JPH0535325B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a method for monitoring and diagnosing the combustion state of a boiler, and more particularly to a combustion state monitoring method for estimating and predicting exhaust gas components at an exhaust gas detection position.

[Background of the invention]

Conventionally, the amount of exhaust gas components generated during boiler operation has been detected by providing a detection end at an outlet in a furnace, a flue, or the like.

On the other hand, during combustion, harmful substances such as NO _x and SO _x are generated due to unburned components or chemical changes and are contained in the exhaust gas, but it takes a long time to separate and analyze the detected components.

For this reason, there is a problem in that reduction of these harmful substances or unburned substances during operation must rely on the experience and intuition of the operator. In particular, issues such as NO _x (nitrogen oxides) and SO _x (sulfur oxides), whose production amounts are being regulated from an environmental perspective, and the reduction of unburned matter in ash, which affects combustion efficiency, must be resolved as soon as possible. This is an issue that must be addressed.

Furthermore, in recent years, coal has been reconsidered as an energy alternative to oil, and pulverized coal and CWM (coal/
_Combustion technology for COM (coal/oil slurry) and COM (water slurry) is attracting attention;
Since the amount of unburned matter remaining in the ash increases significantly compared to the combustion of gas, oil, etc., this has a large impact on the environment and efficiency, and new technical measures are required.

Many of these problems can be solved by improving the shape of the combustion flame, and there is a method of measuring the combustion flame near the burner and determining the combustion state near the burner and in the middle and downstream region of the combustion flame. It is greatly influenced by the amount of air mixed in the middle and rear regions or the fuel ratio (bright component/solid carbon component).
Measuring the vicinity of the burner is not enough to quantitatively and accurately evaluate the downstream region of the combustion flame. (Related publicly known examples include JP-A No. 56-23630.) [Object of the Invention] The object of the present invention is to eliminate substances (particularly NO _x , SO _x , dust, etc.) contained in combustion exhaust gas during boiler operation. An object of the present invention is to provide a combustion state monitoring method for quantitatively estimating, in a short period of time, the residual amount of harmful substances or unburned substances that affect efficiency, etc.).

[Summary of the invention]

The present invention focuses on the fact that the position of the high brightness part or the high temperature part of the flame has a correlation with the exhaust gas components, and the amount of the exhaust gas component is determined in advance as a function of the position of the high brightness part or the high temperature part of the flame. , Measure the position of the high brightness part or high temperature part of the actual flame based on the flame image, and calculate the value of the function from the position information of the measured high brightness part or high temperature part of the flame. The present invention is characterized by quantitatively estimating the amount of components of exhaust gas.

A further preferred embodiment is one that takes into account the influence of after-air and also takes into account information in the trailing region of the flame.

[Embodiments of the invention]

First, I will explain the basics.

Regulation values have been set for substances contained in the combustion exhaust gas during boiler operation, especially harmful substances such as NO _x , SO _x , soot and dust, etc., and it is necessary to operate the boiler by keeping the amount generated below the regulation value. There must be.

On the other hand, it is desirable that the combustion efficiency of the boiler is always kept at its maximum during operation. The standard for calculating this efficiency is the unburned content contained in the exhaust gas. As the amount of unburned matter in the exhaust gas increases, the combustion efficiency decreases, resulting in increased fuel consumption to obtain the same output. However, since it takes a long time to detect unburned components, efficiency during operation must rely on experience and intuition.

Recently, the use of coal as a fuel instead of gas and oil is being reconsidered, and pulverized coal,
CWM (coal/water slurry), COM (coal/oil slurry), etc. are beginning to be used as fuel.

In particular, when coal is used as a fuel, the nitrogen components contained in the coal itself are converted into _NOx through combustion, resulting in a large amount of NOx produced. moreover,
Since the combustion speed is much slower than that of gas or oil, the amount of unburned matter remaining in the ash tends to increase as the furnace temperature decreases.

For this reason, the case where pulverized coal is used as fuel will be described below as an illustrative example.

