Image recognition-based method, apparatus, and system for real-time monitoring of boiler temperature field

Abstract

An image recognition-based method, apparatus, and system for real-time monitoring of a boiler temperature field, the method comprising: acquiring detection data from a boiler, and performing digital processing on the detection data to obtain a boiler combustion image (S101); on the basis of the boiler combustion image, determining three primary color values corresponding to each of a plurality of pixel points (S102); and sequentially inputting the respective three primary color values corresponding to the plurality of pixel points into a temperature measurement grid model to output a temperature value corresponding to each of the plurality of pixel points, and generating a temperature field distribution of the boiler on the basis of the respective temperature values corresponding to the plurality of pixel points (S103).

Description

A method, device, and system for real-time monitoring of boiler temperature field based on image recognition.

[0001] This application claims priority to Chinese Patent Application No. 202411937243.8, filed with the Chinese Patent Office on December 26, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of boiler detection technology, and for example to a method, device and system for real-time monitoring of boiler temperature field based on image recognition. Background Technology

[0003] Combustion temperature within the furnace is one of the key factors affecting the operating efficiency of power plant boilers; therefore, accurate detection of the combustion temperature within the furnace is of great significance. With increasing energy demand and stricter environmental standards, the efficient and stable operation of boilers has become particularly important.

[0004] Real-time monitoring of the combustion temperature field can help optimize the combustion process, reduce fuel consumption, lower harmful gas emissions, and prevent boiler malfunctions due to overheating or uneven temperature. Operating power plant boilers typically use thermocouples placed inside the furnace for temperature measurement, using the temperature information collected by the thermocouples to determine the combustion state within the furnace. However, due to the heat resistance limitations of the sensing element, thermocouples can only perform short-term measurements. When problems such as flame deflection, uneven temperature distribution, or vertical shift of high-temperature zones occur within the furnace, the limited number of thermocouples present results in insufficient temperature information collected, making it difficult to accurately reflect the combustion state and thus failing to depict the overall temperature distribution of the furnace. Summary of the Invention

[0005] This application provides a method, device, and system for real-time monitoring of boiler temperature field based on image recognition, in order to solve the technical problem in related technologies that the amount of temperature information collected is too small, making it difficult to accurately reflect the combustion state inside the furnace and to depict the temperature distribution of the entire furnace.

[0006] This application provides a method for real-time monitoring of boiler temperature field based on image recognition, including:

[0007] Real-time acquisition of detection data in the boiler, and digital processing of the detection data to obtain a boiler combustion image, wherein the boiler combustion image includes multiple pixels;

[0008] Based on the boiler combustion image, determine the three primary color values ​​corresponding to each of the plurality of pixels;

[0009] The three primary color values ​​corresponding to the plurality of pixels are sequentially input into the temperature measurement grid model, and the temperature value corresponding to each of the plurality of pixels is output. The temperature field distribution of the boiler is generated based on the temperature values ​​corresponding to the plurality of pixels.

[0010] This application embodiment also provides a real-time monitoring device for boiler temperature field based on image recognition, including: an acquisition module, a three-primary-color module, and a temperature module;

[0011] The acquisition module is configured to acquire detection data in the boiler in real time and digitize the detection data to obtain a boiler combustion image, wherein the boiler combustion image includes multiple pixels.

[0012] The three-primary-color module is used to determine the three-primary-color value corresponding to each pixel in a plurality of pixels based on the boiler combustion image.

[0013] The temperature module is configured to sequentially input the three primary color values ​​corresponding to the plurality of pixels into the temperature measurement grid model, output the temperature value corresponding to each of the plurality of pixels, and generate the temperature field distribution of the boiler based on the temperature values ​​corresponding to the plurality of pixels.

[0014] This application also provides a method and system for real-time monitoring of boiler temperature field based on image recognition, including: an optical system, an air-cooling system, a combustion state acquisition system, and an image processing system;

[0015] The optical system includes: a housing, an image transmission optical fiber penetrating the front cavity of the housing, and a lens rod; wherein, the housing includes a front cavity and a rear cavity, and the front end of the front cavity is provided with a perforated plate air outlet;

[0016] The air-cooling system includes: a cooler and a cooling air inlet and a cooling air outlet respectively connected to the optical system;

[0017] The combustion state acquisition system is connected to the second end of the lens rod and includes: a camera disposed in the rear cavity of the housing, a color filter disposed in front of the camera, a microcomputer disposed in the rear cavity of the housing, an image acquisition card connected to the microcomputer and disposed on the rear cavity wall of the housing, and a power supply disposed in the rear cavity of the housing.

[0018] The image processing system is configured to execute the image recognition-based real-time monitoring method for boiler temperature field as described above, comprising: an image acquisition card and an image processing unit; the image acquisition card is used for signal conditioning and analog-to-digital signal conversion; the image processing unit is used for image denoising, image enhancement processing, and temperature field calculation.

[0019] This application also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the image recognition-based real-time monitoring method for boiler temperature field as described above.

[0020] This application also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the image recognition-based real-time monitoring method for boiler temperature field as described above. Attached Figure Description

[0021] Figure 1 is a flowchart of a real-time monitoring method for boiler temperature field based on image recognition provided in an embodiment of this application;

[0022] Figure 2 is a structural diagram of a real-time monitoring method and device for boiler temperature field based on image recognition provided in an embodiment of this application;

[0023] Figure 3 is a structural diagram of a real-time monitoring method system for boiler temperature field based on image recognition provided in an embodiment of this application;

[0024] Figure 4 is a schematic diagram of light incident through the collimating lens provided in the embodiment of this application;

[0025] Figure 5 is a spectral response characteristic curve of the dual bandpass filter and CCD camera provided in the embodiments of this application.

[0026] The reference numerals in the accompanying drawings are as follows: 1-First interface; 2-Power switch; 3-Image processing system; 4-Power supply; 5-CCD camera; 6-Color filter; 7-Housing; 8-Cooling air outlet; 9-Porous support frame; 10-Image transmission fiber optic; 11-Flame arrestor plate; 12-High temperature protective material; 13-Colliding lens; 14-Flame; 15-Lens; 16-Optical lens; 17-Lens rod; 18-Cooler; 19-Cooling air inlet; 20-Handheld handle; 21-Dual bandpass filter; 22-Microcomputer; 23-Image acquisition card. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0028] Example 1

[0029] Please refer to Figure 1, which shows a real-time monitoring method for boiler temperature field based on image recognition provided in this application embodiment, including the following steps S101-S103.

