JPH0627577B2 - Combustion state diagnostic device and combustion state control device - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to a combustion state diagnosis device for monitoring a flame in a combustion furnace and diagnosing a combustion state from the flame, and a combustion state control device for controlling the combustion state. .

[Background of the Invention]

Substance contained in the combustion exhaust gas in a boiler operation, the nitrogen oxides (hereinafter referred to as NO _x) regulation value is such is provided which is particularly harmful substances, protect the product amount below the regulated value To drive.

On the other hand, it is desirable to keep the combustion efficiency of the boiler at the maximum value at all times. By the way, the unburned component contained in the exhaust gas is a standard for calculating the combustion efficiency. The combustion efficiency decreases as the unburnt content in the exhaust gas increases, and the fuel consumption increases even if the same output is obtained. However, conventionally, it takes a long time to detect the unburned component, so that the combustion efficiency during operation has to rely on experience and intuition.

Recently, coal has been increasingly used instead of gas and oil as fuel, and is being converted into pulverized coal, CWM (coal / water slurry), COM (coal / oil slurry), etc. in boilers as well.

In particular, when coal is used as a fuel, the nitrogen content contained in the coal itself is converted into NO _x by combustion, so that the amount produced is large. Furthermore, since the burning rate is much slower than that of gas and oil, the unburned content remaining in the ash tends to increase with the decrease in furnace temperature, which has been a problem.

On the other hand, as a method to know the combustion state during boiler operation,
(1) A flame detector attached to the burner nozzle to detect flames, (2) A detector attached to the exhaust gas outlet or flue to detect the components contained in the exhaust gas, (3) In the furnace ITV (Indastrial Tel) installed with an explosion-proof mechanism at the top of the furnace to obtain information
evision) cameras, etc. Such a detection end is
The item (1) is for detecting ignition or extinguishing of the burner, and the item (2) is attached for detecting whether or not the limit value defined by the environmental regulations is exceeded. The ITV camera of (3) is for monitoring the combustion state of the entire furnace. (Fig. 2).

However, such devices conventionally installed have the following drawbacks.

The flame detector is a device that detects "ignition" and "extinguishing" of the flame at the burner outlet. If the flame flies from the burner nozzle and is "igniting", an incorrect output of "fire extinguishing" is output. There is a possibility of making a decision, and the decision is left to the operator. This is because the flame detector can basically output only ON (ignition) and OFF (fire extinguishing) signals at the burner outlet.

The detector for exhaust gas components requires tens of seconds to several minutes (for ash unburned matter, several hours to several days) in the analysis time, and there is a problem in the real-time property of the analysis value, and the furnace It was only a clue to know the combustion state.

The ITV camera attached to the upper part of the furnace takes a picture of the flame from the opposite burner, but since the flame is seen in a swirling state, it is necessary to rely on the experience and intuition of the operator to judge the combustion state. I didn't get it. (Note that the related device as the above conventional device is disclosed in, for example, Japanese Patent Laid-Open No. 49-93
091 publication).

In order to solve the above-mentioned drawbacks, a combustion state diagnosis device has been developed in recent years, in which the state inside the furnace is captured by an ITV camera and the combustion state is quantitatively grasped based on this.

However, in the conventional combustion state diagnosis device, the analysis method of data obtained from the ITV camera is logically incorporated, so that the analysis method and the like can hardly be changed by the worker or the like, and there is no versatility. Moreover, there is a problem that it is inconvenient to handle.

[Object of the Invention]

An object of the present invention is to provide a combustion state diagnosis device and a combustion state control device which are highly versatile and easy to handle.

[Outline of Invention]

A combustion state diagnosis apparatus according to the present invention includes an image pickup unit that captures a flame image of a flame in a furnace, a knowledge database that stores a relationship between a feature amount related to flame and a combustion state, and image processing of the captured flame image. And a knowledge processing means for inferring a combustion state based on the extracted feature quantity and the relationship stored in the knowledge database. It is a feature.

A combustion state control device according to the present invention includes: an image capturing unit that captures a flame image of a flame in a furnace; and a knowledge database that stores a relationship between a feature amount related to a flame, a combustion state, and an operation amount for adjusting the combustion state. A feature amount extraction means for performing image processing on the captured flame image to extract the feature amount, and the combustion state and the operation based on the extracted feature amount and the relationship stored in the knowledge database. It is characterized by comprising inference means for inferring the amount and operation control means for adjusting the combustion state according to the inferred operation amount.

Here, as the characteristic amount, the effective flame area and /
Alternatively, the distance from the burner mouth to the effective flame may be used.

Further, it is preferable that the combustion state diagnosis device and the combustion state control device are provided with display means for displaying an inference result.

Example of Invention

 Hereinafter, various examples of the present invention will be described.