Figure 2 shows four cases of flames with different shapes during pulverized coal combustion. a, b are the flames when pulverized coal is burned at a fuel ratio (volatile content/solid carbon content) between 1.0 and 1.5; c and d are the flames when pulverized coal is burned at a fuel ratio between 1.6 and 2.0. In the example, the area inside the solid line circle is where the flame was actually measured (primary combustion area), and the area inside the dotted line is where the combustion state was traditionally estimated from the flame shape in the primary combustion area. This is an example of a portion (secondary combustion region) that is actually measured in the present invention.

Here, the primary combustion region is a region where volatile matter combustion is the main activity, and the secondary combustion region is a region where solid carbon content combustion is the main activity.

Hereinafter, the correlation between the unburned content in the ash and the flame during combustion of pulverized coal will be explained as an example.

It is clear from the analysis of various experimental data that the amount of unburned matter remaining in the ash of the flames in the four cases of FIG. 2 increases in the order of a, c, b, and d.

There is an extremely high correlation between the size, positional relationship, and temperature of the primary combustion region and secondary combustion region of the combustion flame and the amount of unburned matter remaining in the ash.

Here, if we focus on flames a and b during pulverized coal combustion of coal with a fuel ratio of 1.0 to 1.5, we can see that the shape, size, positional relationship, brightness, and temperature of the primary combustion area and the relationship between the secondary combustion area and the There is a correlation between the combustion state and the amount of unburned matter remaining in the ash.

However, if we focus on flames c and d during pulverized coal combustion of coal with a fuel ratio of 1.6 to 2.0, even though there is a difference in the amount of unburned matter remaining between the two flames,
The shape of the primary combustion region, the size of the region, the positional relationship,
There is almost no change in brightness and temperature. However, there is an extremely high correlation between the shape, size, positional relationship, brightness, and temperature of the secondary combustion region and the amount of unburned matter remaining in the ash. This is thought to be because high fuel ratio coal with a large solid carbon content has a slower combustion speed than low fuel ratio coal, and characteristics of combustion flame appear in the secondary combustion region, which is the combustion region of solid carbon.

In this way, the shape, size, positional relationship, brightness, temperature, and combustibility from the tip of the primary combustion area as well as the shape, size, and size of the secondary combustion area,
Based on the fact that the positional relationship, brightness, and temperature are correlated with the unburned content in the ash, for example, the characteristic parameters (feature quantities) of the flame shape to obtain the estimation index of the unburned content in the ash (I _UBC ) are as shown in Figure 3. stipulated in

In FIG. 3, the high brightness area in the primary combustion area will be referred to as the oxidation flame. Here, for example, as a characteristic parameter representing the characteristics of a combustion flame,
From the primary combustion area, the distance between oxidizing flames X ₁ , the average value of temperature or brightness from the burner tip to the oxidizing flame
From X ₂ , the secondary combustion region, the integral value of the temperature or brightness of the secondary combustion region was considered.

Distance between oxidizing flames X ₁ = dx/dB (1) Average brightness value from the burner tip to oxidizing flame X ₂ = _d 〓 ^x=a I(x)/X (2) Secondary combustion area brightness integral value X ₃ = _d 〓 ^x=c Using I(x)/X (3), the estimated index I _UBC of unburned content in the ash is defined as, for example, I _UBC = K・X ₁・X ₂・X ₃ (4) do. Here K is the primary aperture coefficient I is the brightness. Here, how to determine G ₁ and G ₂ representing X ₁ in Fig. 3: (1) Let G ₁ and G ₂ be the center of the oxidation flame.

(2) The portion of each oxidation flame closest to or farthest from the burner tip is designated G ₁ and G ₂ .

(3) Let G ₁ and G ₂ be the positions with the highest flame temperature.

(4) Find the oxidation flame from the temperature distribution, and find its center of gravity G ₁ ,
Let it be G ₂ .

How to determine a and b to determine X ₂ in Figure 3: (1) Set the burner tip position to a (2) In G ₁ and G ₂ , the point nearer or farther from the burner tip How to determine c and d to find ₃ : (1) They shall be determined arbitrarily depending on the fuel ratio, air injection method, etc.

etc. are possible.

