Boiler temperature field real-time monitoring method, device and system based on image recognition

Abstract

The invention discloses a boiler temperature field real-time monitoring method, device and system based on image recognition, and the method comprises the steps: obtaining detection data in a boiler, and carrying out the digital processing of the detection data, and obtaining a boiler combustion image; according to the boiler combustion image, determining a three-primary-color value corresponding to any pixel point; and inputting each three-primary-color value into a temperature measurement grid model in sequence, outputting and obtaining a temperature value corresponding to each pixel point, and generating temperature field distribution of the boiler based on the temperature values. The technical problems that in the prior art, the amount of collected temperature information is too small, the combustion state in the furnace is difficult to accurately reflect, and the temperature distribution of the whole hearth cannot be described are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler detection, and in particular to a method, device and system for real-time monitoring of a boiler temperature field based on image recognition. Background Art

[0002] Measuring the combustion temperature inside the furnace is one of the key factors affecting the operating efficiency of power plant boilers, so accurately detecting the combustion temperature inside the furnace is of great significance. With the increase in energy demand and the improvement of environmental protection standards, the efficient and stable operation of boilers has become more and more important.

[0003] Real-time monitoring of the combustion temperature field can help optimize the combustion process, reduce fuel consumption, lower harmful gas emissions, and prevent boiler failures due to overheating or uneven temperatures. Currently 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 status within the furnace. However, due to the heat resistance limitations of the temperature sensing element, thermocouples can only measure for a short period of time. When problems such as flame deflection, uneven temperature distribution, and upward and downward shifts in the high-temperature zone occur within the furnace, the limited number of thermocouples placed inside the furnace will result in insufficient temperature information being collected, making it difficult to accurately reflect the combustion status within the furnace and, consequently, unable to depict the temperature distribution of the entire furnace. Summary of the Invention

[0004] The present invention provides a method, device and system for real-time monitoring of boiler temperature field based on image recognition, so as to solve the technical problem in the prior art that the amount of temperature information collected is too small, making it difficult to accurately reflect the combustion state in the furnace and unable to depict the temperature distribution of the entire furnace.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for real-time monitoring of boiler temperature field based on image recognition, comprising:

[0006] Acquire detection data from the boiler in real time, and digitally process the detection data to obtain a boiler combustion image;

[0007] Determining the three primary color values ​​corresponding to any pixel point according to the boiler combustion image;

[0008] Each of the three primary color values ​​is sequentially input into the temperature measurement grid model, and the temperature value corresponding to each pixel point is output, and the temperature field distribution of the boiler is generated based on the temperature value.

[0009] As a preferred solution, the real-time acquisition of detection data from the boiler and digital processing of the detection data to obtain a boiler combustion image specifically includes:

[0010] Through the lens set in the boiler fire viewing hole, the integration time is adjusted, and the detection signal of the one-dimensional time series is obtained according to the preset sampling frequency;

[0011] Digitally processing the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal;

[0012] An image is generated according to the detection digital signal to obtain a boiler combustion image.

[0013] As a preferred solution, the digital processing of the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal specifically includes:

[0014] Convert the detection signal of the one-dimensional time series into a frequency domain signal, and extract features of the converted frequency domain signal to obtain corresponding amplitude features;

[0015] The corresponding continuous amplitude features are binned to obtain detection data signals involving amplitude size;

[0016] Extracting the cutoff frequency and the original distribution of the detection signal of the one-dimensional time series, and converting the extracted cutoff frequency and the original distribution to obtain a detection data signal corresponding to the spatial coordinates;

[0017] The detection data signal includes a signal involving amplitude and spatial coordinates.

[0018] As a preferred solution, determining the three primary color values ​​corresponding to any pixel point according to the boiler combustion image specifically includes:

[0019] Based on the true color image card, the boiler combustion image is color recognized to obtain the grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the boiler combustion image as the three primary color values ​​corresponding to any pixel.

[0020] As a preferred solution, the step of sequentially inputting each of the three primary color values ​​into a temperature measurement grid model, outputting a temperature value corresponding to each pixel point, and generating a temperature field distribution of the boiler based on the temperature value specifically includes:

[0021] Using each of the three primary color values ​​as an input to a temperature measurement grid model, so that the temperature measurement grid model calculates the input three primary color values ​​and outputs a two-dimensional temperature value corresponding to each pixel in the boiler combustion image;

[0022] Based on the two-dimensional temperature values ​​and their corresponding spatial coordinates, a temperature field distribution of the boiler is generated.

[0023] As a preferred solution, the method of sequentially inputting each of the three primary color values ​​into a temperature measurement grid model, outputting a temperature value corresponding to each pixel point, and generating a temperature field distribution of the boiler based on the temperature value further includes:

[0024] Dividing the pixels in the boiler combustion image into regions, and determining the middle pixel of each region as the target pixel according to the regions of the divided pixels; wherein the number and size of the pixels in each region are equal;

[0025] The three primary color values ​​corresponding to the target pixel points are used as inputs of a temperature measurement grid model, so that the temperature measurement grid model calculates the input three primary color values ​​and outputs a two-dimensional temperature value corresponding to each target pixel point in the boiler combustion image;

[0026] Based on the area where each target pixel point is located, edge smoothing is performed on the target pixel points in the adjacent areas respectively, so as to calculate the two-dimensional temperature values ​​corresponding to the remaining pixel points in each area;

[0027] The temperature field distribution of the boiler is generated through the two-dimensional temperature values ​​of all pixel points and their corresponding spatial coordinates.

[0028] As a preferred solution, the method for constructing the temperature measurement grid model includes:

[0029] Acquire sample data and perform normalization processing on the sample data; wherein the sample data includes: a plurality of groups of flame radiation temperatures and three primary color values ​​corresponding to each flame radiation temperature;

[0030] Setting an initial temperature measurement network model, and determining a penalty factor, a Gaussian kernel function, and kernel parameters of the initial temperature measurement network model;

[0031] The set initial temperature measurement network model is trained based on the sample data, thereby obtaining the optimal penalty factor, the optimal kernel parameter kernel minimum mean standard error corresponding to the initial temperature measurement network model;

[0032] The optimal penalty factor, the optimal kernel parameter kernel minimum mean standard error are updated and set in the initial temperature measurement network model, so as to train and obtain the final temperature measurement grid model.