A combustion state diagnosis apparatus according to a first embodiment of the present invention will be described with reference to FIGS. 1, 3 and 4. The present embodiment shows an example in which the present invention is applied to a power generation boiler that uses pulverized coal as a fuel.

As shown in FIG. 3, the burners 31a, 31b, 3 of each stage are
Air is introduced into the furnace 30 together with the fuel (pulverized coal) from 1c, heat generated by combustion in the furnace 30 is supplied to the heat transfer tubes 32, and exhaust gas is discharged into the atmosphere from a chimney (not shown).

It is empirically known that in such a power generation boiler, the state of the flame at the root of the burner greatly influences the combustion state including the downstream thereof. As shown in FIG. 3, the combustion state diagnosis device of this embodiment has the following configuration.

Protect the flame of the burner root from the protective tubes 21a, 21b, 21c.
Image fibers 22a, 22b, 22c protected by
Is measured as an image by using and is taken into the ITV cameras 23a, 23b, and 23c installed outside the furnace. ITV camera 2
The flame images captured by 3a, 23b, and 23c are stored in the frame memory 25 after A / D conversion 24a, 24b, and 24c, and are taken into the computer (processor) 10.
After performing image processing, feature amount extraction processing, and knowledge processing, the processing results are output to a predetermined display device 29.

As shown in FIG. 1, the computer 10 has an image input unit 1, an image processing unit 2, a feature amount extraction unit 3, a knowledge processing unit 4, and a knowledge database 9 in which various kinds of knowledge are stored. Is configured. Although the contents of the knowledge database 9 are stored in the memory,
A floppy disk or the like may be used for management.

Next, the operation of this embodiment will be described with reference to the flowchart shown in FIG.

In step 1, the image fibers 22a, 22b, 22
c and the ITV cameras 23a, 23b, 23c are used to obtain flame images.

In step 2, the flame image is loaded into the computer 10 and image processing is performed.

In this step, the image input unit 1 checks the flame safety and averages (for example, equation (1)) by determining the presence or absence of a flame, averaging the images, and the like.

: Average brightness of the input image X: Brightness of the input image N: Number of samples of the captured image Further, in step 2, the coordinate conversion unit 2a rotates the flame image due to twisting of the image fiber, and parallels the coordinates by cutting out the image. Performs coordinate conversion such as movement.

In step 2, the image processing unit 2b also removes noise components included in the flame image, and also performs half threshold processing by using the equation (2).

(i, j) X _th , (i, j) = (i, j) (i, j) <X _th , (i, j) = 0 ... (2) X (, j); coordinate position i, the average brightness X _th of j; other half thresholding limit above processing, binarization processing, edge processing, straining optionally facilities image processing and the like.

In step 3, in the feature quantity extraction unit 3, as shown in FIG. 7, the center of gravity of the effective flame, the distance X from the burner port to the effective flame effective flame, the mutual distance Y of the effective flames Y, the effective flame area Z, Features such as effective flame circumference are extracted. In addition,
X, Y, and Z are normalized by the burner primary diameter ratio.

In steps 4, 5, 6 and 7, the knowledge database 9 is set in advance.
“The flame or its characteristic amount and the amount of exhaust gas components (NO _x , SO _x unburned components, etc.)” stored as knowledge in “,” the flame or its characteristic amount and the burner operation amount ((fuel amount / air amount ) Ratio, secondary air amount, tertiary air amount, (secondary air amount / tertiary air amount) ratio, after-air amount, etc.) "and the like.

Here, as a specific example, the amount of exhaust gas (here, NO
_x ) and the operation amount estimation in accordance therewith will be described.

In the knowledge database 9, as shown in FIG. 4 (A),
Relationship between NOx and characteristic quantities X, Y, Z, ...
The relationship between the characteristic quantities X, Y, Z, ... And various manipulated variables relating to flames is stored in advance.

The knowledge processing unit 4 calls the knowledge regarding these relationships from the knowledge database 9 and makes an inference.

In step 4, the knowledge for estimating the exhaust gas amount is called to start the inference.

In step 5, for example, when “X is long and Y is short”, it is estimated that “NOx is high” and the process proceeds to step 6. When "Y is long and Z is wide", it is estimated that "NOx is low" and the process is ended immediately.

In step 6, knowledge for operation amount estimation is called to start this inference.

In step 7, for example, when "X is long and Y is short", it is estimated that "combustion speed is fast", and when "X is short and Y is long", "tertiary air amount / secondary air amount ratio" is calculated. Presumed to be “large”.

And if there is a corresponding item as shown in step 7,
In step 8, for example, a message such as "Combustion speed is high, slow combustion speed" is output from the display device 29. If there is no applicable item, "no applicable item" is output in step 9.

