, and more particularly to a method for collecting, managing, and analyzing enormous information on the furnace body of a coke oven, and generating an appropriate repair guideline. .

[0002]

2. Description of the Related Art In a coke oven, a large number of carbonization chambers and combustion chambers are connected alternately, coal is carried into the carbonization chamber, and high heat of 900 ° C. to 1100 ° C. is generally supplied from the combustion chamber through a furnace wall. The coal is added continuously for 20 hours and the coal is carbonized. That is, coke is produced. At the end of the carbonization, the coke is discharged and the coal is charged and heating is started again. This is a repetition, and the temperature is always high.

FIGS. 4 (a) and 4 (b) show visible shapes of the carbonization chamber when viewed from the kiln entrance (IN) to the kiln exit (EX). The size of the carbonization chamber is an example, and is approximately the height (H).
5m, width (W) 0.4 / 0.46m (tapered,
0.4 m at the coke extruder side (IN side), 0.46 m at the coke kiln exit side (EX side) and length (L) 16 m
Thus, a very narrow and deep furnace space is formed. Each of the refractory bricks constituting the furnace wall has a height of approximately 12 mm.
0 mm, width 260 mm, thickness 110 mm.

The refractory brick used for the inner wall of the coking chamber is exposed to the above-mentioned high heat for a long time, and when the coke is carried out after completion of the coke, it is subjected to coke pressure by a coke extruder to be thermally, chemically or chemically. It is easily damaged by mechanical stress. In other words, the joints on the furnace surface, brick cracks, carbon adhesion, or irregularities on the walls are likely to add to the damage, and the damage is likely to spread further. Affects operations. Further, if the damage is enlarged, gas leakage or black smoke is generated, which has an adverse effect on the environment. The life of the coke oven is defined as the time when the coke is unable to be discharged from the oven finally due to further damage, which is generally said to be 30 to 35 years.

[0005] Since the coke oven is a continuous operation facility, the operation cannot be stopped and cooled after construction operation. Further, it is extremely difficult to renew the internal bricks because of the structure in which the bricks are precisely stacked. For this reason, repairs such as spraying the damaged portion inside the carbonization chamber are currently being performed. Also, the coke oven of a general steel mill is 30
There are about 0 to 400 kilns. Conventionally, there has been no appropriate means for damage evaluation, and therefore, the determination of a kiln to be repaired has been made by visually determining a particularly deteriorated carbonization chamber. In recent years, a method has been proposed in which a two-dimensional camera is inserted to photograph a wall surface of a carbonized room and a state of wall surface damage is determined based on obtained image information in order to discover damage of the coke oven and to grasp its position.

For example, as a technique using image analysis, Japanese Patent Application Laid-Open No. 3-105195 discloses a camera transport lance in which a video camera is simply mounted from a furnace opening of a coke oven as shown in FIG. A method has been reported in which a damage state of each kiln is output as a diagnostic map by an image analyzer from an image obtained by photographing an inner wall surface of the kiln while moving the kiln inside the kiln and moving in the depth direction of the kiln.

[0007]

As described above, there are many coke ovens in a coke oven, and when including these diagnostic information and repair information, the amount of information becomes enormous, leading to insufficient information management. There was a problem. For example, image data of the entire inner wall inside the carbonization chamber has a spatial resolution of 1 mm and a contrast of 255 in order to accurately observe the damage state.
When photographing with a gradation image (1 byte), 10
The capacity is as large as 4 MB (6.5 × 1000 × 16 × 1000). That is, both walls obtained in one shot are 208
MB, and as mentioned above, 300-40
Since there are 0 carbonization chambers, even if each carbonization chamber is photographed only once, the image data amount is 60 GB to 83.2 G.
B and a huge thing. Therefore, in order to manage these huge amount of image data and compare the damage status in a timely manner,
Practical problems have arisen in handling, such as an increase in the size of the image data storage device and the time required to read and write data.

Further, even in the case where diagnosis is performed from the output result of the carbonization indoor wall diagnosis map as in the prior art, in order to repair the damage, the diagnosis map is manually manipulated and the damage situation is checked by looking at the image of the corresponding portion of the damage. Because it is necessary to confirm,
It is necessary to handle a huge amount of image data, and the same problem as described above occurs.