[0033] As a preferred solution, obtaining sample data specifically includes:

[0034] Obtaining a black furnace flame image and each group of flame radiation temperatures in a preset black furnace;

[0035] Based on a true color image card, color recognition is performed on the black furnace flame image to obtain grayscale quantization values ​​corresponding to red, green, and blue colors of each pixel coordinate in the black furnace flame image;

[0036] Determining correction coefficients for any two components of the three colors based on the monochromatic emissivity of the preset black furnace, the flame radiation temperature, and the wavelengths corresponding to the three colors of light;

[0037] According to the correction coefficient and the wavelengths of light corresponding to the three colors, combined with the monochromatic radiation rate of the flame, 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 light;

[0038] Based on the light intensity signal value, a relationship between the corresponding temperature, light wavelength and the monochromatic emissivity of the preset black furnace is obtained, and according to the relationship and the correction coefficient, the grayscale 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.

[0039] As a preferred solution, determining the three primary color values ​​corresponding to any pixel point according to the boiler combustion image specifically further includes:

[0040] Based on a true color image card, color recognition is performed on the boiler combustion image to obtain grayscale quantization values ​​corresponding to red, green, and blue colors of each pixel coordinate in the boiler combustion image;

[0041] According to the relationship and the correction coefficient, the grayscale quantization values ​​corresponding to the three colors of the boiler combustion image are calibrated, so as to determine the three primary color values ​​corresponding to any pixel point.

[0042] Accordingly, the present invention also provides a real-time monitoring device for boiler temperature field based on image recognition, comprising: an acquisition module, a three-primary color module and a temperature module;

[0043] The acquisition module is used to acquire detection data from the boiler in real time and digitally process the detection data to obtain a boiler combustion image;

[0044] The three-primary color module is used to determine the three-primary color value corresponding to any pixel point according to the boiler combustion image;

[0045] The temperature module is used to input each of the three primary color values ​​into the temperature measurement grid model in sequence, output the temperature value corresponding to each pixel point, and generate the temperature field distribution of the boiler based on the temperature value.

[0046] Accordingly, the present invention also provides a method system for real-time monitoring of boiler temperature field based on image recognition, comprising: an optical system, an air cooling system, a combustion state acquisition system and an image processing system;

[0047] The optical system comprises: a housing, an image transmission optical fiber and a mirror rod passing through a front cavity of the housing; wherein the housing comprises a front cavity and a rear cavity, and a front end of the front cavity is provided with a porous plate air outlet;

[0048] The air cooling system includes: a cooler and a cooling air inlet and a cooling air outlet respectively connected to the optical system;

[0049] The combustion state acquisition module is connected to the second end of the mirror rod and includes: a camera disposed in the rear end cavity of the housing, a color filter disposed in front of the camera, a microcomputer disposed in the rear end 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 end cavity of the housing;

[0050] The image processing system is used to execute the real-time monitoring method of the boiler temperature field based on image recognition as described in any one of the above items, including: 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.

[0051] As a preferred solution, 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;

[0052] A lens is arranged at the front end of the mirror rod, and a plurality of optical lenses and a dual-bandpass filter are arranged at intervals inside the mirror rod.

[0053] As a preferred solution, the power supply is connected to the camera and the microcomputer via a power switch.

[0054] As a preferred solution, a plurality of porous support frames are provided in the front end cavity of the shell along the first end and the second end direction of the front end cavity, and the porous support frames are used to support the image transmission optical fiber and the mirror rod.

[0055] As a preferred solution, a fire-blocking plate is further provided on the outside of the front end cavity of the shell, and a hand-held handle is further provided on the second end position of the outside of the front end cavity of the shell.

[0056] Accordingly, the present invention also provides 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 when the processor executes the computer program, it implements the real-time monitoring method of the boiler temperature field based on image recognition as described in any one of the above.

[0057] Accordingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the real-time monitoring method of the boiler temperature field based on image recognition as described in any one of the above.

[0058] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0059] The technical solution of the present invention obtains the detection data in the boiler in real time, and then digitizes the corresponding detection data to obtain the corresponding boiler combustion image, and then determines the three primary color values ​​of any point in the image, so that after each three primary color value is input into the temperature measurement grid model in turn, the temperature value corresponding to each pixel point can be output, and the temperature field distribution of the boiler is generated based on the temperature value, so that the distribution of the flame temperature field in the current boiler furnace can be obtained accurately and quickly, and the temperature field state in the boiler can also be detected more timely, comprehensively and accurately, avoiding the problem that the existing equipment can only perform short-term measurements and is difficult to accurately reflect the combustion state in the furnace, thereby being unable to depict the temperature distribution of the entire furnace. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0061] Figure 1 : A flow chart of a method for real-time monitoring of boiler temperature field based on image recognition provided by an embodiment of the present invention;

[0062] Figure 2 : A structural diagram of a method and device for real-time monitoring of boiler temperature field based on image recognition provided by an embodiment of the present invention;

[0063] Figure 3 : A structural diagram of a method system for real-time monitoring of boiler temperature field based on image recognition provided by an embodiment of the present invention;

[0064] Figure 4: Schematic diagram of light incident on the collimating lens provided by an embodiment of the present invention;

[0065] Figure 5 : Spectral response characteristic curve diagram of the dual-bandpass filter and CCD camera provided in an embodiment of the present invention;

[0066] The accompanying drawings in the specification are numerals as follows:

[0067] 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 optical fiber; 11-fire barrier; 12-high-temperature protective material; 13-collimating lens; 14-flame; 15-lens; 16-optical lens; 17-mirror rod; 18-cooler; 19-cooling air inlet; 20-handle; 21-dual-bandpass filter; 22-microcomputer; 23-image acquisition card. DETAILED DESCRIPTION

[0068] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0069] Example 1

[0070] Please refer to Figure 1 , a method for real-time monitoring of boiler temperature field based on image recognition provided by an embodiment of the present invention, comprising the following steps S101-S103:

[0071] Step S101: Acquire detection data from the boiler in real time, and digitally process the detection data to obtain a boiler combustion image.

[0072] In this embodiment, the corresponding detection data can be obtained through a CCD camera. Preferably, the CCD camera collects and outputs a one-dimensional time series signal. Therefore, the detection data of the collected 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.

[0073] Step S102: Determine the three primary color values ​​corresponding to any pixel point according to the boiler combustion image.

[0074] In this embodiment, based on the boiler combustion image, a 24-bit true color image card is preferably used to represent the grayscale quantization values ​​of red, green, and blue at each coordinate for each pixel in the boiler combustion image. This results in the corresponding three primary color values ​​for each pixel. A 24-bit true color image card can distinguish millions of colors, corresponding to 256 quantization levels and 8 bits per color channel.

[0075] Step S103: Input each of the three primary color values ​​into the temperature measurement grid model in sequence, output the temperature value corresponding to each pixel point, and generate the temperature field distribution of the boiler based on the temperature value.