As described above, according to the present embodiment, the combustion state during the boiler operation can be quantitatively grasped, and also can be qualitatively (descriptively) grasped by the inference using the knowledge processing.

In addition, the knowledge database is highly versatile because it can easily change and expand its contents, unlike the one that grasps the combustion state using all logic circuits, and uses data according to the characteristics of each plant. Therefore, the handling becomes easy, and the combustion state can be grasped more accurately.

Next, a combustion state control device according to a second embodiment of the present invention will be described.

As shown in FIG. 5, the present embodiment is provided with a combustion control unit 35 which is driven according to the manipulated variable inferred by the computer 10 in the first embodiment. In this embodiment, the basic configuration of the boiler, computer, etc., other than the combustion adjusting unit 35 and the relationship between the combustion adjusting unit 35 and the computer 10, is the same as in the first embodiment. Therefore, the operation is basically the same as that of the first embodiment except that the operation amount is finally controlled.

Similar to the first embodiment, the flame at the root of the burner is measured as an image (luminance or temperature image) using the image fibers 22a, 22b, 22c protected by the protective tubes 21a, 21b, 21c, and the furnace is used. ITV camera 2 installed outside
Take in 3a, 23b, 23c. ITV camera 23
The brightness or temperature flame image captured by a, 23b, and 23c is stored in the frame memory 25 after A / D conversion 24a, 24b, and 24c, and stored in the computer (processor) 1.
0, and image processing, feature quantity extraction processing, and knowledge processing are performed, and the combustion adjustment unit 35 adjusts the amount of air input to each stage burner 31a, 31b, 31c based on the operation amount obtained by exhaust gas estimation and operation amount determination. To do.

The flame image captured in the computer 10 is checked and averaged (e.g., in the same manner as in the first embodiment) by the image input unit 1 by determining the presence / absence of a flame, averaging the images, and the like. Formula (1) is attempted.

Here, an example of the flame safety check will be described with reference to FIG.

The captured flame image is divided into a plurality of areas, and the average brightness (or average temperature) of each divided area or the area of the flame included in that area is obtained. FIG. 7 shows an example of division into 9 areas.

Here, A, B, C; average brightness in the x direction of the divided areas, or average temperature, or the sum of flame areas I, II, III; average brightness in the y direction of the divided areas, average temperature, or flame Sum of areas a, b, c; divided areas in the y direction 1, 2, 3; divided areas in the x direction I; average brightness or average temperature of divided areas, or flame area There is no particular limitation as long as it is a quantity to be expressed, but the quantity obtained in each region is calculated using the equation (3), and as shown in the equation (4), the calculated value A,
B is compared with the limit values TH _a , TH _b , ... Preliminarily obtained from the flame during normal combustion.

In the equation (4), when it is smaller than the limit value, it is set to "1".

Then, as shown in FIG. 8, for example, “A = 0, B =
When 0 and C = 0, it is judged as “safe”, and
When "A = 0, B = 1, C = 1", it is "not safe".
Then, the fact that it is "one sided fire" is displayed. The eighth
In the figure, in the item of safety, "0" represents normal and "1" represents abnormal.

The above check can be easily performed by the same method in the first embodiment.

After that, the same processing as the processing from step 2 to step 7 in the first embodiment is performed.

When inferring the manipulated variables in steps 6 and 7, as shown in FIG.
The knowledge processing unit 4 may infer the operation amount from the relationship between the operation amount and the secondary air amount, the tertiary air amount, the inferred NOx amount, and the like stored in advance in. In the ninth, "0" represents normal and "1" represents abnormal. Further, as a method for estimating the amount of exhaust gas and the like, for example, those shown in Japanese Patent Application No. 59-92872 (estimation of NO _x ) and Japanese Patent Application No. 59-83782 (estimation) may be used.

The case where the input air amount is used as the operation amount will be specifically described.

The manipulated variable determined by the knowledge processing unit 4 is output to the combustion adjustment unit 35. In addition, the combustion control part 35 is specifically an air control valve here.

The combustion control unit 35 gradually changes the valve opening degree based on the operation amount input from the computer 10.

At this time, the valve drive amount should be gradually approached to the target operation amount by adding or subtracting (a _1,2 / CON) etc. several times as shown in equations (5) and (6). Go. Note that (5)
The expression is an expression used when increasing the operation amount, and the expression (6) is used when decreasing the operation amount.

Here, a _1,2 ; secondary air amount of the first stage burner a _1,3 ; tertiary air amount of the first stage burner CONT; constant The operation amount, the combustion state, etc. are output from the display device 29 as in the first embodiment.

As described above, according to the present embodiment, the combustion state is appropriately controlled quantitatively according to the combustion state, so that a good combustion state can be maintained. Furthermore, as in the first embodiment,
Since the combustion state is inferred using a knowledge database that can be easily changed and expanded, it is possible to perform control that matches the characteristics of the plant by partially changing the contents of the knowledge database.

