KR100356272B1 - A combustion method for boiler burner - Google Patents

Abstract

Disclosed is a boiler burner combustion diagnosis method for discriminating an image signal of a burner flame by a plurality of discrimination criteria.

According to the present invention, in the combustion diagnosis method of a boiler burner, the image information is read from a flame imager to obtain the average luminance Ia of the effective flame region, and if the average luminance is larger than the set value α1, it is determined as a misfire. If the average luminance is smaller than the set value α2, it is judged as ignition. If the average luminance is an intermediate value between the set values α1 and α2, the number of pixels Hn in the luminance level setting area is obtained. If larger than the value (N1), it is achieved by the combustion diagnostic method of the boiler burner comprising a plurality of phantom determination step of determining the misfire and the ignition if small.

Description

A combustion method for boiler burner

The present invention relates to a combustion diagnosis method of the boiler burner, and more particularly, to a combustion diagnosis method of the boiler burner to discriminate the video signal of the burner flame by a plurality of discrimination criteria.

Conventionally, as a combustion diagnosis system of a boiler burner, a method of monitoring the flame by calculating the temperature as a ratio of G / R of Mitsubishi Heavy Industries in Japan has been known as JP-A-64-28413.

The patent of Mitsubishi Heavy Industries, Japan in accordance with the claims of the above-mentioned patent publication discloses a video signal transmitted by photographing a flame of a flame having a plurality of burners with a color TV camera in a standard method of color TV in a signal processing apparatus. It is a combustion monitoring system which displays the temperature of several preset sections of the photographing space image surface as a ratio of two-color signals separated by R (red), G (green), and B (blue) due to the sensitivity curve.

That is, by comparing the flame image when no fuel is detected with the image when the fuel is supplied, the ignition state is identified using the change amount of the image, and the characteristic shape of the burner flame having an average brightness in the effective area (burner flame area) Determination of misalignment and misfire by combining the detection of the parameters related to the unburned state and the average of the R component as the optimal stability data over the full load region among the image data of the R, G, and B components based on the point where the luminance level is high. The flame stability is determined using the time variation of luminance and average luminance as an evaluation factor, and the flame temperature distribution of each point is calculated based on the two-component ratio of R and G by applying the principle of a two-color thermometer, and then the temperature relation with R and G. And after setting the temperature measurement range of the heavy and crude oil boiler, this is to determine the practical range in the burner combustion state evaluation to measure the flame temperature (low NOX), smoke and dust generation in the burner An evaluation coefficient (flame transparency) indicating the amount of light, the luminance low luminance area of each pixel within the upper and lower limits based on the luminance distribution of the active component, defined as the pixel area within the upper and lower limits, and the transmittance and luminance low luminance area and the dust amount in the graphs presented. Flame transparency is judged as a correlation of.

1 shows an example of application of such a boiler burner combustion diagnosis apparatus.

That is, the burner (2) is installed in the boiler furnace (1) to generate a flame using coal, oil, gas, etc. to heat the heat exchange means of the boiler. The flame generated in this way transmits the image of the flame to the CCD camera 4 through the high temperature image receiving unit 3. The image signal input to the CCD camera 4 outputs a combustion diagnosis result using the flame image as an input signal from the combustion diagnosis discrimination unit 5, thereby indicating a combustion diagnosis state in the combustion diagnosis computer 6.

This combustion diagnosis result is shown on the monitoring diagnostic monitor 7 via the flame detection diagnostic diagnostic circuit.

Such a diagnosis method or other known diagnosis method is pattern matching by extracting a specific signal of an image signal, or by using a physics discrimination determination by a database, or by databaseting characteristics by optical spectrum analysis to establish a state of physics. This diagnostic method is easy to distinguish the combustion state of a specific burner or a specific temperature range, but the change of the specific point becomes larger when the combustion conditions are different, and the noise removal according to the extraction characteristics of the specific data is difficult. It can be a problem.

Therefore, there is a need for a stable detection method that can cope with changes in various combustion conditions while being less affected by disturbance or noise.

The present invention is to input the video signal of the burner flame for the operation to reduce the generation of pollutants while making the combustion state of the burner flame to the optimum combustion state, by determining the combustion state for each image signal by a plurality of discrimination circuits The purpose of the present invention is to provide a diagnostic method for combustion combustion of the burner to improve the accuracy of the burner identification.