[0076] In this example, the R, G, and B values ​​obtained by the image processing system were incorporated into a temperature measurement grid model based on a multi-layer perceptron neural network to determine the temperature distribution of the power plant boiler furnace cross section. It should be noted that the network selected for the temperature measurement grid model is an SVM (Support Vector Machine) network, with three input nodes and one output node. Based on the characteristics of SVM networks, a three-layer SVM network was constructed. The input layer has three items, corresponding to the R, G, and B color values, and the output layer has one item, namely, the two-dimensional temperature T of the flame radiation energy image.

[0077] The implementation of the above embodiment has the following effects:

[0078] The technical solution of the present invention obtains the detection data in the boiler in real time, and then digitizes the corresponding detection data to obtain the corresponding boiler combustion image, and then determines the three primary color values ​​of any point in the image, so that after each three primary color value is input into the temperature measurement grid model in turn, the temperature value corresponding to each pixel point can be output, and the temperature field distribution of the boiler is generated based on the temperature value, so that the distribution of the flame temperature field in the current boiler furnace can be obtained accurately and quickly, and the temperature field state in the boiler can also be detected more timely, comprehensively and accurately, avoiding the problem that the existing equipment can only perform short-term measurements and is difficult to accurately reflect the combustion state in the furnace, thereby being unable to depict the temperature distribution of the entire furnace.

[0079] Example 2

[0080] This embodiment is a preferred implementation of a method for real-time monitoring of boiler temperature field based on image recognition.

[0081] As a preferred solution of this embodiment, real-time acquisition of detection data from the boiler and digital processing of the detection data to obtain a boiler combustion image specifically includes:

[0082] Through the lens set in the boiler fire viewing hole, the integration time is adjusted, and the detection signal of the one-dimensional time series is obtained according to the preset sampling frequency; the detection signal of the one-dimensional time series is digitized so that the detection signal of the one-dimensional time series is discretized into a detection data signal; and an image is generated according to the detection digital signal to obtain a boiler combustion image.

[0083] In this embodiment, after turning on the portable air compressor, the camera power switch, and the computer, the entire system is started and operated. Furthermore, the lens that can be installed in the boiler's fire viewing hole can be a CCD camera lens, through which detection data information within the boiler can be obtained. Preferably, the CCD camera can be controlled by a software interface program installed on the computer to begin collecting radiation data and radiation image data, with the sampling frequency set to 1 Hz. During the sampling process, a suitable integration time can be adjusted to ensure that both the spectral data collected by the spectrometer and the image data collected by the CCD camera are non-saturated and have a high signal-to-noise ratio.

[0084] In this embodiment, preferably, the CCD camera outputs a one-dimensional time series signal, which needs to be digitized, that is, a flame image (i.e., boiler combustion image) is represented by a finite array or sequence, and each frame of the video image is discretized from a continuous analog signal into a digital signal.

[0085] Specifically, the detection signal of the one-dimensional time series is digitized, including the digitization of the amplitude and the digitization of the spatial coordinates.

[0086] As a preferred solution of this embodiment, digital processing is performed on the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal, which specifically includes:

[0087] The detection signal of the one-dimensional time series is converted into a frequency domain signal, and the feature of the converted frequency domain signal is extracted to obtain the corresponding amplitude feature; the corresponding continuous amplitude feature is binned to obtain the detection data signal related to the amplitude size; the cutoff frequency and the original distribution of the detection signal of the one-dimensional time series are extracted, and according to the extracted cutoff frequency and the original distribution, the detection data signal is converted to obtain the corresponding detection data signal related to the spatial coordinates; wherein, the detection data signal includes a signal related to the amplitude size and the spatial coordinates.

[0088] In this embodiment, the digitization of the image consists of two parts: the digitization of the amplitude and the digitization of the spatial coordinates, and image sampling is the digitization process of the spatial coordinates (x, y). For example, a pixel with a cutoff frequency f xg , f ygThe flame image has an original distribution of f (x, y), which will be distributed as f after sampling. g (i, j). It can be understood that image sampling is the process of digitizing continuous image space coordinates, that is, converting the continuous space coordinates of the image into discrete pixel points. For a flame image, its original distribution is f(x,y)f(x,y), and the image distribution obtained after sampling is f g (i,j)f g (i,j), where i and j represent the discretized x and y coordinates respectively.

[0089] In this embodiment, it is necessary to set the sampling frequency, which determines the resolution of the image. In a two-dimensional image, there are usually two sampling frequencies f x and f y , corresponding to the sampling intervals in the x and y directions, respectively, which is 1 Hz in this embodiment. 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 up. During the sampling process, the continuous grayscale or color values ​​of the image also need to be quantized into finite numerical values, where the grayscale or color range of the image is divided into a finite number of levels, and each level is assigned a digital value.

[0090] As a preferred solution of this embodiment, determining the three primary color values ​​corresponding to any pixel point based on the boiler combustion image specifically includes:

[0091] Based on the true color image card, the boiler combustion image is color recognized to obtain the grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the boiler combustion image as the three primary color values ​​corresponding to any pixel.

[0092] In this embodiment, the color information in the image can be grayscaled using a 24-bit true color image card, thereby obtaining a grayscale quantization value as the three primary color values. A 24-bit true color image card can distinguish millions of colors, corresponding to 256 quantization levels, with each color channel being 8 bits. These R, G, and B values ​​represent the grayscale quantization values ​​of red, green, and blue at each coordinate.

[0093] In this embodiment, each pixel in the boiler combustion image has a corresponding spatial coordinate, and each spatial coordinate is the coordinate of the corresponding pixel. Each pixel has a corresponding grayscale quantization value of red, green, and blue, that is, the three primary color values ​​corresponding to the pixel.

[0094] As a preferred solution of this embodiment, each of the three primary color values ​​is sequentially input into the temperature measurement grid model, the temperature value corresponding to each pixel is output, and the temperature field distribution of the boiler is generated based on the temperature value, specifically including:

[0095] Each of the three primary color values ​​is used as the input of 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 pixel point in the boiler combustion image; based on the two-dimensional temperature values ​​and their corresponding spatial coordinates, the temperature field distribution of the boiler is generated.

[0096] In this embodiment, the three primary color values ​​corresponding to each pixel point are respectively used as the input of the temperature measurement grid model, that is, the R, G, and B values ​​corresponding to each pixel point are simultaneously used as the input of the temperature measurement grid model. Among them, the temperature measurement grid model has three input nodes, so that the two-dimensional temperature value corresponding to the pixel point can be directly output. After all the pixels in the image output the corresponding two-dimensional temperature values, the distribution map of the two-dimensional temperature value of the entire image can be obtained.