The above is an example of a method for estimating unburned content in ash using flame shape.

Furthermore, if afterair is injected into the downstream side of such a flame, the unburnt UBC in the ash will increase.
The relationship between this and its estimated index I _UBC is shown in Figure 4. As shown in Figure 4, due to the influence of after-air, the estimated target
It was found that the unburned fraction _UBC in the ash has a binary value region (curve A) with respect to I UBC, and a problem arose that I _UBC could not be used as an estimation index.

On the other hand, analysis of various experimental data shows that the unburned UBC in the ash when the maximum amount of afterair is injected and its estimated index I _UBC have a linear relationship as shown by the broken line B in Figure 4. It became clear from the results.

As a result, the unburned content in the ash due to after-air injection
The influence on UBC is determined by the measurement position and
It was found that each amount of after air is expressed by a function (especially an exponential function), and it was found that the unburned content in the ash at the measurement position can be estimated or predicted with high accuracy.

The above description of unburned content in ash is based on the findings obtained by the inventors from various experiments, but other exhaust gas components (NO _x , SO _x , soot, etc.) also show similar trends. An example based on the present invention will be described below, taking unburnt UBC in ash as an example.

An embodiment of the present invention is shown in FIGS. 1 and 6. Figure 1 shows the case where monitoring and diagnosis of unburned UBC in the ash was carried out for a single burner. An image fiber cooled with water or air is inserted through a viewing window in the furnace wall at a position near the burner of the furnace and at a position where the middle and downstream region of the combustion flame can be seen, and an image of the combustion flame is guided to the outside of the furnace.
The flame images guided outside the reactor are converted into electrical signals by each ITV camera. FIG. 6 shows one configuration example of the combustion state monitoring device. Analog video signals 6 and 7 from each ITV camera are converted into digital video signals 8 and 9 via an A/D converter 1 and written into a frame memory 2. The written image data 10 is taken into the processor 3 and calculates the unburned content estimation index I _UBC defined by equation (4). The manipulated variable and measured variable 13 are input to the processor 3 as a digital signal 12 via the process I/O 5.

On the other hand, in FIG. 1, afterair is injected from the downstream part of the combustion flame, and at the measurement position, the amount of reduction in unburned matter in the ash due to afterair is also superimposed and measured. Therefore, the unburned UBC in the ash is estimated using equation (5), which is obtained by adding an estimation term that takes into account the influence of this after air to equation (4).

P(UBC)=K ₁・I _UBC +K ₂・exp(α)+C (5) Here, P(UBC); Estimated amount of unburned content in ash I _UBC ; Estimated index of unburned content in ash α; After air Representing coefficients K ₁ , K ₂ , C; constants (however, K ₂ is a constant that takes into account the distance from the after-air device to the detection position) In equation (5), α representing the influence of after-air is
It is expressed as a function of the amount of after air as shown in equation (6).

α=g(G _AA ……) (6) Here, G _AA : After air amount In addition, the after air amount shown in equation (6) can also be expressed using the air ratio as shown in equation (7). be.

α=g{(λ− _λBNR ),……) (7) Here, λ: Total air ratio _λBNR : Burner air ratio Furthermore, G _AA can also be expressed using the total air amount and tertiary air amount. can. Equation (5) uses an exponential function to express the estimated term due to the input of after-air, but it is also possible to express it using other functions. That is, P(UBC)=K ₁・I _UBC +K ₂・{g(G _AA ),...}+
CP(UBC)=K ₁・I _UBC +K ₂・{g(λ−λ _BNR ),…
...}+C(8).

As an example of the above processing, an outline of the internal processing flow of the processor 3 is shown in FIGS. 7a and 7b. The outline of the process shown in FIG. 7 will be explained next.

100: Input of each flame image data Burner vicinity flame image data IM ₁ (i, j) and combustion flame intermediate and downstream flame image data IM ₂ (i, j) are input to the processor 3.