[0030] Step S101: Acquire detection data in the boiler in real time, and digitize the detection data to obtain boiler combustion images.

[0031] In this embodiment, the corresponding detection data can be acquired using a charge-coupled device (CCD) camera. Optionally, the CCD camera acquires and outputs a one-dimensional time series signal. Therefore, the detection data of the acquired one-dimensional time series signal is digitized so that the corresponding data is converted into a digital signal to obtain the corresponding boiler combustion image.

[0032] Step S102: Based on the boiler combustion image, determine the three primary color values ​​corresponding to each pixel in the multiple pixels.

[0033] In this embodiment, based on the boiler combustion image, optionally a 24-bit true-color image card is used. The R, G, and B values ​​of each pixel in the boiler combustion image represent the grayscale values ​​of red, green, and blue at the corresponding coordinates, thus obtaining the three primary color values ​​for each pixel. The 24-bit true-color image card can distinguish millions of colors, with a corresponding quantization level of 256 and each color channel being 8 bits.

[0034] Step S103: Input the three primary color values ​​corresponding to multiple pixels into the temperature measurement grid model in sequence, output the temperature value corresponding to each pixel in the multiple pixels, and generate the temperature field distribution of the boiler based on the temperature values ​​corresponding to the multiple pixels.

[0035] In this embodiment, the R, G, and B values ​​obtained from the image processing system are input into the established multilayer perceptron neural network temperature measurement grid model to obtain the temperature field distribution of the power plant boiler furnace cross-section. The temperature measurement grid model network is selected as an SVM (Support Vector Machine) network, with 3 input nodes and 1 output node. Based on the characteristics of the SVM network, a 3-layer SVM network is constructed. The input layer has three terms, corresponding to the R, G, and B color values, and the output layer has one term, namely the two-dimensional temperature T of the flame radiation energy image.

[0036] The technical solution of this application acquires detection data from the boiler in real time, digitizes the corresponding detection data to obtain the corresponding boiler combustion image, and then determines the three primary color values ​​corresponding to each pixel in the image. After the three primary color values ​​corresponding to multiple pixels are sequentially input into the temperature measurement grid model, the temperature values ​​corresponding to multiple pixels can be output. Based on the temperature values, the temperature field distribution of the boiler is generated, thereby accurately and quickly obtaining the distribution of the flame temperature field in the boiler furnace. At the same time, it can detect the temperature field state in the boiler in a relatively timely, comprehensive and accurate manner, avoiding the problem that existing equipment can only perform short-term measurements and cannot accurately reflect the combustion state in the furnace, thus failing to depict the temperature distribution of the entire furnace.

[0037] Example 2

[0038] This embodiment is an optional implementation method in the real-time monitoring method of boiler temperature field based on image recognition.

[0039] As an optional solution in this embodiment, real-time acquisition of detection data in the boiler and digital processing of the detection data to obtain a boiler combustion image include:

[0040] By adjusting the integration time and according to the preset sampling frequency, a one-dimensional time series detection signal is obtained through a lens installed in the boiler observation hole; the one-dimensional time series detection signal is digitized to discretize it into detection data signals; and an image is generated based on the detection digital signals to obtain a boiler combustion image.

[0041] In this embodiment, the entire system starts operating after the portable air compressor, camera power switch, and computer are turned on. The lens installed in the boiler observation hole can be a CCD camera lens, through which detection data information inside the boiler can be acquired; optionally, the CCD camera can be controlled by a software interface program installed on the computer to start acquiring radiation data and radiation image data, and the sampling frequency can be set to 1Hz; wherein, during the sampling process, an appropriate integration time can be adjusted to ensure that the spectral data acquired by the spectrometer and the image data acquired by the CCD camera are both unsaturated and have a high signal-to-noise ratio.

[0042] In this embodiment, optionally, the CCD camera outputs a one-dimensional time series signal, which needs to be digitized. That is, a flame image (i.e., a boiler combustion image) is represented by a finite array or sequence, and then each frame of video image is discretized from a continuous analog signal into a digital signal.

[0043] The detection signal of a one-dimensional time series is digitally processed, including the digitization of amplitude and spatial coordinates.

[0044] As an optional solution in this embodiment, the detection signal of the one-dimensional time series is digitally processed to discretize the detection signal of the one-dimensional time series into detection data signals, including:

[0045] The detection signal of a one-dimensional time series is converted into a frequency domain signal, and features are extracted from the frequency domain signal to obtain the corresponding continuous amplitude features. The corresponding continuous amplitude features are binned to obtain detection data signals involving amplitude magnitude. The cutoff frequency and original distribution of the detection signal of the one-dimensional time series are extracted, and based on the extracted cutoff frequency and original distribution, the detection signal of the one-dimensional time series is converted to obtain the corresponding detection data signals involving spatial coordinates. The detection data signals include detection data signals involving amplitude magnitude and spatial coordinates.

[0046] In this embodiment, image digitization consists of two parts: the digitization of amplitude and the digitization of spatial coordinates. Image sampling is the process of digitizing spatial coordinates (x, y). For example, an image with a cutoff frequency f... xg f yg The original distribution of the flame image is f(x,y), which will be obtained after sampling as f(x,y). g The image of (i,j). It can be understood that image sampling is the process of digitizing continuous image spatial coordinates, that is, converting the continuous spatial coordinates of an image into discrete pixels. For a flame image, its original distribution is f(x,y), and the image distribution obtained after sampling is f(x,y). g (i,j)f g (i,j), where i and j represent the discretized x and y coordinates, respectively.

[0047] In this embodiment, a sampling frequency needs to be set, which determines the image resolution. In two-dimensional images, there are typically two sampling frequencies f. x and f y The sampling intervals correspond to the x and y directions, respectively, and in this embodiment, it is 1Hz. During the sampling process, the continuous coordinates (x, y) of the image are mapped to discrete pixel coordinates (i, j) by dividing the continuous coordinates by the sampling interval and rounding them down. During the sampling process, the continuous grayscale or color values ​​of the image also need to be quantized into finite values, whereby the grayscale or color range of the image is divided into finite levels, and a numerical value is assigned to each level.