The conventional technique relates to the observation of the inner wall of an individual coking chamber, and performs a quantitative evaluation of a specific coking chamber. There was a problem that the priority of the carbonization room to be repaired, that is, the priority of the carbonization room to be repaired was unclear, and the repair guidance and repair guideline were not determined. In order to extend the life optimally, it is necessary to effectively utilize the captured inner wall images and accurately evaluate the degree of deterioration in order to perform effective repair.

[0010] In addition, in the repair judgment performed by evaluating the damage state from the photographed inner wall image, individual differences are likely to occur,
There is also a problem that the carbonization room to be repaired is erroneously determined, and repair is performed after the deterioration becomes particularly severe.

The problem to be solved by the present invention is to facilitate the analysis of the degree of damage for each furnace group without managing the image data of the entire furnace wall taken in order to carry out the optimal life extension measures of the coke oven, An object of the present invention is to provide a method for generating effective repair and management information based on a simplified diagnosis result.

[0012]

In order to achieve the above object, a method for diagnosing a coke oven furnace body according to the present invention comprises inserting a device for photographing the entire inner wall of a coke oven from a furnace opening of a coke oven carbonization chamber. A method of diagnosing the carbonized room from the captured carbonized indoor wall image, wherein a plurality of damaged regions are extracted from the carbonized indoor wall image obtained for each inner wall photographing in the carbonized room, and each of the extracted damaged regions is extracted. A damaged site data collection process of calculating and collecting a damaged site data including a partial image and a plurality of feature amounts and a damage name for each inner wall damaged region, and a process of capturing each carbonized interior wall image collected in the damaged site data collection process. Damaged site data is sequentially stored in a storage device together with a coking chamber number and a photographing time, and a coking chamber furnace wall damage database creating process for creating a coking chamber furnace wall damage database; The degree of deterioration of the coke oven is quantified in advance as a characteristic degradation index for each feature value corresponding to the damage name of the damaged part calculated from the furnace wall damage data obtained during the database creation process. By comparing and evaluating a plurality of damaged site data collected every time with the feature quantity deterioration index, the carbonization chambers to be repaired are ranked.

According to this, a plurality of damaged areas are extracted from the carbonized indoor wall image taken in the damaged part data collecting process, and the position and form are quantitatively indicated for each extracted inner wall damaged area. The feature amount of a plurality of items is calculated, the name of the damaged part is separated from the value of the feature amount, and cut out as a partial image with the maximum circumscribed rectangle size of the damaged part,
It is calculated and collected as damage site data.

The collected damaged part data represents the position and form of the damage, and is composed of the minimum amount of images when an operator or the like checks the images. It is possible to analyze the state of deterioration of the furnace body using detailed information without holding an image of the entire area.

Further, in the coking chamber furnace wall damage data obtained by creating the coking chamber furnace wall damage database, a set of the plurality of damaged site data of all the furnace gangs is associated with the coking chamber number and the photographing time. It is possible to easily search the characteristic amount item and the damage name of the data to find out what kind of damage is distributed in which carbonization chamber in the furnace group. For example, when it is desired to display a carbonized room number having a joint breakage having a length of 3 m or more, it is possible to easily display a carbonized room number that is directly related to a feature amount item or a damage name of damaged site data. . That is, on a furnace group basis,
It becomes possible to manage damage and analyze the degree of deterioration.

Further, in the present invention, the degree of deterioration exerted on the coke oven is quantified in advance as a characteristic amount deterioration index for each characteristic amount of the damaged part calculated from the carbonization chamber furnace wall damage data. The calculated feature amount,
By comparing and analyzing the characteristic amount deterioration index, the degree of damage can be obtained for each damaged part in the furnace group.
That is, the order of the degree of deterioration of the damaged portion to be repaired can be determined, and the order of the carbonized room to be repaired can be determined from the related carbonized room number of the damaged portion.
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION

[0018]

FIG. 1 is a block diagram showing an example in which the present invention is implemented by a computer system. A photographing device (not shown) is inserted into the coke oven 1 for photographing the wall surface of the coke oven 1,
The image taken by the imaging device, that is, the image of the carbonized interior wall is stored in real time or temporarily in a storage medium such as a magneto-optical disk or a hard disk.
Given to. FIG. 2 shows an example of the carbonized indoor wall image taken at this time. 2 includes images of damage such as joint breaks 10, carbon adhesion 11, spalling 12, and brick stripping 13. This image is an example shown schematically.

