CN104896506B - Building method for combustion energy radiant energy signal in coal-fired thermal power generating unit boiler - Google Patents

Abstract

A method for constructing combustion energy radiation energy signals in a coal-fired thermal power unit furnace belongs to the technical field of boiler combustion monitoring and control in thermal power plants. First of all, the image information of the burning flame is obtained through multiple flame image detectors arranged at different heights along the furnace, and the gray value of each image is calculated in real time and converted to the same range as the actual power of the unit, which is defined as is the initial radiant energy signal. Thirdly, the initial radiant energy signal is dynamically compensated for data processing using the value of the unit's actual power generation to obtain the final radiant energy signal. Finally, the radiant energy signal is output to the coordinated control system of the thermal power unit as the detection quantity of the energy level in the furnace. The application results show that the radiant energy signal constructed by this method can effectively reflect the pulsation of the furnace flame, and eliminate the signal deviation caused by factors such as detector dust accumulation and coking, and can be applied to continuous online optimization of different types and different capacity units control.

Description

A construction method of combustion energy radiant energy signal in coal-fired thermal power unit furnace

technical field

The invention relates to a method for constructing combustion energy radiation energy signals in a furnace of a coal-fired thermal power unit, which belongs to the boiler combustion monitoring and control technology of a thermal power plant.

Background technique

Using the furnace flame radiation imaging monitoring system to obtain radiant energy signals and accurately reflect the online monitoring technology of the energy released during the combustion process in the furnace (Zhou Huaichun, Principles and Technology of Visual Flame Detection in Furnace, Science Press, May 2005, pp.306 -309) Some achievements have been made in the field of coordinated optimization control of thermal power units.

This technology is to install multiple high-temperature flame image detectors at different heights and positions of pulverized coal combustion boilers, take pictures of flame images in the furnace, and obtain radiant energy signals through image processing technology, radiation heat transfer theory and advanced solution strategies. Since the brightness of the flame color image is directly proportional to the radiant energy received by the image detector, under the condition of fixed image acquisition and processing conditions (lens aperture, camera shutter, gain and white balance and other conditions that affect image information do not change), the Obtaining the relative radiant energy signal directly from the gray value of the image can be called the initial value of the radiant energy signal. This value can quickly reflect the combustion status of the furnace space and has a strong correlation with the unit. However, the detector is affected by coking, dust accumulation, and lens discoloration during use, and the initial radiant energy signal value will be unstable, with large pulsations and poor reliability. Case. On this basis, the Chinese patent literature discloses a method for correcting the static deviation of radiant energy [application number: 201310301791.7], and introduces a specific solution, which eliminates the influence of various factors on the accuracy of radiant energy signals to a certain extent. influences.

In order to improve the optimal calculation of initial radiant energy, better adapt to different types and capacities of units, and apply it in long-term engineering practice, a newest radiant energy signal detection and construction method is proposed.

Contents of the invention

The purpose of the present invention is to propose a method for constructing combustion energy radiant energy signals in the furnace of a coal-fired thermal power unit, so that the obtained radiant energy signals have the characteristics of high accuracy, strong applicability, and no interference.

Technical scheme of the present invention is as follows:

A method for constructing a combustion energy radiant energy signal in a coal-fired thermal power unit furnace is characterized in that the method includes the following steps:

1) Install multiple flame image detectors at different heights along the furnace to obtain the flame radiation image information in the furnace, and calculate the gray value GSU _i,j of the image in real time through image processing technology, the expression:

GSU _i,j = 0.11R+0.59G+0.23B (1)

Among them, i is the number of flame detectors, j is the number of images collected by each detector, and each image is calculated to obtain a gray value; R, G, and B are the information values of the three primary colors of the image, red, green, and blue, respectively, And define the gray value GSU _i,j of the image as the initial radiant energy;

2) The dynamic range conversion of the initial radiant energy is performed based on the actual generating power of the unit, according to the formula (2-3):

E _i,M =k _1,i GSU _i,M (3)

Among them, k _1,i is the proportional coefficient of the initial radiant energy signal obtained by the i-th detector, M is the length of the data segment for calculating the proportional coefficient, P _j is the actual power of the unit in this data segment, and GSU _i,M is The value of the initial radiant energy signal obtained by the i-th detector at time M, E _i,M is the radiant energy value converted to the same range as the unit load at time M;