[0048] As an optional solution in this embodiment, based on the boiler combustion image, the three primary color values ​​corresponding to each pixel in the plurality of pixels are determined, including:

[0049] Based on a true-color image card, color recognition is performed on the boiler combustion image to obtain the grayscale values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the boiler combustion image, and these grayscale values ​​are used as the three primary color values ​​corresponding to that pixel.

[0050] In this embodiment, the color information in the image can be grayscaled using a 24-bit true-color image card to obtain a grayscale value as the three primary color values. The 24-bit true-color image card can distinguish millions of colors, with a corresponding quantization level of 256. Each color channel is 8 bits, and R, G, B are used to represent the grayscale values ​​of red, green, and blue at the corresponding coordinates.

[0051] In this embodiment, each pixel in the boiler combustion image has its corresponding spatial coordinates, and each spatial coordinate is the coordinate of the pixel. Each pixel also has its corresponding grayscale values ​​of red, green and blue, which are the three primary color values ​​of the pixel.

[0052] As an optional solution in this embodiment, the three primary color values ​​corresponding to multiple pixels are sequentially input into the temperature measurement grid model, and the temperature value corresponding to each pixel is output. Based on the temperature value, the temperature field distribution of the boiler is generated, including:

[0053] The three primary color values ​​corresponding to multiple pixels are used as inputs to the temperature measurement grid model, so that the temperature measurement grid model can calculate the input three primary color values ​​and output the two-dimensional temperature value corresponding to each pixel in the boiler combustion image; based on the two-dimensional temperature values ​​corresponding to multiple pixels and their corresponding spatial coordinates, the temperature field distribution of the boiler is generated.

[0054] In this embodiment, the three primary color values ​​corresponding to multiple pixels are used as inputs to the temperature measurement grid model. That is, the R, G, and B values ​​of each pixel are used as inputs to the temperature measurement grid model. The temperature measurement grid model has three input nodes, so it can directly output the two-dimensional temperature value corresponding to the pixel. After all pixels in the image have output their corresponding two-dimensional temperature values, the distribution map of the two-dimensional temperature values ​​of the entire image can be obtained.

[0055] As an optional solution in this embodiment, the three primary color values ​​corresponding to multiple pixels are sequentially input into the temperature measurement grid model, and the temperature value corresponding to each of the multiple pixels is output. The temperature field distribution of the boiler is then generated based on the temperature values ​​corresponding to the multiple pixels. The method further includes:

[0056] The boiler combustion image is divided into regions based on the number of pixels within each region. The central pixel of each region is then designated as the target pixel. Each region contains an equal number and size of pixels. The primary color values ​​of the target pixels in each region are used as input to a temperature-measuring grid model. This model calculates the two-dimensional temperature value for each target pixel in the boiler combustion image. Edge smoothing is then applied to the target pixels in adjacent regions within each target pixel's region to calculate the two-dimensional temperature values ​​for the remaining pixels in each region. Finally, the boiler's temperature field distribution is generated using the two-dimensional temperature values ​​of all pixels and their corresponding spatial coordinates.

[0057] In this embodiment, since the number of pixels in the boiler combustion image is very large, it would be extremely time-consuming to calculate the corresponding temperature value for each pixel sequentially using a temperature-measuring grid model. Instead, by dividing the pixels in the boiler combustion image into regions—optionally, 3x3 or 5x5 pixels can be considered as one region—the center pixel of each region can be determined as the target pixel. The three primary color values ​​corresponding to the target pixel are then used as inputs to the temperature-measuring grid model, allowing the model to calculate the two-dimensional temperature value corresponding to the center pixel of each region.

[0058] In this embodiment, by performing temperature edge smoothing calculations on each region and its corresponding adjacent regions, the temperature values ​​corresponding to the remaining pixels in each region, excluding the middle pixel, can be directly calculated. This avoids directly inputting a large amount of pixel information into the temperature measurement grid model for calculation, reducing the model's computational load and optimizing computing resources. At the same time, smoothing algorithms such as Gaussian filtering can ensure the accuracy of temperature detection.

[0059] Understandably, for a boiler combustion image, the most important information for detecting the combustion temperature field is obtaining the flame temperature. The temperature abrupt changes within the flame itself during combustion are not significant, so a smoothing algorithm can be directly applied to greatly reduce the computational load. Similarly, the temperature abrupt changes around the flame are also relatively small, thus reducing computational load considerably. Since the temperature data of the flame itself already exists, a clear temperature abrupt change edge exists between the flame and its edges. For example, if the center pixel in one region corresponds to the flame itself with temperature 'a', and the center pixel in an adjacent region corresponds to the edge space with temperature 'b', then determining that the temperature between 'a' and 'b' is greater than a preset value identifies this as a temperature abrupt change edge. This edge can be further calculated by setting appropriate smoothing weights and other coefficients, ultimately reducing the computational burden of calculating large amounts of similar data.

[0060] As an optional solution in this embodiment, the method for constructing the temperature measurement grid model includes:

[0061] Acquire sample data and normalize it; the sample data includes: multiple flame radiation temperatures and the three primary color values ​​corresponding to each flame radiation temperature; set up an initial temperature measurement network model and determine the penalty factor, Gaussian kernel function, and kernel parameters of the initial temperature measurement network model; train the set initial temperature measurement network model based on the sample data to obtain the optimal penalty factor, optimal kernel parameters, and minimum average standard error of the kernel for the corresponding initial temperature measurement network model; update the optimal penalty factor and optimal kernel parameters in the initial temperature measurement network model to train the final temperature measurement mesh model.

[0062] In this embodiment, sample data is acquired and then normalized. This data is then used to train a temperature measurement grid model, which can be an SVM model. An initial SVM model is set, with 3 input nodes and 1 output node. The sample data used for model training is shown in Table 1.

[0063] Table 1

[0064] In this embodiment, the initial SVM model uses a radial basis function (RBF) Gaussian kernel function:

[0065] σ is the standard deviation of the Gaussian function, which affects the width of the kernel function. A larger value of σ results in a wider kernel function and a smoother decision boundary; a smaller value of σ results in a narrower kernel function and a more complex decision boundary. x and x i These are the kernel function values ​​for two pixel data points, respectively.

