Method for identifying production efficiency of steam increasing machine

Abstract

The invention relates to the technical field of image processing, in particular to a steam increasing machine production efficiency identification method. According to the method, the abnormal degree of each position is calculated according to different temperature distributions in infrared thermodynamic diagrams shot at the same position of the surface of a Venturi tube of the steam increasing machine at different angles due to different radiation quantities, the disorder degree of the temperature distributions in the connected domains and the fluctuation degree of the area of the connected domains, and the production efficiency of the steam increasing machine is judged. According to the scheme, the abnormal position can be found more accurately, and maintenance personnel can repair or maintain the abnormal position quickly.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to a method for identifying production efficiency of a steam booster. Background Art

[0002] In the traditional field of steam booster production efficiency detection, the evaluation of whether the performance and efficiency of the steam booster are abnormal mainly relies on a variety of sensor technologies and infrared thermal imaging equipment. These detection methods collect key parameters during the operation of the steam booster, such as temperature distribution, pressure fluctuations, and flow, to achieve comprehensive monitoring of the working status of the steam booster. In particular, the temperature distribution on the surface of the Venturi tube, as a core component of the steam booster, is regarded as an important indicator for judging the overall efficiency of the steam booster. Through the precise analysis of the temperature distribution on the surface of the Venturi tube, it is possible to effectively identify whether the steam booster has abnormal production efficiency.

[0003] However, in actual operation, since the venturi tube design of the steam turbine usually adopts two quasi-conical structures, this special shape leads to significant differences in the radiation amount of the venturi tube surface at different shooting angles. Specifically, even at the same position, the infrared thermal images obtained at different shooting angles will show different temperature distribution characteristics. This difference in temperature distribution caused by angle changes has brought great trouble to subsequent image processing and data analysis.

[0004] In particular, in the image processing stage, different manifestations of temperature distribution often lead to increased errors in the connected domain extraction algorithm. Connected domain extraction is a key step in image processing, which is used to identify and distinguish different temperature areas or feature areas in the image. However, in the infrared thermal image analysis of the steam booster venturi tube, due to the above-mentioned angle dependence problem, the accuracy of connected domain extraction is greatly reduced, which may lead to the misidentification of abnormal production efficiency of the steam booster. This misidentification not only affects the accuracy of production efficiency detection, but also may mislead subsequent troubleshooting and maintenance strategies, bringing potential risks to the safe operation and energy efficiency management of the steam booster.

[0005] In view of the above problems, an innovative solution is urgently needed, which aims to overcome the temperature distribution recognition error caused by changes in shooting angles by improving image processing technology, thereby improving the accuracy and reliability of abnormal identification of steam turbine production efficiency. Summary of the invention

[0006] The purpose of the present invention is to provide a method for identifying the production efficiency of a steam booster, so as to solve the problem of abnormal misidentification of the production efficiency of a steam booster due to different temperature distributions in infrared thermal images taken at different angles at the same position on the surface of the venturi tube of the steam booster, resulting in extraction errors of connected domains due to different radiation amounts in infrared thermal images taken at different angles. To this end, the present invention provides the following technical solutions.

[0007] The present invention provides a method for identifying the production efficiency of a steam booster, comprising: Obtain infrared thermal images of any position on the surface of the venturi tube of the steam booster at multiple angles at different times; The probability that a pixel point in each infrared thermogram belongs to an edge at each moment is calculated, and the probability is positively correlated with the gradient amplitude of any pixel point in the corresponding infrared thermogram in any of the three channels R, G, and B, and the importance of any channel; the importance represents the difference change between the pixel values ​​of the pixel points in the target area and the pixel points in the background area in any channel in the historical infrared thermogram of each historical abnormal production; If the probability is greater than a preset threshold, the corresponding pixel point belongs to an edge pixel point, and a connected domain of the corresponding infrared thermal map is obtained based on all edge pixels, and the overlapping area of ​​the connected domains at all angles at each moment is taken as the target connected domain at the corresponding moment; Calculate the degree of abnormality at any position; the degree of abnormality is positively correlated with the degree of temperature distribution disorder of the target connected domain at each moment, and negatively correlated with the area of ​​the corresponding target connected domain, and the degree of temperature distribution disorder represents the degree of fluctuation of the temperature gradient of each pixel point in the target connected domain; If the abnormality degree is greater than a threshold value, the production efficiency of the steam booster is abnormal.