3) Screen the initial radiant energy signal that correctly reflects the load change of the unit, according to formula (4-5):

|E _i,M -P _M |≤E _thr (4)

Among them, P _M is the actual transmission power corresponding to time M, E _thr is the set threshold, N is the total number of detectors, N′ is the number of detectors satisfying formula (4), and E ₀ is the filtered The mean value of the radiant energy signal that correctly reflects the energy change;

4 ₎ Optimizing the average value E0 of the radiant energy signal of the unit in the stage of severe load changes:

E=k ₂ ·E ₀ (7)

Among them, k ₂ is the second correction coefficient, a', b' are coefficients calculated by the least squares method between the mean value of the power change of k _1,i random group and the generated power P, is the average value of power generation, and E is the energy level in the reactor and the final output radiation energy.

Compared with the corresponding technology, the present invention has the following advantages and outstanding technical effects:

The present invention mainly uses two proportional coefficient correction links and one threshold value judgment link. The first proportional correction is to convert the gray value of the image, that is, the initial radiant energy value into the same range as the actual power of the unit; the threshold value judgment link It is to eliminate the influence of error information caused by factors such as coking and dust accumulation during the operation of the image detector. The second proportional correction is based on the linear relationship between the first correction coefficient and the power of the unit to eliminate the deviation of the radiant energy signal that is too large or too small during the severe load change stage; the final radiant energy signal is the detection signal of the energy level in the furnace. When the combustion instruction changes, the radiant energy signal can be reflected in time, and its value is too large or too small to reflect that the combustion rate in the furnace is either "supply exceeds demand" or "supply is less than demand", and the radiant energy signal constructed by this method It is not affected by the measurement noise, and can be used as the operating parameters of the unit to be sent to the coordinated control system to participate in the combustion optimization control.

The practical results prove that the output radiant energy signal can not only effectively reflect the pulsation of the furnace flame, but also eliminate the signal deviation caused by the ash accumulation and coking of the detector, and can be applied to coal-fired units of different models and capacities. It provides a good online detection method for the optimal control of boiler combustion and unit coordination.

Description of drawings

Figure 1 is a structural diagram of the online flame monitoring system.

Fig. 2 is a flowchart of a method for constructing a radiant energy signal.

Figure 3 is the comparison curve between the gray value of the image and the actual transmission power.

Figure 4 shows the comparison between the gray value of the image obtained by the detector CCD13 and the actual transmission power. It can be found that there are large error messages at 9:10 and 9:40.

Fig. 5 is the judgment of the threshold value of the original radiant energy signal, which shows the representative of each layer in the 4-layer flame detector, and the error value is eliminated by the upper and lower limits of the threshold value.

Figure 6 shows the mean value of the original radiant energy signal after threshold judgment. The number of sources for calculating the mean value is dynamic. Compared with the actual generating power of the unit, the difference between the two can be seen.

Figure 7a and Figure 7b show the initial radiant energy signal error in the lifting and lowering phase, and those that are too small and too large are marked with circles in the figure.

Figure 8a and Figure 8b show the comparison between the final radiant energy signal and the actual power of the unit.

Figure 9a, Figure 9b and Figure 9c are the application of the radiation energy detection system in 200MW, 300MW, and 600WM boilers respectively, and the comparison curves of the final output radiation energy signal and the actual power in one day.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

The present invention provides a method for constructing combustion energy radiant energy signals in a coal-fired thermal power unit furnace, and its specific implementation method is as follows:

1. The flame detectors installed at different heights in the furnace 1 of the coal-fired boiler obtain the flame radiation image information in the furnace, and send it to the hard disk video recorder through the video acquisition system 2 to synthesize an image 3, which is calculated in real time by image processing technology The RGB value of the image is calculated to represent the gray value GSU (gray-scale-unit) of the relative radiant energy signal, such as the expression:

GSU _i,j = 0.11R+0.59G+0.23B (1)

Among them, i is the number of flame detectors, j is the number of images collected by each detector, and each image is calculated to obtain a gray value.