[0066] In this embodiment, during network training, the penalty factor (c) is set to a range of -10 to 10, and the kernel parameter (g) is set to a range of -5 to 5 for parameter optimization. Here, the values ​​of c and g are logarithmic. Through training, the optimal c value (bestc), optimal g value (bestg), and minimum mean-square error (MSE) of the SVM network parameters are obtained as follows: bestc = 90.5097; bestg = 0.03125; minimum MSE = 214.061.

[0067] In this embodiment, an SVM temperature measurement grid model is established. A large sample database is used to train the temperature measurement grid model. The SVM network has 3 input nodes and 1 output node. Utilizing the characteristics of the SVM network, a 3-layer SVM network is constructed. The input layer contains three terms: R, G, and B color values, and the output layer contains one term: the two-dimensional temperature value T of the flame radiation energy image.

[0068] As another optional embodiment, a multi-dimensional three-layer SVM network can also be constructed, in which each dimension uses an SVM network as a classifier. In this case, the model is a hybrid model, which combines the characteristics of deep learning and SVM, so that the corresponding detected temperature value can be output more accurately through the temperature measurement grid model.

[0069] As an optional solution in this embodiment, obtaining sample data includes:

[0070] The process involves acquiring an image of the flame in a pre-defined black furnace and multiple flame radiation temperatures; performing color recognition on the flame image using a true-color image card to obtain the grayscale values ​​corresponding to the red, green, and blue coordinates of each pixel in the flame image; determining correction coefficients for any two components of the three colors based on the monochromatic emissivity of the pre-defined black furnace, the flame radiation temperature, and the wavelengths corresponding to the red, green, and blue light; determining the monochromatic radiation intensity of the combustion flame spectrum for each color based on the correction coefficients, the wavelengths corresponding to the three colors, and the monochromatic emissivity, thereby obtaining the light intensity signal value at the wavelength corresponding to each color; obtaining the relationship between the flame radiation temperature, the light wavelength, and the monochromatic emissivity of the pre-defined black furnace based on the light intensity signal value; and calibrating the grayscale values ​​corresponding to the three colors in the flame image based on the relationship and the correction coefficients, thereby obtaining the three primary color values ​​corresponding to each flame radiation temperature.

[0071] In this embodiment, by using a black furnace, an image of the furnace flame is acquired and stored using a temperature control system, thereby obtaining multiple sets of sample data, optionally 30 sets. Based on a true-color image card, color recognition is performed on the black furnace flame image, thereby identifying the grayscale values ​​corresponding to the red, green, and blue colors for each pixel coordinate in the black furnace flame image. Based on the preset monochromatic emissivity of the black furnace, the flame radiation temperature, and the wavelengths corresponding to the three colors of light, correction coefficients for any two components of the three colors are determined.

[0072] ε(λ b C1 represents the monochromatic emissivity of the artificial blackbody; C1 = 3.742 × 10⁻⁶. -16 W·m;λ r ,λ g ,λ b The wavelengths represented are 700nm, 546.1nm, and 435.8nm for the R, G, and B primary colors; T is the temperature of the black furnace flame, in K.

[0073] According to Wien's formula:

[0074] E(λ,T) is the monochromatic radiance of the combustion flame spectrum [W / m]. 3 ]; ε(λ) is the flame monochromatic emissivity; λ is the wavelength (m); T is the absolute temperature (K); C1 and C2 are Planck constants; C1 = 3.742 × 10 -16 W·m,C2=hc / k=1.439×10 -2 m·k.

[0075] Taking the light intensity signal values ​​L(λ1,T), L(λ2,T), and L(λ3,T) at three different wavelengths of 700nm, 546.1nm, and 435.8nm, and taking the ratio of two light intensity values ​​at each pair of different wavelengths (optionally, the ratio of light intensity signal values ​​between R and G, and between G and B), we have:

[0076] Therefore, dividing the two equations yields:

[0077] This allows us to obtain the relationship between multiple flame radiation temperatures and the grayscale values ​​of the radiation thermal image output by the color CCD camera. System calibration is then performed, and the calibration results are used to obtain an ideal fitting curve through an appropriate fitting method. The calibrated expression is: R'=R G'=c g ×G B'=c b ×B

[0078] c g c bR, G, and B are the coefficients used to correct components G and B, respectively; R, G, and B are the three primary color values ​​corresponding to any pixel in the flame image acquired by the CCD camera, and R', G', and B' are the results after R, G, and B calibration, respectively.

[0079] As an optional solution in this embodiment, determining the three primary color values ​​corresponding to each pixel in a plurality of pixels based on the boiler combustion image further includes:

[0080] Based on a true-color image card, color recognition is performed on the boiler combustion image to obtain the gray quantization values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the boiler combustion image. According to the relational formula and correction coefficient, the gray quantization values ​​corresponding to the three colors of the coordinates of multiple pixels in the boiler combustion image are calibrated to determine the three primary color values ​​corresponding to each pixel in the multiple pixels.

[0081] In this embodiment, since there is a corresponding calibration expression when establishing the temperature measurement grid model, when obtaining the gray quantization values ​​corresponding to the red, green, and blue coordinates of multiple pixels in the boiler combustion image, a corresponding system calibration can also be performed. That is, according to the corresponding relationship and correction coefficient, the gray quantization values ​​corresponding to R, G, and B of the boiler combustion image are calibrated to obtain the three primary color values ​​that can be input into the temperature measurement grid model after final calibration. This greatly improves the accuracy of temperature detection by acquiring image data and has higher accuracy than direct single model detection.

[0082] The technical solution of this application acquires detection data from the boiler in real time, digitizes the corresponding detection data to obtain the corresponding boiler combustion image, and then determines the three primary color values ​​of each pixel in the image. After the three primary color values ​​of multiple pixels are sequentially input into the temperature measurement grid model, the temperature value corresponding to each pixel can be output. Based on the temperature values ​​corresponding to multiple pixels, the temperature field distribution of the boiler is generated. This allows for accurate and rapid determination of the current flame temperature field distribution in the boiler furnace. It also enables timely, comprehensive, and accurate detection of the temperature field state inside the boiler, avoiding the problems of existing equipment that can only perform short-term measurements and cannot accurately reflect the combustion state inside the furnace, thus failing to depict the temperature distribution of the entire furnace.

[0083] Example 3

[0084] Please refer to Figure 2, which shows the apparatus for real-time monitoring of boiler temperature field based on image recognition provided in this application, including: acquisition module 201, three-primary-color module 202 and temperature module 203;

[0085] The acquisition module 201 is configured to acquire detection data in the boiler in real time and digitize the detection data to obtain a boiler combustion image;

[0086] The three-primary-color module 202 is configured to determine the three-primary-color values ​​corresponding to each pixel in a plurality of pixels based on the boiler combustion image.