[0097] As a preferred solution of this embodiment, each of the three primary color values ​​is sequentially input into the temperature measurement grid model, the temperature value corresponding to each pixel is output, and the temperature field distribution of the boiler is generated based on the temperature value, which specifically includes:

[0098] The pixels in the boiler combustion image are divided into regions, and the middle pixel of each region is determined as the target pixel according to the regions of the divided pixels; wherein the number and size of the pixels in each region are equal; the three primary color values ​​corresponding to the target pixels are respectively used as the input of 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 the boiler combustion image; based on the region where each target pixel is located, the edge of the target pixels in the adjacent regions are smoothed respectively, so as to calculate the two-dimensional temperature values ​​corresponding to the remaining pixels in each region; the temperature field distribution of the boiler is generated through the two-dimensional temperature values ​​of all pixels and their corresponding spatial coordinates.

[0099] In this embodiment, since the number of pixels in the boiler combustion image is very large, using a temperature measurement grid model to sequentially calculate the pixels in the boiler image to obtain the corresponding temperature values ​​would be extremely time-consuming. Instead, the pixels in the boiler combustion image are divided into regions, preferably, pixels of a size of 3*3 or 5*5 can be used as a region. The middle pixel in each region is then determined as the target pixel. Specifically, the three primary color values ​​corresponding to the target pixel are used as inputs to the temperature measurement grid model. The temperature measurement grid model then calculates the input three primary color values ​​to obtain the two-dimensional temperature value corresponding to the middle pixel in each region.

[0100] In this embodiment, by performing temperature edge smoothing calculation on each area and its corresponding adjacent areas, the temperature values ​​corresponding to the remaining pixels except the middle pixel in each area can be directly calculated, so that a large amount of pixel point information can be avoided from being directly input into the temperature measurement grid model for calculation, the model calculation amount is reduced, and the computing resources are optimized. At the same time, the accuracy of temperature detection can be guaranteed through smoothing algorithms such as Gaussian filtering.

[0101] It is understandable that for a boiler combustion image, the most important information for detecting the combustion temperature field is to obtain the temperature of the flame combustion. The temperature mutation inside the flame body during flame combustion is not very large, so the corresponding smoothing algorithm can be directly used to greatly reduce the corresponding calculation amount. The temperature mutation around the flame body is also not very large, so the corresponding calculation amount can also be greatly reduced. Since the temperature data of the flame combustion body already exists, there will be an obvious temperature mutation edge between the flame body and its edge. For example, the position corresponding to the central pixel point in a certain area is the flame body, and its temperature is a. The position corresponding to the central pixel point in the adjacent area is the edge space, and its temperature is b. Then, after directly determining that the temperature between a and b is greater than the preset value, it can be determined that the temperature mutation edge between the two areas is. The temperature mutation edge can be calculated by setting the corresponding smoothing weight and other coefficients, and then the temperature value of the corresponding edge pixel point can be finally calculated, which can greatly avoid the calculation of a large amount of similar data.

[0102] As a preferred solution of this embodiment, a method for constructing a temperature measurement grid model includes:

[0103] Acquire sample data and perform normalization on the sample data; wherein the sample data includes: several groups of flame radiation temperatures and three primary color values ​​corresponding to each flame radiation temperature; set 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, thereby obtaining the optimal penalty factor, optimal kernel parameter and kernel minimum mean standard error corresponding to the initial temperature measurement network model; update the optimal penalty factor and optimal kernel parameter and kernel minimum mean standard error and set them in the initial temperature measurement network model, thereby training to obtain the final temperature measurement grid model.

[0104] In this embodiment, sample data is obtained and normalized, and then a temperature measurement grid model can be trained. The temperature measurement grid model can be an SVM model. An initial SVM model is then set, with the number of input nodes and output nodes of the SVM network set to 3 and 1, respectively. The sample data used for model training is shown in Table 1.

[0105]

[0106] In this embodiment, the kernel function of the initial SVM model selects the RBF Gaussian kernel function:

[0107]

[0108] Where σ is the standard deviation of the Gaussian function, which affects the width of the kernel function. The larger the value of σ, the wider the kernel function and the smoother the decision boundary. The smaller the value of σ, the narrower the kernel function and the more complex the decision boundary. i are the kernel function values ​​of two pixel data points respectively.

[0109] In this embodiment, during network training, the penalty factor (c) was set to a range of -10 to 10, and the kernel parameter (g) was set to a range of -5 to 5 to optimize the parameters. Here, the values ​​of c and g were logarithmized. The training yielded the optimal c (bestc), optimal g (bestg), and minimum mean standard error (MSE) values ​​for the SVM network: bestc = 90.5097; bestg = 0.03125; and minimum MSE = 214.061.

[0110] In this example, an SVM temperature measurement grid model was established. A large sample database was used to train the temperature measurement grid model. The number of input nodes in the SVM network was selected to be 3, and the number of output nodes was selected to be 1. Leveraging the characteristics of the SVM network, a three-layer SVM network was constructed. The input layer contained three items, namely, the R, G, and B color values, and the output layer contained one item, namely, the two-dimensional temperature value T of the flame radiation energy image.

[0111] As another preferred 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 that combines the characteristics of deep learning and SVM, so that the corresponding detection temperature value can be output more accurately through the temperature measurement grid model.

[0112] As a preferred solution of this embodiment, obtaining sample data specifically includes:

[0113] Obtain a black furnace flame image and each group of flame radiation temperatures in a preset black furnace; perform color recognition on the black furnace flame image based on a true color image card, and obtain grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the black furnace flame image; determine correction coefficients for any two components of the three colors based on the monochromatic emissivity of the preset black furnace, the flame radiation temperature and the wavelengths corresponding to the three colors of light; determine the monochromatic radiation intensity of the combustion flame spectrum of each color based on the correction coefficients and the wavelengths of light corresponding to the three colors, combined with the monochromatic emissivity of the flame, and thus obtain the light intensity signal value at the wavelength corresponding to each color of light; based on the light intensity signal value, obtain a relationship between the corresponding temperature, the light wavelength and the monochromatic emissivity of the preset black furnace, and calibrate the grayscale quantization values ​​corresponding to the three colors of the black furnace flame image based on the relationship and the correction coefficient, and thus obtain the three primary color values ​​corresponding to each flame radiation temperature.

[0114] In this embodiment, a black body furnace flame image is collected and stored using a temperature control system, thereby obtaining several sets of sample data, preferably 30 sets. Furthermore, color recognition is performed on the black body flame image based on a true color image card, thereby identifying the grayscale quantization values ​​corresponding to the red, green, and blue colors of each pixel coordinate in the black body flame image. Based on the preset monochromatic emissivity of the black body furnace, the flame radiation temperature, and the wavelengths corresponding to the three colors of light, the correction coefficients for any two components of the three colors are determined:

[0115]

[0116]

[0117] in, is the monochromatic radiance of an artificial black body; ; It represents the three primary colors of light R, G, and B with wavelengths of 700nm, 546.1nm, and 435.8nm respectively; T is the flame temperature of the blackbody furnace, in K.