[0008] The above scheme obtains infrared thermal images of multiple angles at different times at any position on the surface of the venturi tube of the steam booster to calculate the probability that the pixel point in each infrared thermal image at each time belongs to the edge. The probability is positively correlated with the gradient amplitude of any pixel point in any of the three channels R, G, and B in the corresponding infrared thermal image and the importance of any channel, wherein the importance represents the difference change between the pixel values ​​of the pixel points in the target area and the pixel points in the background area in any channel in the historical infrared thermal images of each historical abnormal production. If the probability is greater than a preset threshold, the corresponding pixel point belongs to the edge pixel point, and the connected domain of the corresponding infrared thermal image is obtained based on all edge pixels, and the overlapping area of ​​the connected domains at all angles obtained at each time is used as the target connected domain at the corresponding time, and the abnormality degree of any position is calculated, wherein the abnormality degree is positively correlated with the degree of temperature distribution disorder of the target connected domain at each time, and negatively correlated with the area of ​​the corresponding target connected domain, and the degree of temperature distribution disorder represents the degree of fluctuation of the temperature gradient of each pixel point in the target connected domain; if the abnormality degree is greater than the threshold, the production efficiency of the steam booster is abnormal. Therefore, the solution of the present invention can find the abnormal position more accurately, allowing maintenance personnel to quickly repair or maintain the abnormal position.

[0009] Optionally, the abnormality degree is specifically: ; Among them, Q cIndicates the abnormality of the cth position, M c represents the total number of times the infrared thermal image is taken at the cth position, S k,c represents the area of ​​the target connected domain at the c-th position at the k-th time, represents the average area of ​​the target connected domain at the cth position at all times, H k It indicates the degree of disorder of temperature distribution of the target connected domain at the kth moment. exp( ) is an exponential function with the natural constant e as the base.

[0010] The above scheme provides an important reference for monitoring the production efficiency of the steam turbine by calculating the degree of abnormality.

[0011] Optionally, the importance level specifically includes: ; Among them, D i represents the importance of channel i; m represents the number of historical infrared thermal images of historical abnormal production collected, a i,j The average pixel value of all pixels in the target area on channel i in the jth historical infrared thermal map representing historical abnormal production, b i,j The target area is the abnormal temperature distribution area obtained by segmenting the j-th historical infrared thermogram, and the background area is the normal temperature distribution area obtained by segmenting the j-th historical infrared thermogram.

[0012] The above scheme provides a method to accurately calculate the importance of RGB channels.

[0013] Optionally, the probability is specifically: ; Among them, P is the probability that the pixel belongs to the edge point, D i Indicates the importance of channel i, F i is the gradient amplitude of the pixel in channel i, max(F) is the maximum value of the gradient amplitude, and channel i is any of the three channels R, G, and B.

[0014] The above scheme extracts the connected domain by calculating the probability that the pixel point belongs to the edge pixel point, and can obtain an accurate connected domain.

[0015] Optionally, the degree of disorder of the temperature distribution is specifically: ; Among them, H k represents the degree of temperature distribution disorder of the target connected domain at the kth moment, N represents the collection of different directions, It represents the standard deviation of the temperature gradient in the lth direction of any pixel in the target connected domain, and norm() is the standardization function.

[0016] The above scheme obtains the temperature distribution on the surface of the venturi tube of the steam booster by calculating the degree of temperature disorder, thereby providing an important reference for monitoring the production efficiency of the steam booster.

[0017] Optionally, the degree of disorder of the temperature distribution is obtained by calculating the entropy of the temperature gradient of each pixel in the target connected domain.

[0018] Optionally, the method further includes: using Gaussian filtering to perform denoising on each infrared thermal image.

[0019] Optionally, after determining that the production efficiency of the steam booster is abnormal, an abnormality report is generated, wherein the abnormality report includes the abnormality location, the abnormality degree and recommended inspection and maintenance measures.

[0020] Optionally, after determining that the production efficiency of the steam booster is abnormal, the method further includes: A database is established and maintained for storing the infrared thermal map, the calculation results of the importance of the RGB channels, the abnormality analysis data and the inspection and maintenance records.

[0021] Optionally, the historical infrared thermodynamic map of historical abnormal production is an infrared thermodynamic map of historical steam booster turbines producing steam in abnormal amounts per unit time.