[0087] The temperature module 203 is configured to input the three primary color values ​​corresponding to multiple pixels into the temperature measurement grid model in sequence, output the temperature value corresponding to each of the multiple pixels, and generate the temperature field distribution of the boiler based on the temperature values ​​corresponding to the multiple pixels.

[0088] As an optional solution, real-time detection data from the boiler is acquired and digitally processed to obtain boiler combustion images, including:

[0089] By adjusting the integration time using a lens installed in the boiler observation hole and according to the preset sampling frequency, a one-dimensional time series detection signal is obtained.

[0090] The detection signal of a one-dimensional time series is digitally processed to discretize the one-dimensional time series detection signal into a detection data signal;

[0091] An image of boiler combustion is generated based on the detected digital signal.

[0092] As an optional approach, the detection signal of the one-dimensional time series is digitally processed to discretize the one-dimensional time series detection signal into detection data signals, including:

[0093] The detection signal of the one-dimensional time series is converted into a frequency domain signal, and the frequency domain signal is used to extract features to obtain the continuous amplitude features corresponding to the frequency domain signal.

[0094] The frequency domain signal is binned to obtain the detection data signal involving amplitude.

[0095] Extract the cutoff frequency and original distribution of the detection signal from the one-dimensional time series, and based on the extracted cutoff frequency and original distribution, transform the one-dimensional time series detection signal to obtain the corresponding detection data signal involving spatial coordinates.

[0096] The detection data signals include those involving amplitude and spatial coordinates.

[0097] As an optional approach, based on the boiler combustion image, the three primary color values ​​corresponding to each pixel in a plurality of pixels are determined, including:

[0098] Based on a true-color image card, color recognition is performed on boiler combustion images to obtain the grayscale values ​​corresponding to the red, green, and blue coordinates of each pixel in the boiler combustion image. The grayscale values ​​corresponding to the red, green, and blue coordinates of each pixel are then used as the three primary color values ​​corresponding to each pixel.

[0099] As an optional approach, the three primary color values ​​corresponding to multiple pixels are sequentially input into the temperature measurement grid model, and the temperature value corresponding to each of the multiple pixels is output. Based on the temperature values ​​corresponding to the multiple pixels, the temperature field distribution of the boiler is generated, including:

[0100] The three primary color values ​​corresponding to multiple pixels are used as inputs to the temperature measurement grid model, so that the temperature measurement grid model can calculate the input three primary color values ​​and output the two-dimensional temperature value corresponding to each pixel in the multiple pixels of the boiler combustion image.

[0101] The temperature field distribution of the boiler is generated based on the two-dimensional temperature values ​​and their corresponding spatial coordinates of multiple pixels.

[0102] As an optional approach, the three primary color values ​​corresponding to multiple pixels are sequentially input into the temperature measurement grid model, and the temperature value corresponding to each of the multiple pixels is output. Based on the temperature values ​​corresponding to the multiple pixels, the temperature field distribution of the boiler is generated, which also includes:

[0103] The boiler combustion image is divided into regions based on the number of pixels in each region. The middle pixel of each region is then determined and used as the target pixel. Each region has the same number and size of pixels.

[0104] The three primary color values ​​corresponding to the target pixels in all regions are used as inputs to the temperature measurement grid model, so that the temperature measurement grid model can calculate the input three primary color values ​​and output the two-dimensional temperature value corresponding to each target pixel in all regions of the boiler combustion image.

[0105] Based on the region where each target pixel is located in all regions, edge smoothing is performed on the target pixels in its adjacent regions, thereby calculating the two-dimensional temperature value corresponding to the remaining pixels in each region of all regions.

[0106] The temperature field distribution of the boiler is generated by using the two-dimensional temperature values ​​of all pixels and their corresponding spatial coordinates.

[0107] As an optional approach, methods for constructing the temperature-sensing grid model include:

[0108] Acquire sample data and normalize the sample data; the sample data includes: multiple flame radiation temperatures and the three primary color values ​​corresponding to each flame radiation temperature.

[0109] Set up an initial temperature measurement network model and determine the penalty factor, Gaussian kernel function, and kernel parameters of the initial temperature measurement network model;

[0110] The initial temperature measurement network model is trained based on sample data to obtain the optimal penalty factor, optimal kernel parameter, and kernel minimum average standard error of the corresponding initial temperature measurement network model.

[0111] The optimal penalty factor and the optimal kernel parameter kernel minimum average standard error are updated and set in the initial temperature measurement network model, thereby training the final temperature measurement grid model.

[0112] As an optional solution, sample data can be obtained, including:

[0113] Acquire images of the flames in a preset black furnace and multiple flame radiation temperatures;

[0114] Based on a true-color image card, color recognition is performed on the image of a black furnace flame to obtain the gray quantization values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the image of the black furnace flame.

[0115] Based on the monochromatic emissivity of the preset black furnace, the radiation temperatures of multiple flames, and the wavelengths corresponding to red, green, and blue light, the correction coefficients for any two components of the three colors are determined at each flame radiation temperature.

[0116] Based on the correction coefficient and the wavelengths corresponding to the three colors of light, combined with the monochromatic emissivity, the monochromatic radiation intensity of the combustion flame spectrum of each color is determined, thereby obtaining the light intensity signal value at the wavelength corresponding to each color of light.

[0117] Based on the light intensity signal values ​​of the red, green and blue light respectively, the relationship between the corresponding temperature and the light wavelength and the monochromatic emissivity of the preset black furnace is obtained. According to the relationship and the correction coefficient, the gray quantization values ​​corresponding to the three colors of the black furnace flame image are calibrated, thereby obtaining the three primary color values ​​corresponding to the radiation temperature of each flame.

[0118] As an optional approach, determining the three primary color values ​​corresponding to each pixel in a multi-pixel array based on the boiler combustion image also includes:

[0119] Based on a true-color image card, color recognition is performed on boiler combustion images to obtain the gray quantization values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the boiler combustion image.

[0120] Based on the formula and correction coefficient, the gray quantization values ​​corresponding to the red, green and blue colors of the coordinates of multiple pixels in the boiler combustion image are calibrated, thereby determining the three primary color values ​​corresponding to each pixel in the multiple pixels.

