Thermal power boiler flame-permeable high-temperature filter coking identification display method and device

Abstract

The embodiment of the invention provides a thermal power boiler flame-permeable high-temperature filter coking recognition display method and device, electronic equipment and a medium, and the method comprises the steps: continuously collecting images of the surface of a flame-permeable high-temperature filter in real time through an industrial camera, and carrying out the preprocessing of the images; performing feature extraction and classification on the preprocessed image by using a neural network model, determining a coking area, analyzing the distribution difference between a normal area and the coking area, determining coking distribution, and calculating a coking quantification result; and displaying coking distribution and a coking quantification result by adopting a three-dimensional thermodynamic diagram. According to the embodiment of the invention, the operation safety and economy of the boiler can be improved, the cooperative monitoring of the combustion state of the boiler and the coking condition of the filter is realized, the intelligent upgrading of equipment management in the thermal power industry is promoted, and the application range is expanded.

Description

TECHNICAL FIELD

[0001] The present document relates to the technical field of power plant boiler coking monitoring, in particular to a kind of power plant boiler flame-through high temperature filter coking identification display, device, electronic equipment and medium. BACKGROUND

[0002] Power station boiler, also known as power plant boiler, is a large-scale boiler in thermal power plant, which provides steam for steam turbine, mainly used for power generation or heating, with large evaporation capacity, main unit is 600MW, ultra-supercritical boiler capacity can reach 1000MW, steam parameters include subcritical (15.7-19.6MPa) and supercritical (>22.1MPa) levels, according to combustion mode, it is divided into pulverized coal boiler and circulating fluidized bed boiler (CFB), the latter uses poly-state fluidized combustion technology, with wide fuel adaptability and low pollution emission characteristics.

[0003] In the thermal power industry, as the core thermal equipment, the combustion state in the furnace directly affects the power generation efficiency and operation safety. Flame-through high temperature filter is a key component of the boiler furnace visual monitoring system, which is used to clearly obtain the flame image inside the furnace in high temperature and high dust environment, providing an important basis for operation personnel to judge the combustion stability and adjust the operation parameters.

[0004] The coking detection of power plant boiler flame-through high temperature filter mainly depends on manual inspection, which has long detection period, strong subjectivity, high missed detection and false detection rate, inaccurate prediction of coking state and degree, and cannot be directly displayed.

[0005] In summary, there is an urgent need for a technical solution to solve the problem of coking detection of power plant boiler flame-through high temperature filter, which mainly depends on manual inspection, has long detection period, strong subjectivity, high missed detection and false detection rate, inaccurate prediction of coking state and degree, and cannot be directly displayed. SUMMARY

[0006] The purpose of the present application is to provide a kind of power plant boiler flame-through high temperature filter coking identification display method, device, electronic equipment and medium, to solve the above problems in the prior art.

[0007] The present application provides a kind of power plant boiler flame-through high temperature filter coking identification display method, which comprises: The image of the flame-through high temperature filter surface is continuously and real-time collected by industrial camera, and the image is pretreated; The pretreated image is subjected to feature extraction and classification by using neural network model, the coking area is determined, the distribution difference between normal area and coking area is analyzed, the coking distribution is determined, and the coking quantization result is calculated; The coking distribution and coking quantization result are displayed by using three-dimensional thermal map.

[0008] The present application provides a kind of thermal power boiler flame-through high temperature filter coking identification display device, comprising: The acquisition preprocessing module is used for continuously collecting the image of the flame-through high temperature filter surface in real time through the industrial camera, and pre-processing the image; The extraction calculation module is used for extracting and classifying the features of the pre-processed image using a neural network model, determining the coking area, analyzing the distribution difference between the normal area and the coking area, determining the coking distribution, and calculating the coking quantification result. The display module is used to display the coking distribution and the coking quantification result using a three-dimensional heat map.

[0009] The present application also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, which implements the steps of the above-mentioned thermal power boiler flame-through high temperature filter coking identification display method when executed by the processor.

[0010] The present application also provides a computer-readable storage medium having an information transmission implementation program stored thereon, which implements the steps of the above-mentioned thermal power boiler flame-through high temperature filter coking identification display method when executed by a processor.