2. Furnace radiant energy signal is a reflection of the energy released by fuel burning in the furnace. It is transmitted to the steam-water system through radiation heat transfer, convective heat transfer, etc., and finally converted into output power of the generator. Therefore, the change of the furnace radiant energy will eventually lead to the change of the power generation of the unit, and the direction of the change of the furnace radiant energy and the actual power generation of the unit is consistent. Based on this, the dynamic tracking calculation of the gray value GSU _i,j representing the radiant energy signal is performed based on the actual power of the unit as a reference, according to formula (2):

k _1,i is the proportional coefficient of the initial radiant energy signal obtained by the i-th detector, M is the length of the data segment for calculating the proportional coefficient, P _j is the actual power of the unit in this data segment, GSU _i,M is the i-th The value of the initial radiant energy signal obtained by a detector at time M:

E _i,M =k _1,i GSU _i,M (3)

E _{i, M} is the radiant energy value converted to the same range as the generating power of the unit at time M;

3. Screening the original radiant energy signal that correctly reflects the load change of the unit, the method is to compare with the generating power of the unit

Calculate the difference between the radiant energy signal E _i,M of the same range and the current actual power P _M , as follows:

|E _i,M -P _M |≤E _thr (4)

E _thr is the set threshold, and P _M is the actual transmission power value corresponding to time M. The radiant energy signal that does not satisfy the above formula indicates that the error information contained in the picture is large and needs to be eliminated, according to formula (5):

Among them, N is the total number of detectors, N′ is the number of detectors satisfying formula (4), and E ₀ is the mean value of the radiant energy signal that correctly reflects the energy change after screening. The selection of E _thr not only preserves the information of detectors capable of normal energy fluctuations in the reaction furnace, but also eliminates the information of detectors that are invalid or subject to major factors such as coking, so as to limit the error range. It should be noted that due to the difference between the combustion working condition and the detector working condition, the number N′ of the original radiant energy signal which correctly reflects the energy change in the furnace also changes dynamically.

4. When the single radiant energy signal is corrected in the period of severe unit load changes, the radiant energy is obtained by the product of the current measured value and the correction coefficient k _1,i . Due to the rapid response capability of the radiant energy signal itself, it will increase rapidly when the amount of combustion increases. At this time, the power of the unit is affected by the delay of the steam-water heat transfer system, and the increase speed is slower than that of the radiant energy signal. The correction coefficient k _1,i calculated by the average value will be too small, resulting in a small average value of the radiant energy signal E ₀ . Similarly, when the load of the unit is reduced, the average value of the radiant energy signal E ₀ will be too large. Therefore, a second proportional correction coefficient k ₂ is introduced to set is the mean value of the power change of k _1,i random group, with Dynamically correct the linear relationship with the generating power of the unit, as shown in formula (6-7):

Among them, P is the unit load at the current moment, is the average load over time.

After the processing of this step, the expression of the final radiant energy signal E is:

E=k ₂ ·E ₀ (8)

Example:

The present invention has corresponding hardware and software support, and its specific implementation method is as follows:

The flame detectors installed at different heights of the furnace obtain the flame radiation image information in the furnace. Taking 16 flame detectors divided into 4 layers, each layer of 4 is applied to a 300MW subcritical boiler as an example. Its structure is shown in Figure 1 shown.

First set the parameters of the CCD camera in the flame detector to be fixed to ensure that the image is clear and not saturated, and drive the SDK2000 video acquisition card to complete the acquisition of flame radiation images. The 16-channel video signal is synthesized by the hard disk video recorder into a flame radiation image and sent to the computer system.

The steps of constructing the radiant energy signal are shown in Fig. 2 . The normal flame image brightness captured by each detector represents the effective radiant energy of the furnace wall surface and the space within the CCD field of view through emission, absorption, and scattering to reach the target surface, so its value has the same variation trend as the actual power generation , but affected by the working environment and state of the detector, there will be coking and dust accumulation, which will cause the image to be unreal, and the corresponding gray value of the image representing radiant energy will deviate, such as the detector CCD13 in Figure 3. Partially enlarged Figure 4, it can be clearly found that there are large fluctuations at 9:10 and 9:40, which are caused by the coking of the detector. After the focus falls off, the gray value of the image returns to the normal value.