[0121] Those skilled in the art will understand that, for the sake of convenience and brevity, the working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0122] The technical solution of this application acquires detection data from the boiler, digitizes the data to obtain a boiler combustion image, and then determines the three primary color values ​​corresponding to each pixel in the image. These values ​​are then sequentially input into a temperature measurement grid model, outputting the temperature value corresponding to each pixel. Based on these temperature values, the temperature field distribution of the boiler is generated, enabling accurate and rapid determination of the flame temperature field distribution within the boiler furnace. This also allows for timely, comprehensive, and accurate detection of the boiler's internal temperature field, avoiding the limitations of existing equipment that can only perform short-term measurements and accurately reflect the combustion state within the furnace, thus failing to depict the overall temperature distribution of the furnace.

[0123] Example 4

[0124] Please refer to Figure 3, which also provides a boiler temperature field real-time monitoring method system based on image recognition, including: an optical system, an air-cooling system, a combustion state acquisition system, and an image processing system.

[0125] The optical system includes: a housing 7, an image transmission fiber 10 penetrating the front cavity of the housing 7, and a mirror rod 17; wherein, the housing 7 includes a front cavity and a rear cavity, and the front end of the front cavity is provided with a perforated plate air outlet.

[0126] As an optional solution in this embodiment, please refer to Figure 4. The first end of the image transmission fiber 10 is connected to a collimating lens 13, the second end of the image transmission fiber 10 is connected to a color filter 6, and the image transmission fiber 10 is wrapped with a high-temperature resistant protective material 12. A lens 15 is provided at the front end of the lens rod 17, and a plurality of optical lenses 16 and a dual bandpass filter 21 are arranged at intervals inside the lens rod 17.

[0127] In this embodiment, the number of optical lenses 16 can be set according to actual needs. Optionally, the number of optical lenses 16 is greater than or equal to 3.

[0128] As an optional solution in this embodiment, a plurality of porous support frames 9 are provided in the front cavity of the housing 7 along the direction of the first end and the second end of the front cavity. The porous support frames 9 are configured to support the image transmission fiber 10 and the mirror rod 17.

[0129] As an optional solution in this embodiment, a fire-resistant plate 11 is also provided on the outer side of the front cavity of the housing 7, and a hand handle 20 is also provided at the second end position on the outer side of the front cavity of the housing 7.

[0130] In this embodiment, the flame arrestor plate can prevent the high-temperature gas or coal ash ejected from the fire observation hole from burning the operator.

[0131] The air-cooling system includes a cooler 18 and a cooling air inlet 19 and a cooling air outlet 8, which are respectively connected to the optical system.

[0132] The combustion state acquisition system is connected to the second end of the lens rod 17 and includes: a camera installed in the rear cavity of the housing 7, a color filter 6 installed in front of the camera, a microcomputer 22 installed in the rear cavity of the housing 7, an image acquisition card 23 connected to the microcomputer 22 and installed on the rear cavity wall of the housing 7, and a power supply 4 installed in the rear cavity of the housing 7.

[0133] As an optional solution in this embodiment, the power supply 4 is connected to the camera and the microcomputer 22 via the power switch 2.

[0134] Optionally, the camera can be a CCD camera, which can directly acquire images of the flame 14. Please refer to Figure 5, which is a graph of the spectral response characteristics of the dual bandpass filter and the CCD camera.

[0135] Image processing system 3 is configured to perform any of the above-mentioned boiler temperature field real-time monitoring methods, including: image acquisition card 23 and image processing unit; image acquisition card 23 is used for signal conditioning and analog-to-digital signal conversion; image processing unit is used for image denoising, image enhancement processing, and temperature field calculation.

[0136] In this embodiment, the image processing system 3 includes an image acquisition card 23 and an image processing unit; the image processing unit includes functions such as image denoising, image enhancement processing, and temperature field calculation.

[0137] In this embodiment, the image processing system 3 is connected to the image acquisition card 23 through the first interface 1.

[0138] In this embodiment, in the image processing system, the CCD camera 5 has a 24-bit true color image card and adopts the color CCD three-color temperature measurement method. R, G, and B represent the gray values ​​of red, green, and blue under the corresponding coordinates. Then, the system is calibrated to establish the relationship between the flame radiation temperature and the gray value of the radiation heat image output by the color CCD camera 5, and finally obtains the R, G, and B three-color values.

[0139] In this embodiment, the R, G, B values ​​obtained by the image processing system are input into the established multilayer perceptron neural network to obtain the temperature field distribution of the furnace section of the power plant boiler.

[0140] Optionally, the first interface 1 is a gigabit network interface, which can be connected to a computer.

[0141] Optionally, the image transmission fiber 10 is an armored fiber, the lens 15 is a pinhole lens, and the power supply 4 is a lithium battery.

[0142] In this embodiment, compressed air introduced into the boiler enters the device through cooling air inlet 19. The cooling air then splits into two paths: one flows out through cooling air outlet 8, and the other through perforated plate outlet 2. The perforated plate outlet 2 cools the lens 10 and collimating lens 6, and the cooling air also cleans and protects them after entering the device. The collimating lens 6 only allows parallel light directly in front of it to enter, as shown in Figure 4. The radiation from the pulverized coal combustion flame in the boiler furnace enters the collimating lens 6 and then travels along the image transmission fiber 10 to the lens 15. After passing through a lens group composed of multiple optical lenses 16, it is incident on the dual bandpass filter 21, where the image is captured by the CCD camera 5. The color CCD camera decomposes the incident light into red (R), green (G), and blue (B) images based on their wavelengths, which are 700 nm, 546.1 nm, and 435.8 nm, respectively. The porous structure of the porous support frame 17 also allows for cooling airflow. Control commands and data transmission for the CCD camera 5 are conducted via a gigabit network cable. The gigabit network interface connects to the subsequent image processing system 3, serving both power supply and data transmission control functions. The CCD camera 5 is powered by a built-in lithium battery, and the power can be turned on or off via the power switch 2.