[0118] According to Wien's formula:

[0119]

[0120] in, is the monochromatic radiation intensity of the burning flame spectrum [W / m 3 ]; is the flame monochromatic radiance; is the wavelength (m); T is the absolute temperature (K); are Planck constants respectively; .

[0121] Take the light intensity signal values ​​at three different wavelengths: 700nm, 546.1nm, and 435.8nm , take the ratio of the two light intensity values ​​at different wavelengths (preferably, the ratio of the light intensity signal values ​​between R and G and between G and B), then:

[0122]

[0123]

[0124] Therefore, by dividing the two equations, we can get:

[0125] Then, the relationship between the flame radiation temperature of each group and the gray value of the radiation thermal image output by the color CCD camera can be obtained. Then the system is calibrated. The calibration result is used to obtain the ideal fitting curve through appropriate fitting method. The expression after calibration is:

[0126]

[0127] in, , are the coefficients for correcting the G and B components respectively; R, G, and B are the three primary color values ​​corresponding to any pixel on the flame image obtained by the CCD camera, and R', G', and B' are the results of R, G, and B calibration respectively.

[0128] As a preferred solution of this embodiment, determining the three primary color values ​​corresponding to any pixel point based on the boiler combustion image specifically includes:

[0129] Based on the true color image card, the boiler combustion image is color recognized to obtain the grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the boiler combustion image; according to the relationship and correction coefficient, the grayscale quantization values ​​corresponding to the three colors of the boiler combustion image are calibrated to determine the three primary color values ​​corresponding to any pixel.

[0130] In this embodiment, since there is a corresponding calibration expression when establishing the temperature measurement grid model, when obtaining the grayscale quantization values ​​corresponding to the red, green and blue colors of the coordinates of each pixel point in the specific boiler combustion image, the corresponding system calibration can also be performed, that is, according to the corresponding relationship and correction coefficient, the grayscale quantization values ​​corresponding to R, G, and B of the boiler combustion image are calibrated, so as to obtain the three primary color values ​​that can be input into the temperature measurement grid model after the final calibration, thereby greatly improving the accuracy of obtaining image data for temperature measurement detection, and having higher accuracy than direct single model detection.

[0131] The implementation of the above embodiment has the following effects:

[0132] The technical solution of the present invention obtains the detection data in the boiler in real time, and then digitizes the corresponding detection data to obtain the corresponding boiler combustion image, and then determines the three primary color values ​​of any point in the image, so that after each three primary color value is input into the temperature measurement grid model in turn, the temperature value corresponding to each pixel point can be output, and the temperature field distribution of the boiler is generated based on the temperature value, so that the distribution of the flame temperature field in the current boiler furnace can be obtained accurately and quickly, and the temperature field state in the boiler can also be detected more timely, comprehensively and accurately, avoiding the problem that the existing equipment can only perform short-term measurements and is difficult to accurately reflect the combustion state in the furnace, thereby being unable to depict the temperature distribution of the entire furnace.

[0133] Example 3

[0134] See also Figure 2 , which is a method and device for real-time monitoring of boiler temperature field based on image recognition provided by the present invention, comprising: an acquisition module 201, a three-primary color module 202 and a temperature module 203;

[0135] The acquisition module 201 is used to acquire the detection data in the boiler in real time and digitally process the detection data to obtain the boiler combustion image;

[0136] The three-primary color module 202 is used to determine the three-primary color value corresponding to any pixel point based on the boiler combustion image;

[0137] The temperature module 203 is used to input each of the three primary color values ​​into the temperature measurement grid model in sequence, output the temperature value corresponding to each pixel point, and generate the temperature field distribution of the boiler based on the temperature value.

[0138] As a preferred solution, real-time acquisition of boiler detection data is performed, and the detection data is digitally processed to obtain a boiler combustion image, specifically including:

[0139] Through the lens set in the boiler fire viewing hole, the integration time is adjusted, and the detection signal of the one-dimensional time series is obtained according to the preset sampling frequency;

[0140] Digitally processing the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal;

[0141] An image is generated based on the detected digital signal to obtain a boiler combustion image.

[0142] As a preferred solution, digital processing is performed on the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal, which specifically includes:

[0143] The detection signal of the one-dimensional time series is converted into a frequency domain signal, and the converted frequency domain signal is subjected to feature extraction to obtain the corresponding amplitude feature;

[0144] The corresponding continuous amplitude features are binned to obtain detection data signals involving amplitude size;

[0145] Extracting the cutoff frequency and original distribution of the detection signal of the one-dimensional time series, and converting the extracted cutoff frequency and original distribution into a detection data signal corresponding to the spatial coordinates;

[0146] The detection data signal includes a signal involving amplitude and spatial coordinates.

[0147] As a preferred solution, the three primary color values ​​corresponding to any pixel point are determined based on the boiler combustion image, specifically including:

[0148] Based on the true color image card, the boiler combustion image is color recognized to obtain the grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the boiler combustion image as the three primary color values ​​corresponding to any pixel.

[0149] As a preferred solution, each of the three primary color values ​​is sequentially input into the temperature measurement grid model, and the temperature value corresponding to each pixel is output. The temperature field distribution of the boiler is generated based on the temperature value, specifically including:

[0150] Each of the three primary color values ​​is used as an input of the temperature measurement grid model, so that the temperature measurement grid model calculates the input three primary color values ​​and outputs a two-dimensional temperature value corresponding to each pixel in the boiler combustion image;

[0151] Based on the two-dimensional temperature values ​​and their corresponding spatial coordinates, the temperature field distribution of the boiler is generated.

[0152] As a preferred solution, each of the three primary color values ​​is sequentially input into the temperature measurement grid model, and the temperature value corresponding to each pixel is output. The temperature field distribution of the boiler is generated based on the temperature value, which specifically includes:

[0153] The pixels in the boiler combustion image are divided into regions, and the middle pixel of each region is determined as the target pixel according to the regions of the divided pixels; wherein the number and size of pixels in each region are equal;

[0154] The three primary color values ​​corresponding to the target pixel points are used as inputs of 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 point in the boiler combustion image;

[0155] Based on the area where each target pixel point is located, edge smoothing is performed on the target pixel points in the adjacent areas respectively, so as to calculate the two-dimensional temperature values ​​corresponding to the remaining pixel points in each area;

[0156] The temperature field distribution of the boiler is generated through the two-dimensional temperature values ​​of all pixel points and their corresponding spatial coordinates.