110: Averaging each flame image data Find the flame shape with the highest probability of indicating the combustion state (an example is shown in equations (9) and (10)) IM ₁ (i, j) = 1/N _N 〓 ^k=1 {IM ₁ (i, j)} _k (9) IM ₂ (i, j)=1/N _N 〓 ^k=1 {IM ₂ (i, j)} _k (10) Here, IM ₁ (i, j): Averaged burner vicinity flame image IM ₂ (i, j): Averaged combustion flame intermediate and trailing flame image k: Number of samples for averaging (k = 1 to N) 120: Each flame Shape feature extraction Using image processing, high brightness and high temperature areas (oxidation flame) of the flame are extracted from the flame image data near the burner, and the distance between the centers of gravity of these extracted areas and the distance from the burner to the center of gravity are calculated. The brightness average value is calculated, and for the combustion flame intermediate and downstream flame images, flames are extracted from the flame image data and the brightness area is calculated.

130: Calculation of unburned content estimation index I _UBC in ash Calculate unburned content estimation index I _UBC in ash using equation (11).

I _UBC = X・Y・Z・K+C _{1 (} 11) where, : Is after-air provided? Determine whether it is necessary to consider after-air effects.

Judgment Necessary (Yes): G _AA > 0 Not necessary (No): G _AA > 0 Where, G _AA : After air amount 150: Estimated amount of reduction in unburned content in ash due to after air injection After air is input and combustion Estimate the amount by which the amount of unburned matter in the ash decreases as the process progresses.

P={g(G _AA ,...)}+C ₂ (12) Here, C ₂ : Constant P : Estimated amount of decrease G _AA : After air amount In equation (12), the function g(G _AA ,...) is , shows that it is a function containing at least G _AA .

160: Estimating the unburned content in the ash Using the I _UBC and P obtained earlier, estimate the unburned content in the ash using equation (13).

P(UBC)=K ₁・I _UBC +K ₂・P+C (13) Here, P(UBC): Estimated unburned content in ash I _UBC : Indicator for estimating unburned content in ash K ₁ , K ₂ , C: Constant P: Estimated amount of decrease in unburned content in ash 170: Output of estimation result The estimated amount P (UBC) of unburned content in ash is output to the output device.

Further, the outline processing of FIG. 7b is as follows.

121: Extraction of high brightness and high temperature areas from flame image data near burner (half threshold processing) High brightness and high temperature areas (oxidation flame) as flame features
, the area is extracted by half-threshold processing. Here, half-threshold processing refers to
It refers to processing an image using equation (9).

When IM ₁ (i, j) TH, IM ₁ (i, j) = IM ₁
(i, j) When IM ₁ (i, j)<TH, IM ₁ (i, j)=0 (14) Here, IM ₁ (i, j): Averaged flame image near the burner.

TH: Half-threshold level 122: Calculate the center of gravity of the high-brightness, high-temperature region Calculate the center of gravity of the high-brightness, high-temperature region (oxidation flame) extracted using half-threshold processing. In this embodiment, the center of gravity of the area is taken as its representative point, but the same effect can be expected if the maximum brightness, maximum temperature point, etc. are used as the representative point.

123: Flame extraction from image data of combustion flame middle and trailing region flame (half-threshold processing) In order to obtain the brightness area of the combustion flame middle and trailing region flame, extract the combustion flame middle and trailing region flame from the image data by half-threshold processing. Here, half-threshold processing means (14)
Similarly to the formula, when IM ₂ (i, j) TH, IM ₂ (i, j) = IM ₂
(i, j) When IM ₂ (i, j) < TH, IM ₂ (i, j) = 0 (15) where, IM ₂ (i, j): averaged combustion flame trailing region flame image TH: Half-threshold level 124: Calculate distance between centers of gravity (X) Using the center of gravity of the oxidizing flame in the flame image data near the burner, which has been processed, the unburned content in the ash is estimated.
I Find X (distance between centers of gravity of oxidizing flame), which is one of the characteristic parameters for finding _UBC .

125: Calculate the average brightness value from the burner tip to the center of gravity (Y) Using the oxidation flame center of gravity position, burner tip position, and flame brightness information of the flame image data near the burner, which has been processed, calculate the unburned content in the ash. Y (average brightness from the burner tip to the oxidizing flame), which is one of the characteristic parameters for determining the estimated index I _UBC , is determined.