Using the correction coefficient one, the initial gray value is dynamically calculated, and the processing of the same range as the actual power of the unit is completed. The threshold E _thr =30 is set, that is, when the difference between the original radiant energy value and the actual emitted power is greater than the threshold, the corresponding radiant energy signal will be rejected and will not participate in subsequent radiant energy calculations. As shown in Figure 5, the intermediate value judged by the upper and lower limits of the threshold is a normal signal, and the original radiant energy signal that meets the requirements is averaged to obtain the initial value of the radiant energy signal, as shown in Figure 6.

When the unit is running in the variable load stage, such as when the load is rising, because the response speed of the radiation energy signal to the increase in combustion is faster than the unit’s actual power, the first correction coefficient will be too small, so the initial value of the radiation energy obtained is too small , similarly, the initial value of radiant energy will be too large when the load of the unit is reduced. As shown by circle marks in Fig. 7a and Fig. 7b. Using the average value of the correction coefficient k _1,i under each load For the linear relationship of unit power, use the second correction coefficient k ₂ obtained by formula (7) to fine-tune the ratio of the initial value of the radiant energy signal, and pay attention to the calculation of the average value of the actual power The length of the time period can be adjusted according to the actual effect. Finally, the final radiant energy signal E is calculated by formula (8), and the results are shown in Fig. 8a and Fig. 8b.

The radiant energy signal constructed by this method can correctly reflect the change of combustion energy in the furnace, and its physical meaning represents the corresponding relationship between the current combustion heat release in the furnace and the power energy demand of the unit. The radiation energy calculation method of the present invention is applicable to units of various types and capacities, as shown in Fig. 9a, Fig. 9b and Fig. 9c. The difference between the radiant energy signal and the unit load reflects that the combustion adjustment in the furnace is not in place. Further narrowing the difference between the two is the goal of optimizing the control of the thermal system of the boiler and steam turbine, and it is also the meaning of the present invention.

Claims (1)

1. A construction method of combustion energy radiant energy signal in a coal-fired thermal power unit furnace, characterized in that the method comprises the steps: 1) Multiple flame image detectors are installed at different heights along the furnace to obtain the flame radiation image information in the furnace, and the gray value GSU _i,j of the image is calculated in real time through image processing technology, the expression of which is: GSU _i,j = 0.11R+0.59G+0.23B (1) Among them, i is the number of flame detectors, j is the number of images collected by each detector, and each image is calculated to obtain a gray value; R, G, and B are the information values of the three primary colors of the image, red, green, and blue, respectively, And define the gray value GSU _i,j of the image as the initial radiant energy value; 2) The dynamic range conversion of the initial radiant energy value is performed on the basis of the actual generating power of the unit, according to the formula (2-3): ${\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; m</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; GSU</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$ E _i,M =k _1,i GSU _i,M (3) Among them, k _1,i is the proportional coefficient of the initial radiant energy signal obtained by the i-th detector, M is the length of the data segment for calculating the proportional coefficient, P _j is the actual power of the unit in this data segment, and GSU _i,M is The value of the initial radiant energy signal obtained by the i-th detector at time M, E _i,M is the radiant energy value converted to the same range as the unit load at time M; 3) Screen the initial radiant energy signal that correctly reflects the load change of the unit, according to formula (4-5): |E _i,M -P _M |≤E _thr (4) $\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}}{{\mathrm{<span\; class="notranslate">\; N</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}& <span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; N</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; )</span>\end{array}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$ Among them, P _M is the actual transmission power corresponding to time M, E _thr is the set threshold, N is the total number of detectors, N′ is the number of detectors satisfying the formula (4), and E ₀ is the filtered The mean value of the radiant energy signal that correctly reflects the energy change; 4 ₎ Optimizing the average value E0 of the radiant energy signal of the unit in the stage of severe load changes: ${\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; P</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$ E=k ₂ ·E ₀ (7) Among them, k ₂ is the second correction coefficient, a', b' are coefficients calculated by the least squares method between the mean value of the power change of k _1,i random group and the generated power P, is the average value of power generation, and E is the energy level in the reactor and the final output radiation energy.