[0143] It is understandable that existing detection methods can also reconstruct the furnace temperature field distribution using acoustic wave methods or laser absorption spectroscopy, but both methods have certain technical drawbacks. Acoustic wave methods require the installation of a certain number of transmitters and receivers on the furnace, which increases construction difficulty and hardware costs. Furthermore, factors such as on-site vibrations can affect the stability of system operation. Additionally, the large size of the acoustic wave transmitters and the large openings in the furnace wall can also affect the temperature distribution within the furnace to some extent. Laser absorption spectroscopy is costly and requires relatively high operational expertise, making it inconvenient for power plant personnel. Therefore, the embodiment of this application is simple in construction, low in cost, highly stable, and easy to operate. This device and detection method enable wireless image recognition-based real-time monitoring of the boiler temperature field, allowing for relatively accurate and comprehensive detection of the temperature state within the boiler.

[0144] The optical system of this embodiment includes a housing, an image transmission fiber and a lens rod penetrating the front cavity, a collimating lens connected to the first end of the image transmission fiber, a lens disposed at the front end of the lens rod, a plurality of optical lenses spaced apart from the first end to the second end of the lens rod inside the lens rod, and a dual bandpass filter; the air-cooling system includes an air cooler, a cooling air inlet connected to the optical device, and a cooling air outlet; the combustion state acquisition system includes: a CCD camera connected to the second end of the lens rod and disposed in the rear cavity; a color filter placed in front of the CCD camera; a microcomputer disposed in the rear cavity; a first interface connected to the microcomputer for transmitting spectral data and power supply; an image acquisition card disposed on the rear cavity wall; and a power supply disposed in the rear cavity connecting the CCD camera and the microcomputer; the image processing system includes an image acquisition card and an image processing unit. Through detailed analysis of the flame image, a temperature distribution map of the boiler combustion zone can be obtained, capturing minute details of temperature changes, thereby providing a more comprehensive understanding of the combustion process. This device is simple in structure, low in cost, highly stable, and easy to operate. With this device and the corresponding detection method, a wireless image recognition-based real-time monitoring method for boiler temperature field can be realized, which can accurately and comprehensively detect the temperature status inside the boiler.

[0145] Example 5

[0146] This application also provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the image recognition-based real-time monitoring method for boiler temperature field as described in any of the above embodiments.

[0147] The terminal device of this embodiment includes a processor, a memory, and a computer program and computer instructions stored in the memory and executable on the processor. When the processor executes the computer program, it implements multiple steps as described in Embodiment 1 above, such as steps S101 to S103 shown in FIG1. ​​Alternatively, when the processor executes the computer program, it implements the functions of multiple modules / units in the above device embodiment, such as temperature module 203.

[0148] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete this application. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device. For example, the temperature module 203 is configured to sequentially input the three primary color values ​​corresponding to multiple pixels into a temperature measurement grid model, output the temperature value corresponding to each pixel, and generate the temperature field distribution of the boiler based on the temperature values ​​corresponding to the multiple pixels.

[0149] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the schematic diagram is merely an example of a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0150] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting multiple parts of the terminal device via various interfaces and lines.

[0151] The memory can be configured to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc.; the data storage area can store data created based on the use of the mobile terminal, etc. In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0152] If the modules / units integrated in the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the above multiple method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The content contained in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0153] Example 6

[0154] This application also provides a computer-readable storage medium, which includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the image recognition-based real-time monitoring method for boiler temperature field as described in any of the above embodiments.

Claims

1. A method for real-time monitoring of boiler temperature field based on image recognition, comprising: Real-time acquisition of detection data in the boiler, and digital processing of the detection data to obtain a boiler combustion image, wherein the boiler combustion image includes multiple pixels; Based on the boiler combustion image, determine the three primary color values ​​corresponding to each of the plurality of pixels; The three primary color values ​​corresponding to the plurality of pixels are sequentially input into the temperature measurement grid model, and the temperature value corresponding to each of the plurality of pixels is output. The temperature field distribution of the boiler is generated based on the temperature values ​​corresponding to the plurality of pixels.

2. The method as described in claim 1, wherein, The real-time acquisition of detection data in the boiler and the digitization of the detection data to obtain a boiler combustion image include: By adjusting the integration time using a lens installed in the boiler observation hole and according to the preset sampling frequency, a one-dimensional time series detection signal is obtained. The detection signal of the one-dimensional time series is digitally processed to discretize the detection signal of the one-dimensional time series into detection data signals; An image is generated based on the detected digital signal to obtain a boiler combustion image.

3. The method as described in claim 2, wherein, The step of digitizing the detection signal of the one-dimensional time series to discretize it into detection data signals includes: The detection signal of the one-dimensional time series is converted into a frequency domain signal, and the frequency domain signal is subjected to feature extraction to obtain the continuous amplitude features corresponding to the frequency domain signal. The frequency domain signal is binned to obtain a detection data signal involving amplitude magnitude; Extract the cutoff frequency and original distribution of the detection signal of the one-dimensional time series, and transform the detection signal of the one-dimensional time series according to the extracted cutoff frequency and original distribution to obtain the detection data signal involving spatial coordinates. The detection data signal includes detection data signals involving amplitude and spatial coordinates.

4. The method according to any one of claims 1-3, wherein, The step of determining the three primary color values ​​corresponding to each pixel among the plurality of pixels based on the boiler combustion image includes: Based on a true-color image card, color recognition is performed on the boiler combustion image to obtain the grayscale values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the boiler combustion image. The grayscale values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel are then used as the three primary color values ​​corresponding to each pixel.

5. The method of claim 1, wherein, The step of sequentially inputting the three primary color values ​​corresponding to the plurality of pixels into the temperature measurement grid model, outputting the temperature value corresponding to each of the plurality of pixels, and generating the temperature field distribution of the boiler based on the temperature values ​​corresponding to the plurality of pixels includes: The three primary color values ​​corresponding to the plurality of pixels are respectively used as inputs to the temperature measuring grid model, so that the temperature measuring grid model calculates the input three primary color values ​​and outputs the two-dimensional temperature value corresponding to each of the plurality of pixels in the boiler combustion image. The temperature field distribution of the boiler is generated based on the two-dimensional temperature values ​​and their corresponding spatial coordinates of the multiple pixels.