[0022] The beneficial effects of the present invention are: The solution of the present invention calculates the channel degrees of different channels of pixel points on the historical infrared thermogram produced using historical anomalies, and analyzes the current infrared thermogram based on the channel degrees to extract the connected domain, and then integrates the connected domains under multiple angles to determine the final target connected domain. By analyzing the target connected domain, the abnormal position can be found more accurately, allowing maintenance personnel to quickly repair or maintain the abnormal position. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A flowchart of the steps of a method for identifying production efficiency of a steam booster in this embodiment is schematically shown. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] It should be understood that the order of execution of the steps in the following embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0026] The present invention is directed to a steam booster production efficiency detection scheme, which requires accurate identification of the steam booster production efficiency by using image processing technology.

[0027] Therefore, based on the above content, the present invention provides a method and system for identifying the production efficiency of a steam booster to accurately detect the production efficiency of the steam booster.

[0028] Specifically, Figure 1 As shown, a method for identifying the production efficiency of a steam turbine in this embodiment includes the following steps: Step S1: Obtain infrared thermal images of any position on the surface of the venturi tube of the steam booster at multiple angles at different times.

[0029] Specifically, the production efficiency of the steam booster is mainly measured by the temperature, pressure and flow rate of the output steam. When the steam booster is blocked or leaking, the flow rate will be reduced and the production efficiency will be abnormal. This is because the steam booster is blocked or leaking, and the main manifestation of the blockage or leakage of the steam booster is the abnormal temperature distribution on the surface of the venturi tube of the steam booster. Therefore, infrared thermal imaging equipment is placed at different angles on the surface of the venturi tube of the steam booster, and the infrared thermal imaging equipment is used to continuously shoot infrared thermal images of the multi-angle surface of the venturi tube of the steam booster. The infrared thermal image is denoised using Gaussian filtering, and the multi-angle infrared thermal image of the venturi tube surface and the temperature values ​​corresponding to the pixel points in the infrared thermal image can be obtained.

[0030] Among them, the infrared thermogram obtained is the current thermogram, and each moment corresponds to a set of multi-angle infrared thermograms. The multi-angle infrared thermogram of the surface of the venturi tube of the steam booster refers to an image representation of the temperature distribution measurement of the surface of the venturi tube in the steam booster from multiple angles using infrared thermal imaging technology. This thermogram can intuitively display the temperature distribution of the venturi tube under different working conditions, thereby helping to analyze the heat exchange performance of the venturi tube and possible heat loss or hot spot areas. The denoising processing parameters of the Gaussian filter can be adjusted according to the noise level requirements of the infrared thermogram.

[0031] In this embodiment, infrared thermal images of multiple positions on the surface of the steam booster venturi tube at multiple angles at different times are obtained, and the infrared thermal images of each position are preprocessed to obtain multiple infrared thermal images corresponding to each time, so as to analyze the abnormal temperature distribution on the surface of the steam booster venturi tube.

[0032] Step S2: Calculate the probability that a pixel in each infrared thermal map at each moment belongs to an edge. If the probability is greater than a preset threshold, the corresponding pixel belongs to an edge pixel. Based on all edge pixels, the connected domain of the corresponding infrared thermal map is obtained, and the overlapping area of ​​the connected domains at all angles at each moment is used as the target connected domain at the corresponding moment.

[0033] Specifically, due to the different radiation amounts of the steam booster venturi tube at different angles, the same position in the infrared thermal images of the steam booster venturi tube taken at different angles presents different temperatures and temperature distributions, resulting in local temperature fluctuations being mistakenly identified as steam flow blockage or steam booster leakage.

[0034] Therefore, in this embodiment, by obtaining the historical infrared thermograms of historical abnormal production, the importance of the pixels in the historical infrared thermograms on different channels is calculated, and then the gradient amplitude of each channel and the importance of each channel are used to obtain the edge pixels in each current infrared thermogram to obtain the connected domain of each image, and then the overlapping parts of the connected domains of each current infrared thermogram are used as the target connected domain.

[0035] It should be noted that the location of the acquired historical infrared thermal map is the same as the location of the current infrared thermal map of any location that needs to be detected.

[0036] In this embodiment, the importance of the pixels in the historical infrared thermal map in the three channels of R, G, and B is calculated, and the probability that each pixel in the current infrared thermal map belongs to an edge pixel is calculated based on each importance. If the probability is greater than a preset threshold, the corresponding pixel belongs to an edge pixel, and the edge pixel is used to obtain the connected domain corresponding to the current infrared thermal map.

[0037] The connected domain helps to identify and analyze the temperature distribution on the surface of the steam turbine. In the infrared thermal map, the connected domain usually represents a set of pixels with similar temperature values. These pixel sets are connected in sequence by continuous edge pixels to obtain a complete area.