1 is a schematic diagram showing a combustion diagnosis process of a typical boiler burner

Figure 2 is a block diagram of the combustion diagnostic apparatus of the present invention burner burner

Figure 3 is a flow chart showing a combustion burner burner diagnostic process of the present invention

※ Explanation of code for main part of drawing

8: Flame video signal ..... 9: Computation processing unit

10: Flame image variable extracting unit ..... 11: combustion monitoring diagnosis processing unit

12: Exhaust gas O2 index calculation unit ........................ 13: NOX index calculation unit

14: CO Indicator Calculation ............ 15: Flame Stability Calculation

The present invention for achieving the above object is to read the image information from the flame imager to obtain the average luminance (Ia) of the effective flame region, and if the average luminance is larger than the set value (α1), it is determined as a misfire, the average luminance Is smaller than the set value α2, determining that the image is ignited; and when the average luminance is an intermediate value between the reference value α1 and the deformation value α2, the reference value N1 in which the number of pixels Hn of the luminance level setting region is defined. If larger than), it is preferable to use a multiple realization discrimination step for determining misfire and small misfire.

That is, after reading the image information from the flame imager, the average luminance of the effective flame region, which is the first discrimination criterion, is obtained. If both values are larger than the two reference values, it is determined as an ignition. If the average luminance is two intermediate values, the number of pixels of the HUE LEVEL specific region, which is the second discrimination criterion, is obtained. Through this process, it is possible to minimize the possibility of misjudgment by ignition by mixing the flame light of the surrounding burner or the reflected light of the furnace wall. In this way, by determining whether the combustion stability with a step-by-step discrimination criteria, it is possible to accurately determine whether misfire during low load operation.

When described in more detail based on the preferred embodiment of the present invention shown in the accompanying drawings as follows.

1 is a schematic diagram illustrating a combustion diagnosis process of a general boiler burner, and FIG. 2 is a block diagram of a combustion diagnosis apparatus of a boiler burner of the present invention.

According to this figure, the burner 2 is installed in the boiler furnace 1 to generate a flame using coal, oil, gas, etc. to heat the heat exchange means of the boiler. The flame generated in this way transmits the image of the flame to the CCD camera 4 through the high temperature image receiving unit 3. The image signal input to the CCD camera 4 outputs a combustion diagnosis result using the flame image as an input signal from the combustion diagnosis discrimination unit 5, thereby indicating a combustion diagnosis state in the combustion diagnosis computer 6.

This combustion diagnosis result is shown on the monitoring diagnostic monitor 7 via the flame detection diagnostic diagnostic circuit.

2 is a configuration diagram of a combustion diagnosis determining system. When the flame image signal 8 is inputted to the combustion diagnosis computer 6, the signal is converted into an RGB value by an image processing card from the HIS (Hue, Intensity, Saturation) calculation processor 9 to 0-255. It is stored as a value with a level up to. Each pixel of the image is statistically processed for each of H, S, and I, and the result value is input to the flame image variable extracting unit 10. The flame image variable extracting unit 10 extracts the flame image variable through a correlation calculation with the exhaust gases by a cross correlation operation. The combustion monitoring diagnosis processing unit 11 outputs four output values using the flame image variable value, which is an exhaust gas O2 index calculation unit 12, a NOX index calculation unit 13, a CO index calculation unit 14, and a flame. It is used as the data of the stability calculation unit 15. Reference numeral 16 denotes a combustion monitoring unit.

The process of the present invention according to the flow chart of the combustion monitoring system of the boiler burner of FIG. 3 is as follows.

<Step 1>

Image information is read from the flame imager and the average luminance Ia of the effective flame region is obtained.

<Step 2>

After setting the average luminance value to 3 sections, if it is larger than the reference value, it is determined as misfire, and if it is small, it is determined as ignition. As is done.

In the multiple illusion determination step, if the average luminance Ia is smaller than the predetermined value α1, it is determined as true. At this time, if the average luminance Ia is greater than the predetermined value α1, it is determined whether the average luminance Ia is smaller than the set deformation value α2, and if it is larger than the deformation value α2, it is immediately determined as ignition.

However, if the average luminance Ia is smaller than the modified value α2, the number of pixels Hn of the luminance level setting region is obtained again. If the number of pixels Hn is larger than the predetermined value N1, it is determined as true. If it is, it is judged to be ignition.

As described above, the present invention has the effect of increasing the accuracy of discriminating the burner by discriminating the burn state of the burner flame with the plurality of discrimination circuits.

Claims (1)

In the combustion diagnosis method of the boiler burner such as a thermal power plant, Reading the image information from the flame imager to obtain an average luminance Ia of the effective flame region; And If the average luminance Ia is greater than the set reference value α1, the determination is true, and if the average luminance Ia is less than the deformation value α2, the determination is ignition; and the average luminance Ia is the reference value α1. And a plurality of realization determination steps for determining if the number of pixels of the luminance level setting region is greater than the reference value N1 if it is a middle value of the deformation value α2, and if it is small, ignition.