6. The method of claim 1, wherein, The step of sequentially inputting the three primary color values ​​corresponding to the plurality of pixels into the temperature measuring grid model, outputting the temperature value corresponding to each of the plurality of pixels, and generating the temperature field distribution of the boiler based on the temperature values ​​corresponding to the plurality of pixels, further includes: The boiler combustion image is divided into regions based on the number of pixels in each region. The middle pixel of each region is determined according to the region where the pixels are located, and the middle pixel is used as the target pixel. The number and size of pixels in each region are equal. The three primary color values ​​corresponding to the target pixels in all regions are used as inputs to the temperature measurement grid model, so that the temperature measurement grid model calculates the input three primary color values ​​and outputs the two-dimensional temperature value corresponding to each target pixel in all regions of the boiler combustion image. Based on the region where each target pixel is located in all regions, edge smoothing is performed on the target pixels in its adjacent regions, thereby calculating the two-dimensional temperature value corresponding to the remaining pixels in each region of all regions. The temperature field distribution of the boiler is generated by using the two-dimensional temperature values ​​of all pixels and their corresponding spatial coordinates.

7. The method as described in any one of claims 1, 2, 3, 5, or 6, wherein, The method for constructing the temperature measurement grid model includes: Acquire sample data and normalize the sample data; wherein, the sample data includes: multiple flame radiation temperatures and the three primary color values ​​corresponding to the multiple flame radiation temperatures; Set up an initial temperature measurement network model and determine the penalty factor, Gaussian kernel function, and kernel parameters of the initial temperature measurement network model; The initial temperature measurement network model is trained based on the sample data to obtain the optimal penalty factor, optimal kernel parameter, and minimum average standard error of the corresponding initial temperature measurement network model. The optimal penalty factor and the optimal kernel parameter kernel minimum average standard error are updated and set in the initial temperature measurement network model, thereby training the final temperature measurement grid model.

8. The method of claim 7, wherein, The acquisition of sample data includes: Acquire images of the flames in a preset black furnace and multiple flame radiation temperatures; Based on a true-color image card, color recognition is performed on the black furnace flame image to obtain the gray quantization values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the black furnace flame image. Based on the monochromatic emissivity of the preset black furnace, the radiation temperatures of the multiple flames, and the wavelengths corresponding to the red, green, and blue light, the correction coefficients for any two components of the three colors are determined at each flame radiation temperature. Based on the correction coefficient and the wavelengths corresponding to the red, green and blue lights respectively, and combined with the monochromatic emissivity, the monochromatic radiation intensity of the combustion flame spectrum of each color is determined, thereby obtaining the light intensity signal value at the wavelength corresponding to each color of light. Based on the light intensity signal values ​​of the red, green and blue light respectively, the relationship between the corresponding flame radiation temperature and the light wavelength and the monochromatic emissivity of the preset black furnace is obtained. According to the relationship and the correction coefficient, the gray quantization values ​​corresponding to the three colors of the black furnace flame image are calibrated to obtain the three primary color values ​​corresponding to each flame radiation temperature.

9. The method of claim 8, wherein, The step of determining the three primary color values ​​corresponding to each pixel among the plurality of pixels based on the boiler combustion image further includes: Based on a true-color image card, color recognition is performed on the boiler combustion image to obtain the gray quantization values ​​corresponding to the red, green, and blue colors of the coordinates of each pixel in the boiler combustion image. Based on the relationship and the correction coefficient, the gray quantization values ​​corresponding to the red, green and blue colors of the coordinates of the plurality of pixels in the boiler combustion image are calibrated, thereby determining the three primary color values ​​corresponding to each of the plurality of pixels.

10. A method and apparatus for real-time monitoring of boiler temperature field based on image recognition, comprising: Acquisition module, three primary color module, and temperature module; The acquisition module is configured to acquire detection data in the boiler in real time and digitize the detection data to obtain a boiler combustion image, wherein the boiler combustion image includes multiple pixels. The three-primary-color module is configured to determine the three-primary-color value corresponding to each pixel among the plurality of pixels based on the boiler combustion image; The temperature module is configured to sequentially input the three primary color values ​​corresponding to the plurality of pixels into the temperature measurement grid model, output the temperature value corresponding to each of the plurality of pixels, and generate the temperature field distribution of the boiler based on the temperature values ​​corresponding to the plurality of pixels.

11. A method and system for real-time monitoring of boiler temperature field based on image recognition, comprising: Optical system, air-cooling system, combustion status acquisition system, and image processing system; The optical system includes: a housing, an image transmission optical fiber penetrating the front cavity of the housing, and a lens rod; wherein, the housing includes a front cavity and a rear cavity, and the front end of the front cavity is provided with a perforated plate air outlet; The air-cooling system includes: a cooler and a cooling air inlet and a cooling air outlet respectively connected to the optical system; The combustion state acquisition system is connected to the second end of the lens rod and includes: a camera disposed in the rear cavity of the housing, a color filter disposed in front of the camera, a microcomputer disposed in the rear cavity of the housing, an image acquisition card connected to the microcomputer and disposed on the rear cavity wall of the housing, and a power supply disposed in the rear cavity of the housing. The image processing system is configured to execute the real-time monitoring method for boiler temperature field based on image recognition as described in any one of claims 1-9, comprising: an image acquisition card and an image processing unit; the image acquisition card is used for signal conditioning and analog-to-digital signal conversion; the image processing unit is used for image denoising, image enhancement processing, and temperature field calculation.

12. The system of claim 11, wherein, The first end of the image transmission fiber is connected to a collimating lens, the second end of the image transmission fiber is connected to a color filter, and the image transmission fiber is wrapped with a high-temperature resistant protective material. A lens is provided at the front end of the lens rod, and multiple optical lenses and a dual bandpass filter are arranged at intervals inside the lens rod.

13. The legal system as claimed in claim 11, wherein, The power supply is connected to the camera and the microcomputer via a power switch.

14. The system according to any one of claims 11-13, wherein, Multiple porous support frames are provided in the front cavity of the housing along the direction of the first end and the second end of the front cavity. The porous support frames are configured to support the image transmission optical fiber and the mirror rod.

15. The system according to any one of claims 11-13, wherein, A fire-resistant plate is provided on the outer side of the front cavity of the housing, and a hand handle is provided at the second end of the outer side of the front cavity of the housing.

16. A terminal device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the image recognition-based real-time monitoring method for boiler temperature field as described in any one of claims 1 to 9.

17. A computer-readable storage medium comprising a stored computer program, wherein, When the computer program is running, it controls the device containing the computer-readable storage medium to perform the real-time monitoring method for boiler temperature field based on image recognition as described in any one of claims 1 to 9.

Images