[0011] With the present application, the image acquisition, processing and identification process takes less than 0.3 seconds, meeting the real-time monitoring requirements of the boiler and avoiding the lag of manual inspection. Through multi-feature fusion, the interference factors of coking and stains are effectively distinguished, and the algorithm parameters can be flexibly adjusted according to the parameters of the flame-through high temperature filter of different types of thermal power boilers, adapting to different furnace temperature and dust concentration conditions. The present coking quantification evaluation and grading early warning reduces the subjectivity of manual judgment, provides data support for boiler operation optimization, and can be widely used in boiler flame-through high temperature filter monitoring in the thermal power, thermal power and other industries. Not only can it improve the safety and economy of boiler operation, realize the collaborative monitoring of boiler combustion state and filter coking, promote the intelligent upgrading of equipment management in the thermal power industry, and expand the scope of application. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the one or more embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art without creating additional labor.

[0013] Figure 1 is the flow chart of the thermal power boiler flame-through high temperature filter coking identification display method of the present application; Figure 2 is a detailed process flowchart of the fire-through flame high-temperature filter coking identification display method of the embodiment of the present application; Figure 3 is a schematic diagram of the fire-through flame high-temperature filter coking identification display device of the embodiment of the present application; Figure 4 is a schematic diagram of the electronic device of the embodiment of the present application. DETAILED DESCRIPTION

[0014] In order for those skilled in the art to better understand the technical solutions in the one or more embodiments of the present specification, the technical solutions in the one or more embodiments of the present specification will be clearly and completely described below in conjunction with the drawings in the one or more embodiments of the present specification. Obviously, the described embodiments are only a part of the embodiments of the present specification, rather than all the embodiments. Based on the one or more embodiments of the present specification, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present document.

[0015] Method embodiment According to the embodiment of the present application, a fire-through flame high-temperature filter coking identification display method is provided, Figure 1 is a flowchart of the fire-through flame high-temperature filter coking identification display method of the embodiment of the present application, as Figure 1 shown, the fire-through flame high-temperature filter coking identification display method according to the embodiment of the present application specifically includes: Step S101, continuously and real-time collecting images of the surface of the fire-through flame high-temperature filter by an industrial camera, and pre-processing the images; specifically including: Collecting images of the surface of the fire-through flame high-temperature filter continuously and real-time by an industrial camera, acquiring 1 frame of image every predetermined time interval, transmitting to a computer terminal through an image acquisition card to form a real-time image stream; wherein the industrial camera is installed outside the boiler observation hole through a high-temperature resistant support, and the lens thereof is aligned with the center of the fire-through flame high-temperature filter.

[0016] The median filter is used to remove the salt and pepper noise in the image, the wavelet threshold denoising is used to process the Gaussian noise, the image is decomposed into low-frequency coefficient, medium-frequency coefficient and high-frequency coefficient, and the high-frequency coefficient is processed by soft threshold to reconstruct the image; the gray scale normalization is performed to map the image gray scale value to the interval of 0-255, and the influence of light change on the image gray scale feature is eliminated.

[0017] Step S102, using a neural network model to extract features and classify the pre-processed images, determining the coking area, analyzing the distribution difference between the normal area and the coking area, determining the coking distribution, and calculating the coking quantization result; the coking quantization result includes coking area ratio data and coking thickness data.

[0018] Step S102 specifically comprises: Using a neural network model, image denoising, Otsu threshold segmentation and morphological dilation are used to extract features and classify the preprocessed image, and the data diversity is enhanced through rotation, scaling, and flipping operations, and the image is normalized to determine the coking area. The Canny edge detection algorithm is used to set the high threshold, medium threshold and low threshold, and the edge profile of the coking area is extracted to distinguish the coking and the slight stains on the filter surface, analyze the distribution difference between the normal area and the coking area, and determine the coking distribution.

[0019] Step S103, using a three-dimensional heat map to display the coking distribution and coking quantification results. Specifically, it comprises: The left side of the display interface of the computer terminal displays the original image collected in real time on the filter surface; the middle area displays the clear image after preprocessing, and the coking area is marked, and the contour line of the coking area is superimposed; the right side area displays the coking quantification result, and the change trend of the coking area ratio in the past predetermined time is displayed in real time in the line chart.

[0020] The above technical solutions of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0021] As shown in Figure 2 The power plant boiler flame-through high-temperature filter coking identification display method comprises: S1, real-time image acquisition, continuously acquiring images of the flame-through high-temperature filter surface through an industrial camera, acquiring 1 frame of image every 0.5s, transmitting to the computer terminal through the image acquisition card to form a real-time image stream; first, remove the salt and pepper noise in the image by using median filtering to avoid the interference of noise on the edge of the coking area; then, remove the Gaussian noise by wavelet threshold denoising, decompose the image into low-frequency coefficient, medium-frequency coefficient and high-frequency coefficient, and reconstruct the image after soft threshold processing of the high-frequency coefficient to improve the signal-to-noise ratio of the image; finally, perform gray scale normalization to map the image gray value to the interval of 0-255, eliminate the influence of light changes on the image gray feature, and ensure the consistency of the image feature under different working conditions.