Note that according to this embodiment, the following effects can also be obtained.

(1) It becomes possible to estimate or predict exhaust gas components in real time, and highly efficient operation can be achieved.

(2) It is possible to grasp the production amount of the component of interest in each stage, enabling fine-grained operation control.

(3) By directly measuring the combustion state of the burner,
Boiler operating status can be accurately grasped.

(4) Exhaust gas components can be accurately determined even under various combustion conditions, such as pulverized coal with a high fuel ratio and CWM.

(5) The burden on the operator can be reduced.

A similar effect can also be expected by averaging the temperature of the oxidizing flame from the tip of the burner.

126: Calculate the brightness integral value of the flame in the middle and trailing region of the combustion flame (Z) Using the image data of the flame in the middle and trailing region of the combustion flame that has been processed, the characteristic parameters for calculating the unburned content estimation index I _UBC in the ash are calculated. One of these, Z (luminance integral value of the middle and rear region of the combustion flame), is determined.

Furthermore, a similar effect can be expected with the temperature integral value in the downstream region of the combustion flame.

As described above, by using the present invention, it is possible to estimate the unburned content in the ash from the flame image and accurately estimate or predict the unburned content in the ash at the measurement position.

As another embodiment, FIG. 8 shows a case where a plurality of different burners are monitored by the combustion state monitoring and control method according to the present invention. In this case, there is a method of switching the image input section of the combustion state monitoring and control device when inputting images of stages A, B, and C (Fig. 9a), and switching the A/D converter and frame memory respectively to stages A, A', and B. , B', C, and C' stages and simultaneously input images to the frames of the three stages (Fig. 9b). The internal processing of the processor is basically the same as that shown in FIGS. 7a and 7b. The schematic process is shown in FIG.

For example, by using the present invention to monitor the combustion state of an actual boiler, it is possible to monitor the burner combustion state of each stage, that is, the boiler operating state, and it is possible to realize detailed and highly efficient operation that takes into account the influence of afterair. In addition, from the estimated value of unburned content in the ash of the present invention,
By controlling the amount of operation (air amount, air ratio, etc.), the burden on the operator can be further reduced.

Furthermore, the invention is not dependent on burner type. For example, even if the burner types are different as shown in FIGS. 11a and 11b, when the combustion flame is measured from the cross-sectional direction of the burner, it is clear that the flames formed have the same shape in both a and b.

〔Effect of the invention〕

By implementing the present invention, exhaust gas components, for example, unburned content in ash (UBC), can be monitored with high precision.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing one embodiment of the present invention, Fig. 2 is a diagram comparing the flame shapes that are the basis of the present invention, and Fig. 3 is a diagram showing one embodiment of the present invention.
Figure 4 shows the characteristic parameters extracted from the flame shape, Figure 4 compares the unburned content in the ash and its estimated index with regard to the influence of after-air, and Figure 5 shows the overestimation of the decrease in the unburned content in the ash. A diagram showing the input amount and distance, FIG. 6 is a diagram showing an example of the device configuration of the present invention,
7a and 7b are diagrams showing a schematic processing flow of the processor, FIG. 8 is a diagram showing another embodiment of the present invention, and FIG. 9 is a diagram showing an example of an image input method of another embodiment.
FIG. 10 is a diagram showing a schematic processing flow of another embodiment, and FIG. 11 is a diagram showing a different type of burner. 1... A/D converter, 8, 9... Digital video signal.

Claims (1)

[Claims] 1. A method for monitoring the combustion state in a boiler using an image of the combustion flame, in which the component amount of the exhaust gas is predetermined as a function of the position of the high-intensity part or the high-temperature part of the flame, and The position of the high-intensity part or high-temperature part of the flame is measured based on images of the base and trailing area of the flame, and the value of the function is calculated from the position information of the measured high-intensity part or high-temperature part of the flame. A combustion state monitoring method characterized by quantitatively estimating the amount of exhaust gas components by calculation.