Various image processing is performed on the carbonized interior wall image (original image data) transmitted to the diagnostic analysis computer 2 or input via a storage medium such as a magneto-optical disk or a hard disk. A plurality of damaged regions are extracted over the entire wall image, and respective values of the feature quantity items as shown in Table 1 are calculated for each of the extracted damaged regions, which accurately represent the position and form. The image data of the damaged part image shown in FIG. 3 is extracted from the inner wall image based on the position and size of the circumscribed rectangle of the damaged part, and the damage name is sent from the value of the above-described feature quantity item. Is the damage site data.

[0020]

[Table 1]

In the method of determining the name of a damaged part, that is, in the assignment of name data to the value of a feature item, a range defined by both values using the lower limit value and the upper limit value of the feature value corresponding to each damage name. If there is a feature value in the table, the name is assigned (classification determination method). Note that other methods such as an AI method may be used. The damaged part image data extracted at this stage is assumed to have been subjected to original image data (no compression processing), compression encoding processing or reduction processing in accordance with the image data processing mode specified by the operator. .

Next, the damage site data (each value of the characteristic amount item, the damaged part image data and the damage name data) collected as described above is applied to the entire carbonized room wall image. Number and shooting time (in this example, date and time)
In addition to the above, the data is sequentially stored in the database storage device 3 shown in FIG. When the accumulation is completed, the carbonized indoor wall image (original image data) used for the search for the damaged portion is discarded. Therefore, the storage device 3
The damaged part data in the carbonization chamber furnace wall damage database describes the position and form of the damaged part in the form of a numerical value as each value of the feature quantity item. When it is desired to observe the damaged part image data, when it is compressed or reduced, it is subjected to restoration decoding processing or enlargement processing and given to a display for reference with sufficient accuracy. It is possible to hold sufficient information for diagnosis and analysis. That is,
The data of damage to the furnace wall of the coking chamber depends on the size of the damaged area of the inner wall of the coking chamber, but the amount of data is much smaller than that of the entire image of the inner wall of the coking chamber. A database is created with the amount.

When analyzing damage for each furnace group, the diagnostic analysis computer 2 reads the coking chamber furnace wall data from the storage device (storage device) 3 in which the coking chamber furnace wall damage database is stored, and performs each damage. Analyze the site data. The content is compared with the coking chamber furnace wall damage data stored in the storage device 3 of all coking chambers in the coking chamber furnace wall damage data by the value of the characteristic quantity item of the damaged part data or the damage name, etc. To explain an example of examining the distribution of damage in the case, for example, when analyzing the distribution of a carbonized chamber having joint breakage with a length of 3 m or more,
First, all the damaged part data are checked for the circumscribed rectangle length or the damaged part main axis length, which are the characteristic amounts in Table 1, and among them, the damaged part data whose characteristic value is 3 m or more is extracted. I do. Here, a set of damaged part data extracted assuming that two or more are extracted is referred to as a damaged part data group.

Among the extracted damaged part data groups, those whose damage names are broken are narrowed down to analysis targets. Next, the carbonization chamber number attached to the damaged site data narrowed down to the analysis target is extracted, and all the carbonization chambers in the furnace group where the target damage (here, joint breakage) exists are displayed in a list.
Here, when the carbonization chamber number is specified and the reading of the damaged position, form, damaged image, and the like is instructed, the diagnostic analysis computer 2 displays it.

As described above, it becomes possible to search for a damaged carbonization chamber, a damaged part, a damaged image, and the like on a per furnace group basis and under arbitrary conditions. Therefore, other diagnostic data such as the carbonization time and the extrusion time can be obtained. By analyzing the correlation between the actual data such as the applied force and the damaged site data, it becomes possible to analyze the effective degree of deterioration in each furnace group.

Next, the ranking of the carbonized chambers to be repaired based on the carbonized chamber damage database will be described. In the diagnostic analysis computer 2, a feature quantity deterioration index corresponding to the degree of deterioration of the carbonization chamber is set in advance for each damage name. The feature quantity deterioration index is obtained by setting a threshold value for each damage name for the feature quantity item of the damaged part data in Table 1. It is desirable to set the characteristic quantity deterioration index in consideration of the operating conditions and the blending of the raw coal charged to the coke oven by brand. Table 2
Is an example of a characteristic quantity deterioration index for carbon adhesion and joint breakage damage.