[0022] The industrial camera in step S1 is a high-temperature resistant industrial camera with a resolution of 1920x1080 or more, a frame rate of 25fps or more, a long-focus high-temperature resistant lens with a focal length of 25-100mm and an aperture of F1.8-F4.0, an image acquisition card supporting HDMI / USB3.0 interface, a computer processing terminal configured with GPU to improve the algorithm running efficiency, and a high-temperature resistant bracket for fixing the camera installed outside the boiler observation hole to ensure that the lens is aligned with the center of the flame-through high-temperature filter.

[0023] The computer in the step S1 realizes the full-process automation from image acquisition to early warning output through pre-installed special software, and the computer needs to have anti-interference and high computing power characteristics.

[0024] S2, a neural network model is used to extract features and classify the preprocessed image, to determine whether there is coking and the severity, to analyze the distribution difference between the normal area and the coking area, and to determine the segmentation threshold; the coking area extraction in the step S2 adopts image denoising, Otsu threshold segmentation and morphological dilation method, and through rotation, scaling, flipping and other operations to enhance the data diversity, and the high-temperature flame image needs to be normalized to reduce the influence of light difference.

[0025] The step S2 adopts a Canny edge detection algorithm, sets high, medium and low threshold values, extracts the edge profile of the coking area, and distinguishes the coking and the slight stain on the filter surface. The slight stain has a fuzzy stain edge and no obvious profile. The extraction of the edge profile of the coking area further assists in distinguishing the coking and the normal area because the coking area has lower texture entropy and weaker contrast due to impurity accumulation.

[0026] S3, the coking area ratio or thickness parameters are calculated to provide a basis for subsequent early warning; S4, a three-dimensional thermal map is used to display the coking distribution, and the color gradient corresponds to the thickness level. The real-time ratio data are displayed on the special display interface developed on the computer terminal. The display interface developed on the computer terminal is divided into three areas: the left side displays the original image collected in real time on the filter surface; the middle displays the clear image after preprocessing, and marks the coking area with a red frame and superimposes the profile line of the coking area; the right side displays the coking quantization result, and uses a line chart to display the change trend of the coking area ratio in the past 1 hour in real time, to help the operator intuitively master the coking development.

[0027] When the coking level reaches slight coking, the system issues a yellow early warning, and suggests that the operator strengthen the monitoring; when the coking level reaches moderate coking, the system issues an orange early warning, and prompts to adjust the boiler air distribution and reduce the furnace temperature to slow down the coking; when the coking level reaches severe coking, the system issues a red early warning, triggers an emergency shutdown suggestion signal, the control room receives a pop-up window prompt at the same time, and the device is automatically triggered to stop or clean, to avoid filter rupture or furnace accidents.

[0028] The coking identification and display method for the high-temperature filter of the boiler transparent flame of the embodiment of the application can be widely applied to the monitoring of the high-temperature filter of the boiler transparent flame in the power and thermal power industries, can not only improve the safety and economy of the boiler operation, realize the collaborative monitoring of the boiler combustion state and the filter coking condition, promote the intelligent upgrading of the equipment management in the power industry, and expand the application range of the method.

[0029] Device embodiment one According to the embodiment of the present application, a fire-through flame high-temperature filter coking identification display device of a thermal power boiler is provided, Figure 3 is a schematic diagram of the fire-through flame high-temperature filter coking identification display device of the thermal power boiler according to the embodiment of the present application, as Figure 3 shown, the fire-through flame high-temperature filter coking identification display device of the thermal power boiler according to the embodiment of the present application specifically comprises: The acquisition and preprocessing module 30 is configured to continuously and real-timely acquire images of the surface of the fire-through flame high-temperature filter through an industrial camera, and pre-process the images. The extraction and calculation module 32 is configured to perform feature extraction and classification on the pre-processed images by using a neural network model, determine a coking area, analyze distribution differences between a normal area and the coking area, determine coking distribution, and calculate a coking quantization result. The display module 34 is configured to display the coking distribution and the coking quantization result by using a three-dimensional heat map.

[0030] The embodiment of the present application is a device embodiment corresponding to the above-mentioned method embodiment, and the specific operations of each module can be understood with reference to the description of the method embodiment, which will not be described here.