〔The invention's effect〕

According to the present invention, since the combustion state and the like are inferred using the knowledge database that can be easily changed and expanded, the versatility can be improved and the handling can be facilitated.

[Brief description of drawings]

FIGS. 1, 3 and 4 show a combustion state diagnosis device according to a first embodiment of the present invention. FIG. 1 is a functional block diagram of essential parts of the combustion state diagnosis device, and FIG. 4 is an explanatory diagram showing the overall configuration of the combustion state diagnostic device, FIG. 4 (A) is an explanatory diagram showing the stored contents of the knowledge database, and FIG. 4 (B) is an illustration showing the operation of the combustion state diagnostic device. A flow chart, FIG. 2 is an explanatory view showing a conventional device for diagnosing a combustion state, and FIGS. 5 to 9 show a combustion state control device of a second embodiment according to the present invention. FIG. 5 is an explanatory diagram for showing the overall configuration of the combustion state control device,
FIG. 7 is a functional block diagram of a main part of a combustion state control device, FIG. 7 is an explanatory diagram for explaining a flame image input process, FIG. 8 is an explanatory diagram for explaining determination of flame safety, and FIG. FIG. 4 is an explanatory diagram showing the stored contents of a knowledge database used when inferring an operation amount. 1 ... Image input unit, 2 ... Image processing unit, 3 ... Feature amount extraction unit, 4 ... Knowledge processing unit, 9 ... Knowledge database, 1
0 ... Computer, 22a, 22b, 22c ... Image fiber, 23a, 23b, 23c ... ITV camera, 29 ... Display device, 30 ... Furnace, 31a, 31
b, 31c ... Burner, 35 ... Combustion control section.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) The inventor, Hisanori Miyagaki, 5-2-1 Omika-cho, Hitachi-shi, Ibaraki, Ltd. Inside the Omika factory, Hitachi, Ltd. (56) Reference JP-A 59-7824 (JP, A) JP-A-60-159515 (JP, A)

Claims (3)

[Claims] 1. A combustion state diagnosing device for diagnosing a combustion state of fuel injected from a burner into a combustion furnace, and an image pickup means for picking up an image of a flame formed in the combustion furnace; and an image of the imaged flame. From this, image processing means for extracting a flame region of a predetermined brightness or more as an effective flame, the area of the effective flame extracted by the image processing means, the distance from the fuel injection port of the burner to the effective flame, the effective flame When there are a plurality of, the feature amount extraction means for extracting the mutual distance of the effective flames as a feature amount, respectively, and the relationship between the feature amount and the combustion state, and in order to make the feature amount and the combustion state appropriate states Based on the knowledge database that stores the relationship with the operating method, the relationship between the feature quantity and the combustion state stored in the knowledge database, and the extracted feature quantity. Combustion state inference means for inferring the state, determination means for determining whether or not the combustion state inferred by the combustion state inference means is proper, and the combustion state is determined not to be proper In case,
An operation method inference means for inferring an operation method for making the combustion state into an appropriate state, based on the relationship between the characteristic amount and the operating method stored in the knowledge database and the extracted characteristic amount; A combustion state diagnosis device comprising: a display unit that displays the inferred combustion state and the inferred operating method. 2. A combustion state control device for combustion equipment, comprising: a combustion furnace; a burner for injecting fuel into the combustion furnace; and a combustion adjusting section for adjusting the combustion state of the fuel injected into the combustion furnace. In the image capturing means for capturing an image of a flame formed in the combustion furnace, an image processing means for extracting a flame region having a predetermined brightness or higher as an effective flame from the captured image of the flame, and the image processing means. Feature amount extraction means for extracting the area of the effective flame, the distance from the fuel injection port of the burner to the effective flame, and the mutual distance of the effective flames when there are a plurality of effective flames as the characteristic amount. And a knowledge database that stores the relationship between the characteristic amount and the combustion state, and the relationship between the characteristic amount and the operation amount of the combustion control unit for setting the combustion state to an appropriate state, and the knowledge database. The combustion state inference means for inferring a combustion state based on the relation between the characteristic amount and the combustion state, and the extracted characteristic amount, and the combustion state inferred by the combustion state inference means are appropriate. And a determination means for determining whether the combustion state is not proper,
Based on the relationship between the characteristic amount and the operation amount stored in the knowledge database and the extracted characteristic amount, the operation amount for making the combustion state into an appropriate state is inferred, and the operation amount is calculated. A combustion state control device comprising: a manipulated variable inferring means for instructing the combustion control section. 3. A display means for displaying the combustion state inferred by the combustion state inference means and the operation amount inferred by the operation amount inference means. The combustion state control device described.