[0027]

[Table 2]

The diagnostic analysis computer 2 compares and analyzes each of the damaged part data in the storage device 3 in which the carbonization chamber furnace wall damage database is stored with the above-mentioned characteristic quantity deterioration index, thereby determining the degree of deterioration of the damaged part. judge. Specifically, in the data of all the damaged parts in the furnace group read from the storage device 3, the damaged part data with the damage name of carbon adhesion is the characteristic quantity deterioration index of carbon adhesion, and similarly, the damage name is the joint. The damaged part data of the cut is compared with the characteristic deterioration index of the joint break and the characteristic deterioration index of the corresponding damage name for the other damage names to evaluate the degree of deterioration of each damaged part. In other words, in the damaged part data with the damage name of carbon adhesion, the values of all the feature quantity items of the damaged part data are sequentially compared with the corresponding feature quantity deterioration index threshold, and if the values fall within the range defined by the threshold, 0 as the judgment value
Is exceeded, the difference is calculated as a judgment value. Then, the number of non-zero judgment values is counted, and the damaged parts are first ordered in the order of the number. If the judgment values are the same, the damaged parts are ordered in descending order of the total value of the judgment values. Thereby, the degree of deterioration of the damaged part data is quantitatively ranked. Therefore, since the damaged site data of the coking chamber in the furnace group was ordered in descending order of the degree of deterioration, the coking chamber number related to the damaged site data was immediately identified, and from which coking chamber the repair should be performed, It is objectively required without individual differences. That is, it is possible to order the carbonization chambers having a remarkable degree of deterioration.

[0029]

According to the present invention, it is not necessary to handle a huge image of the entire wall of the coke oven chamber, so that the diagnostic data management of the coke oven becomes easy, and the analysis of the degree of damage per furnace group becomes easy. Effective repair and furnace body management based on the objective diagnosis results can be realized. In other words, conventionally, qualitative judgment was followed, but according to the present invention, an appropriate repair kiln can be grasped and an optimal repair guideline can be determined based on the diagnosis result, which can contribute to extending the life of the coke oven.

[Brief description of the drawings]

FIG. 1 is a block diagram of a computer system according to an embodiment of the present invention.

FIG. 2 is a plan view schematically showing a captured carbonized indoor wall image.

FIG. 3 is a plan view schematically showing an example of a partial image of a damaged site collected in the present invention.

FIG. 4 (a) is a perspective view of the inner space of one coking chamber of a coke oven viewed from the outside of the kiln mouth to the kiln outlet through the kiln mouth, and FIG. It is a top view which shows the aspect which inserts a two-dimensional camera and image | photographs an inner wall surface.

[Explanation of symbols]

1: coke oven 2: diagnostic analysis computer 3: database storage device 4: diagnostic analysis computer terminal 10: joint breakage damage 11: carbon adhesion damage 12: spalling damage 13: brick exfoliation damage 20: damage area maximum outer angle rectangle 40 : 2D camera EX: kiln exit H: carbonization chamber height IN: kiln entrance L: carbonization chamber length W: carbonization chamber width W1: right vertical wall surface W2: left vertical wall surface

Claims (1)

[Claims] 1. A method for diagnosing said coking chamber from an image of the coking chamber inner wall, wherein a device for photographing the entire inner wall of the coke oven is inserted from a kiln opening of the coking furnace coking chamber. A plurality of damaged regions are extracted from the carbonized indoor wall image obtained for each inner wall photographing, and damaged portion data including a partial image, a plurality of feature amounts, and a damage name is calculated for each of the extracted inner wall damaged regions, and A damage site data collection process to be collected, and sequentially store in the storage device the damage site data for each carbonized indoor wall image taken in the damaged site data collection process together with the carbonized room number and the imaging time. A process for creating a coking chamber furnace wall damage database for creating a wall damage database, and a process for generating a damage name of a damaged part calculated from the coking chamber furnace wall damage data obtained in the coking chamber furnace wall damage database creation process. For each feature to be performed, the degree of deterioration of the coke oven is quantified in advance as a feature degradation index, and a plurality of damaged site data collected for each coking chamber of the coke oven are compared with the feature degradation index and evaluated. Thus, a coke oven furnace body diagnosis method is characterized in that carbonization chambers to be repaired are ranked.