[0031] Device embodiment two The embodiment of the present application provides an electronic device, as Figure 4 shown, comprising a memory 40, a processor 42, and a computer program stored on the memory 40 and executable on the processor 42, wherein the computer program is executed by the processor 42 to implement the steps as described in the method embodiment.

[0032] Device embodiment three The embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores an implementation program of information transmission, and the program is executed by the processor 42 to implement the steps as described in the method embodiment.

[0033] The computer-readable storage medium described in the embodiment includes but is not limited to ROM, RAM, magnetic disk or optical disk, etc.

[0034] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying and displaying the coking of a high-temperature filter in a boiler of a thermal power plant, characterized in that, The method comprises the following steps: Collecting images of the high-temperature flame-transmitting filter surface through an industrial camera in real time and continuously, and pre-processing the images; Using a neural network model to extract features and classify the pre-processed images, determine the coking area, analyze the distribution difference between the normal area and the coking area, determine the coking distribution, and calculate the coking quantification result; Using a three-dimensional heat map to display the coking distribution and the coking quantification result.

2. The method of claim 1, wherein, The specific steps of collecting images of the high-temperature flame-transmitting filter surface through an industrial camera in real time and continuously include: Collecting images of the high-temperature flame-transmitting filter surface through an industrial camera in real time and continuously, and pre-processing the images; 3. The method of claim 1, wherein, The specific steps of pre-processing the images include: Using median filtering to remove salt and pepper noise in the image, using wavelet threshold denoising to process Gaussian noise, decomposing the image into low-frequency coefficients, medium-frequency coefficients and high-frequency coefficients, and reconstructing the image after soft threshold processing of the high-frequency coefficients; performing gray scale normalization to map the image gray scale value to the 0-255 interval, and eliminating the influence of light changes on the image gray scale features.

4. The method of claim 1, wherein, The specific steps of using a neural network model to extract features and classify the pre-processed images, determine the coking area, analyze the distribution difference between the normal area and the coking area, and determine the coking distribution include: Using a neural network model, using image denoising, Otsu threshold segmentation and morphological dilation method to extract features and classify the pre-processed images, and through rotation, scaling, and flipping operations to enhance data diversity, and normalizing the image to determine the coking area, using Canny edge detection algorithm to set high threshold, medium threshold and low threshold, extract the edge profile of the coking area, distinguish coking and slight stains on the filter surface, analyze the distribution difference between the normal area and the coking area, and determine the coking distribution.

5. The method of claim 1, wherein, The specific steps of using a three-dimensional heat map to display the coking distribution and the coking quantification result include: Displaying the original image collected in real time on the filter surface in the left area of the display interface of the computer terminal; displaying the clear image after pre-processing in the middle area, and labeling the coking area, and superimposing the contour line of the coking area; displaying the coking quantification result in the right area, and displaying the change trend of the coking area ratio in the past predetermined time in a line chart in real time.

6. The method of claim 1, wherein, The coking quantification result includes coking area ratio data and coking thickness data.

7. The method of claim 1, wherein, The method further specifically includes: When the coking level reaches slight coking, the system issues a yellow warning to prompt the operator to strengthen monitoring; when the coking level reaches moderate coking, an orange warning is issued to prompt adjustment of the boiler air distribution and reduction of the furnace temperature to slow down coking; when the coking level reaches severe coking, a red warning is issued to trigger an emergency shutdown suggestion signal, and the control room receives a pop-up window prompt at the same time, and automatically triggers the equipment shutdown or cleaning program.

8. A device for identifying and displaying the coking of a high-temperature filter of a boiler of a thermal power plant, characterized in that, The method comprises the following steps: Collecting images of the high-temperature flame-transmitting filter surface through an industrial camera in real time and continuously, and pre-processing the images; The extraction calculation module is configured to use the neural network model to perform feature extraction and classification on the preprocessed image, determine a coking area, analyze distribution differences between a normal area and the coking area, determine coking distribution, and calculate a coking quantification result. The display module is configured to display the coking distribution and the coking quantification result by using a three-dimensional heat map.

9. An electronic device, comprising: The method comprises the following steps: The memory, the processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is executed by the processor to implement the steps of the method for identifying and displaying coking of a transparent flame high-temperature filter of a thermal power boiler according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an implementation program of information transmission, and the program is executed by the processor to implement the steps of the method for identifying and displaying coking of a transparent flame high-temperature filter of a thermal power boiler according to any one of claims 1 to 7.

Images