[0157] As a preferred solution, a method for constructing a temperature measurement grid model includes:

[0158] Acquire sample data and perform normalization processing on the sample data; wherein the sample data includes: a plurality of sets of flame radiation temperatures and three primary color values ​​corresponding to each flame radiation temperature;

[0159] Setting an initial temperature measurement network model, and determining the penalty factor, Gaussian kernel function, and kernel parameters of the initial temperature measurement network model;

[0160] The set initial temperature measurement network model is trained based on the sample data to obtain the optimal penalty factor, optimal kernel parameter kernel minimum mean standard error corresponding to the initial temperature measurement network model;

[0161] The optimal penalty factor, the optimal kernel parameter kernel minimum mean standard error update are set in the initial temperature measurement network model, so as to train the final temperature measurement grid model.

[0162] As a preferred solution, obtaining sample data specifically includes:

[0163] Obtaining a black furnace flame image and each group of flame radiation temperatures in a preset black furnace;

[0164] Based on the true color image card, the color recognition of the black furnace flame image is performed to obtain the grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the black furnace flame image;

[0165] Based on the preset monochromatic emissivity of the black furnace, the flame radiation temperature and the wavelengths corresponding to the three colors of light, the correction coefficients of any two components of the three colors are determined;

[0166] According to the correction coefficient and the wavelengths of light corresponding to the three colors, combined with the monochromatic radiation rate of the flame, 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 light;

[0167] Based on the light intensity signal value, 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 grayscale 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.

[0168] As a preferred solution, the three primary color values ​​corresponding to any pixel point are determined based on the boiler combustion image, which specifically includes:

[0169] Based on the true color image card, the boiler combustion image is color recognized to obtain the grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel coordinate in the boiler combustion image;

[0170] According to the relationship and correction coefficient, the grayscale quantization values ​​corresponding to the three colors of the boiler combustion image are calibrated to determine the three primary color values ​​corresponding to any pixel point.

[0171] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0172] The implementation of the above embodiment has the following effects:

[0173] The technical solution of the present invention obtains the detection data in the boiler, and then digitizes the corresponding detection data to obtain the corresponding boiler combustion image, and then determines the three primary color values ​​of any point in the image, so that after each three primary color value is input into the temperature measurement grid model in turn, the temperature value corresponding to each pixel point can be output, and the temperature field distribution of the boiler is generated based on the temperature value, so that the distribution of the flame temperature field in the current boiler furnace can be obtained accurately and quickly, and at the same time, the temperature field state in the boiler can be detected more timely, comprehensively and accurately, avoiding the problem that the existing equipment can only perform short-time measurements and it is difficult to accurately reflect the combustion state in the furnace, thereby failing to depict the temperature distribution of the entire furnace.

[0174] Example 4

[0175] See also Figure 3The present invention also provides a method system for real-time monitoring of boiler temperature field based on image recognition, which includes: an optical system, an air cooling system, a combustion state acquisition system and an image processing system.

[0176] The optical system includes: a housing 7, an image transmission optical fiber 10 passing through 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 a porous plate air outlet is provided at the front end of the front cavity.

[0177] As a preferred solution of this embodiment, please refer to Figure 4 The first end of the image transmission optical fiber 10 is connected to a collimating lens 13, the second end of the image transmission optical fiber 10 is connected to a color filter 6, and the image transmission optical fiber 10 is wrapped with a high-temperature resistant protective material 12; a lens 15 is provided at the front end of the mirror rod 17, and a plurality of optical lenses 16 and a dual-bandpass filter 21 are arranged at intervals inside the mirror rod 17.

[0178] In this embodiment, the number of optical lenses 16 can be set according to actual needs. Preferably, the number of optical lenses 16 is greater than or equal to 3.

[0179] As a preferred solution of this embodiment, a plurality of porous support frames 9 are arranged in the front end cavity of the shell 7 along the first end and the second end direction of the front end cavity. The porous support frames 9 are used to support the image transmission optical fiber 10 and the mirror rod 17.

[0180] As a preferred solution of this embodiment, a fire-blocking plate 11 is further provided on the outside of the front end cavity of the shell 7 , and a hand-held handle 20 is further provided on the second end position of the outside of the front end cavity of the shell 7 .

[0181] In this embodiment, the fire-blocking plate can prevent the high-temperature gas or coal ash ejected from the fire-viewing hole from burning the operator.

[0182] The air cooling system includes a cooler 18 and a cooling air inlet 19 and a cooling air outlet 8 respectively connected to the optical system.

[0183] The combustion state acquisition module is connected to the second end of the mirror rod 17, and includes: a camera arranged in the rear end cavity of the shell 7, a color filter 6 arranged in front of the camera, a microcomputer 22 arranged in the rear end cavity of the shell 7, an image acquisition card 23 connected to the microcomputer 22 and arranged on the rear cavity wall of the shell 7, and a power supply 4 arranged in the rear end cavity of the shell 7.

[0184] As a preferred solution of this embodiment, the power supply 4 is connected to the camera and the microcomputer 22 through the power switch 2.

[0185] Optionally, the camera may be a CCD camera that can directly capture images of the flame 14, see Figure 5 , which is a spectral response characteristic curve of the dual-bandpass filter and the CCD camera.

[0186] The image processing system 3 is used to execute any of the above furnace image combustion temperature field detection methods, including: an image acquisition card 23 and an image processing unit; the image acquisition card 23 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.

[0187] 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.

[0188] In this embodiment, the image processing system 3 is connected to the image acquisition card 23 via the first interface 1 .

[0189] In this embodiment, in the image processing system, the CCD camera 5 has a 24-bit true color image card and adopts a color CCD three-color temperature measurement method. The grayscale quantization values ​​of red, green, and blue at each coordinate are represented by R, G, and B. Subsequently, the system is calibrated to establish the relationship between the flame radiation temperature and the grayscale value of the radiation thermal image output by the color CCD camera 5, and finally the R, G, and B three-color values ​​are obtained.

[0190] In this embodiment, the R, G, and B values ​​obtained by the image processing system are brought into the established multi-layer perceptron neural network to obtain the temperature field distribution of the power station boiler furnace section.

[0191] Optionally, the first interface 1 is a gigabit network interface, which can be connected to a computer.

[0192] Optionally, the image transmission optical fiber 10 is an armored optical fiber, the lens 15 is a pinhole lens, and the power supply 4 is a lithium battery.

[0193] In this embodiment, the compressed air introduced into the boiler can enter the device through the cooling air inlet 19. After that, the cooling air is divided into two paths; one path flows out through the cooling air outlet 8, and the other path flows out through the porous plate outlet 2. The porous plate outlet 2 can cool the lens 10 and the collimating lens 6, and after the cooling air enters the device, it will also purge the lens 10 and the collimating lens 6 to keep them clean. The collimating lens 6 only allows parallel light directly in front of it to enter. The incident light diagram is shown in FIG. Figure 4As shown. The radiant light from the coal powder burning flame in the boiler furnace enters the collimating lens 6 and then enters the lens 15 along the image transmission optical fiber 10. After passing through the lens group composed of multiple optical lenses 16, it is incident on the dual-bandpass filter 21. The image on the dual-bandpass filter 21 is captured by the CCD camera 5. The color CCD camera decomposes the incident light into three-color images of red (R), green (G), and blue (B) according to the different wavelengths. The wavelengths are 700nm, 546.1nm and 435.8nm respectively. The porous structure of the porous support frame 17 can also pass through cooling air. The control instructions and data transmission of the CCD camera 5 are carried out through the gigabit network cable. The gigabit network interface is connected to the subsequent image processing system 3, which can play the role of power supply and data transmission control at the same time. The CCD camera 5 is powered by a built-in lithium battery, and the power switch 2 can be used to control the power on or off.

[0194] It is understandable that existing detection methods can also reconstruct the temperature field distribution inside the furnace by utilizing the acoustic wave method or laser absorption spectroscopy, but both methods have certain technical disadvantages. The acoustic wave method requires the installation of a certain number of transmitters and receivers on the furnace, which increases the construction difficulty and hardware cost, and factors such as on-site vibration will affect the stability of the system operation. At the same time, the acoustic wave transmitter is large in size and the furnace wall opening is also large, which will also affect the temperature distribution inside the furnace to a certain extent. The laser absorption spectroscopy method is relatively expensive and has relatively high requirements for on-site operation, which is not convenient for power plant site staff to use. Therefore, the embodiment of the present invention has a simple structure, low cost, high stability, and is easy to operate. Through this device and detection method, a wireless real-time monitoring method of the boiler temperature field based on image recognition can be implemented, which can more accurately and comprehensively detect the temperature status inside the boiler.

[0195] The implementation of the above embodiment has the following effects:

[0196] The optical system of this embodiment includes a housing, an image transmission fiber and a mirror rod extending through a front cavity, a collimating lens connected to a first end of the image transmission fiber, a lens disposed at the front end of the mirror rod, a plurality of optical lenses spaced from the first end to the second end of the mirror rod within the mirror 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 mirror rod and disposed within the rear cavity; a color filter placed in front of the CCD camera; a microcomputer disposed within the rear cavity; a first interface connected to the microcomputer for transmitting spectral data and providing power; an image acquisition card disposed on the rear cavity wall; a power supply disposed within the rear cavity connecting the CCD camera and the microcomputer; and an image processing system including the image acquisition card and an image processing unit. By analyzing the flame image in detail, a temperature distribution map of the boiler combustion zone can be obtained, capturing minute details of temperature changes and providing a more comprehensive understanding of the combustion process. This device has a simple structure, low cost, high stability, and is easy to operate. Through this device and the corresponding detection method, a wireless real-time monitoring method of the boiler temperature field based on image recognition can be realized, which can more accurately and comprehensively detect the temperature status in the boiler.

[0197] Example 5

[0198] Accordingly, the present invention also provides 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 when the processor executes the computer program, it implements the real-time monitoring method of the boiler temperature field based on image recognition as described in any one of the above embodiments.

[0199] The terminal device of this embodiment includes: a processor, a memory, and a computer program and computer instructions stored in the memory and capable of running on the processor. When the processor executes the computer program, each step in the above embodiment 1 is implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of each module / unit in the above device embodiment, such as the temperature module 203, are implemented.

[0200] Exemplarily, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program in the terminal device. For example, the temperature module 203 is used to sequentially input each of the three primary color values ​​into the 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.

[0201] The terminal device may be a computing device such as a desktop computer, laptop, PDA, or cloud server. The terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of a terminal device and does not limit the terminal device. The terminal device may include more or fewer components than shown, or a combination of certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, and the like.

[0202] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of the terminal device and connects various parts of the entire terminal device using various interfaces and lines.

[0203] The memory can be used 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 calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store an operating system, at least one application required for a function, etc.; the data storage area may store data created based on the use of the mobile terminal, etc. In addition, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0204] If the module / unit integrated into the terminal device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0205] Example 6

[0206] Accordingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the real-time monitoring method of the boiler temperature field based on image recognition as described in any one of the above embodiments.

[0207] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A real-time monitoring method for boiler temperature field based on image recognition, characterized in that: include: Acquire detection data in the boiler, and digitally process the detection data to obtain a boiler combustion image; According to the boiler combustion image, determining the three primary color values ​​corresponding to any pixel point; Each of the three primary color values ​​is sequentially input into the temperature measurement grid model, the temperature value corresponding to each pixel point is output, and the temperature field distribution of the boiler is generated based on the temperature value.

2. A method for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 1, characterized in that: The acquiring of the detection data in the boiler and digital processing of the detection data to obtain the boiler combustion image specifically includes: Through the lens set in the boiler fire viewing hole, the integration time is adjusted, and the detection signal of the one-dimensional time series is obtained according to the preset sampling frequency; Digitally process the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal; An image is generated according to the detected digital signal to obtain a boiler combustion image.

3. A method for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 2, characterized in that: The step of digitally processing the detection signal of the one-dimensional time series so as to discretize the detection signal of the one-dimensional time series into a detection data signal specifically includes: Convert the detection signal of the one-dimensional time series into a frequency domain signal, and extract features of the converted frequency domain signal to obtain corresponding amplitude features; The corresponding continuous amplitude features are binned to obtain a detection data signal involving the amplitude size; Extracting the cutoff frequency and the original distribution of the detection signal of the one-dimensional time series, and converting the extracted cutoff frequency and the original distribution to obtain the detection data signal corresponding to the spatial coordinates; The detection data signal includes a signal involving amplitude and spatial coordinates.

4. A method for real-time monitoring of boiler temperature field based on image recognition according to any one of claims 1 to 3, characterized in that: Determining the three primary color values ​​corresponding to any pixel point according to the boiler combustion image specifically includes: Based on the true color image card, the boiler combustion image is color recognized to obtain grayscale quantized values ​​corresponding to the red, green and blue colors of each pixel coordinate in the boiler combustion image as the three primary color values ​​corresponding to any pixel.

5. The method for real-time monitoring of boiler temperature field based on image recognition according to claim 1, characterized in that: The step of sequentially inputting each of the three primary color values ​​into the temperature measurement grid model, outputting a temperature value corresponding to each pixel point, and generating the temperature field distribution of the boiler based on the temperature value specifically includes: Using each of the three primary color values ​​as an input of a temperature measurement grid model, so that the temperature measurement grid model calculates the input three primary color values ​​and outputs a two-dimensional temperature value corresponding to each pixel point in the boiler combustion image; Based on the two-dimensional temperature values ​​and their corresponding spatial coordinates, a temperature field distribution of the boiler is generated.

6. A method for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 1, characterized in that: The step of sequentially inputting each of the three primary color values ​​into the temperature measurement grid model, outputting a temperature value corresponding to each pixel point, and generating the temperature field distribution of the boiler based on the temperature value specifically includes: Divide the pixels in the boiler combustion image into regions, and determine the middle pixel of each region as the target pixel according to the regions of the divided pixels; wherein the number and size of the pixels in each region are equal; The three primary color values ​​corresponding to the target pixel points are respectively used as inputs of the temperature measurement grid model, so that the temperature measurement grid model calculates the input three primary color values ​​and outputs a two-dimensional temperature value corresponding to each target pixel point in the boiler combustion image; Based on the area where each target pixel point is located, edge smoothing is performed on the target pixel points in the adjacent areas respectively, so as to calculate the two-dimensional temperature values ​​corresponding to the remaining pixel points in each area respectively; The temperature field distribution of the boiler is generated through the two-dimensional temperature values ​​of all pixel points and their corresponding spatial coordinates.

7. A method for real-time monitoring of boiler temperature field based on image recognition according to any one of claims 1, 2, 3, 5 or 6, characterized in that: The method for constructing the temperature measurement grid model comprises: Acquire sample data and perform normalization processing on the sample data; wherein the sample data includes: a plurality of groups of flame radiation temperatures and three primary color values ​​corresponding to each flame radiation temperature; Setting an initial temperature measurement network model, and determining a penalty factor, a Gaussian kernel function, and kernel parameters of the initial temperature measurement network model; Based on the sample data, the set initial temperature measurement network model is trained to obtain the optimal penalty factor, the optimal kernel parameter and the minimum average standard error of the kernel corresponding to the initial temperature measurement network model; The optimal penalty factor, the optimal kernel parameter kernel minimum mean standard error are updated and set in the initial temperature measurement network model, so as to train and obtain the final temperature measurement grid model.

8. A method for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 7, characterized in that: The obtaining of sample data specifically includes: Obtaining a black furnace flame image and each group of flame radiation temperatures in a preset black furnace; Based on the true color image card, color recognition is performed on the black furnace flame image to obtain grayscale quantization values ​​corresponding to the red, green and blue colors of each pixel point coordinate in the black furnace flame image; Determine the correction coefficients of any two components of the three colors based on the monochromatic emissivity of the preset black furnace, the flame radiation temperature and the wavelengths corresponding to the three colors of light; According to the correction coefficient and the wavelengths of light corresponding to the three colors, combined with the monochromatic radiation rate of the flame, the monochromatic radiation intensity of the combustion flame spectrum of each color is determined, so as to obtain the light intensity signal value at the wavelength corresponding to each color light; Based on the light intensity signal value, a relationship between the corresponding temperature and the light wavelength and the monochromatic emissivity of the preset black furnace is obtained, and according to the relationship and the correction coefficient, the grayscale 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. A method for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 8, characterized in that: The step of determining the three primary color values ​​corresponding to any pixel point according to the boiler combustion image specifically includes: Based on a true color image card, color recognition is performed on the boiler combustion image to obtain grayscale quantization values ​​corresponding to red, green and blue colors of each pixel point coordinate in the boiler combustion image; According to the relationship and the correction coefficient, the grayscale quantization values ​​corresponding to the three colors of the boiler combustion image are calibrated, so as to determine the three primary color values ​​corresponding to any pixel point.

10. A method and device for real-time monitoring of boiler temperature field based on image recognition, characterized in that: include: Acquisition module, three-primary color module and temperature module; The acquisition module is used to acquire detection data in the boiler and digitally process the detection data to obtain a boiler combustion image; The three-primary color module is used to determine the three-primary color value corresponding to any pixel point according to the boiler combustion image; The temperature module is used to input each of the three primary color values ​​into the temperature measurement grid model in sequence, output the temperature value corresponding to each pixel point, and generate the temperature field distribution of the boiler based on the temperature value.

11. A real-time monitoring method system for boiler temperature field based on image recognition, characterized in that: include: Optical system, air cooling system, combustion state acquisition system and image processing system; The optical system comprises: a housing, an image transmission optical fiber and a mirror rod penetrating a front cavity of the housing; wherein the housing comprises a front cavity and a rear cavity, and a porous plate air outlet is provided at the front end of the front cavity; The air cooling system comprises: a cooler and a cooling air inlet and a cooling air outlet respectively connected to the optical system; The combustion state acquisition module is connected to the second end of the mirror rod, and includes: a camera arranged in the rear end cavity of the shell, a color filter arranged in front of the camera, a microcomputer arranged in the rear end cavity of the shell, an image acquisition card connected to the microcomputer and arranged on the rear cavity wall of the shell, and a power supply arranged in the rear end cavity of the shell; The image processing system is used to execute the furnace image combustion temperature field detection method as described in any one of claims 1 to 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. A method and system for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 11, characterized in that: The first end of the image transmission optical fiber is connected to a collimating lens, the second end of the image transmission optical fiber is connected to a color filter, and the image transmission optical fiber is wrapped with a high temperature resistant protective material; A lens is arranged at the front end of the mirror rod, and a plurality of optical lenses and a dual-bandpass filter are arranged at intervals inside the mirror rod.

13. A method and system for real-time monitoring of boiler temperature field based on image recognition as claimed in claim 11, characterized in that: The power source is connected to the camera and the microcomputer through a power switch.

14. A method system for real-time monitoring of boiler temperature field based on image recognition according to any one of claims 11 to 13, characterized in that: A plurality of porous support frames are arranged in the front end cavity of the shell along the first end and the second end direction of the front end cavity, and the porous support frames are used to support the image transmission optical fiber and the mirror rod.

15. The method system for real-time monitoring of boiler temperature field based on image recognition according to any one of claims 11 to 13, characterized in that: A fire-blocking plate is also arranged outside the front end cavity of the shell, and a hand-held handle is also arranged at the second end position outside the front end cavity of the shell.

16. A terminal device, characterized in that: The method comprises 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, the method for real-time monitoring of boiler temperature field based on image recognition as described in any one of claims 1 to 9 is implemented.

17. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the real-time monitoring method of boiler temperature field based on image recognition as described in any one of claims 1 to 9.

Images