[0038] In the present invention, by analyzing the connected domains in the infrared thermal image of the steam turbine, the area with abnormal temperature can be identified. The gradient amplitude of the RGB channel refers to the rate or intensity difference of the change of the pixel values ​​of the three channels of the image in red (R), green (G), and blue (B).

[0039] In this embodiment, since the infrared thermogram is an RGB image obtained by color mapping of temperature, the brightness difference of different colors may be small, resulting in the phenomenon that the connected domain is too large or too small when the traditional image connected domain extraction is based on the pixel value or pixel gradient of the pixel point. Therefore, the present invention calculates the importance of each channel based on the pixel points of the target area and the background area in the historical infrared thermogram of historical abnormal production to obtain the importance of the three channels, and obtains the probability of the pixel point being an edge pixel point based on the importance and the gradient amplitude on each channel.

[0040] The importance is calculated as: ; Among them, D i represents the importance of channel i; m represents the number of historical infrared thermal images of historical abnormal production collected, a i,j The average pixel value of all pixels in the target area on channel i in the jth historical infrared thermal map representing historical abnormal production, b i,j Represents the average pixel value of all pixels in the background area of ​​the jth historical infrared thermal map on channel i for historical abnormal production.

[0041] The target area is the The abnormal temperature distribution area is obtained by segmenting the historical infrared thermal map. The background area is the The normal temperature distribution area is obtained by segmenting the historical infrared thermal map. It shows the difference between the abnormally produced steam turbine connected area and the background area on channel i.

[0042] The above segmentation can use image segmentation method to divide the historical infrared thermal map into regions.

[0043] The probability that each pixel in the image belongs to an edge point is calculated using the gradient amplitude of each channel and the importance of each channel. The specific calculation process of the probability is: ; Among them, P is the probability that the pixel belongs to the edge point, D i Indicates the importance of channel i, F i is the gradient amplitude of the pixel in channel i, max(F) is the maximum value of the gradient amplitude, and channel i is any of the three channels R, G, and B.

[0044] After obtaining the probability of each pixel point and the edge pixel point, the pixel points whose probability of each pixel point and the edge pixel point is greater than the target threshold are taken as edge pixels to extract the corresponding connected domain in the current infrared thermal map.

[0045] Step S3: Calculate the abnormality degree of any position.

[0046] Specifically, since the abnormal temperature distribution when the steam booster is blocked or leaking usually manifests itself as an irregular gradient change from the center of the blockage or leakage to the surrounding areas, the present invention calculates the degree of temperature distribution disorder in the connected domain at each moment by obtaining the gradients of the pixel points in different directions in the target connected domain at each position, and calculates the degree of abnormality at the corresponding position according to the degree of temperature distribution disorder and area at each moment.

[0047] In this embodiment, the degree of temperature distribution disorder in the connected domain at each moment is calculated by the gradients of the pixels in the target connected domain in different directions, specifically: ; Among them, H k represents the degree of temperature distribution disorder of the target connected domain at the kth moment, N represents the collection of different directions, It represents the standard deviation of the temperature gradient of any pixel in the target connected domain in the lth direction. ( ) is the normalization function.

[0048] The directions in the direction collection can be multiple directions such as 0°, 45°, 90°, 135°, etc. The larger the value, the more irregular the temperature gradient change in the connected domain in that direction. Therefore, the greater the probability of blockage or anomaly.

[0049] Due to the uneven mixing of the power steam and the inhaled exhaust steam in the steam booster, a temporary abnormal temperature distribution occurs in some areas of the surface of the venturi tube of the steam booster. Using the infrared thermal image at a certain moment will cause the normal production state to be identified as abnormal. Therefore, the present invention calculates the abnormality degree of each position according to the temperature distribution disorder degree and area of ​​the target connected domain at each position. The calculation of the abnormality degree is: ; Q c Indicates the abnormality of the cth position, M c represents the total number of times the infrared thermal image is taken at the cth position, S k,c represents the area of ​​the target connected domain at the c-th position at the k-th time, represents the average area of ​​the target connected domain at the cth position at all times, H k It indicates the degree of disorder of temperature distribution in the target connected domain at the kth moment. The greater the degree of disorder of temperature distribution, the greater the degree of abnormality. k There is a case where it is 0, so 0.001 is introduced, and exp( ) is an exponential function with the natural constant e as the base.

[0050] It represents the fluctuation degree of the target connected domain area at the cth position at the kth moment. Since the blockage or leakage area has abnormal temperature distribution at each moment, the area change of the target connected domain is small, which means that the greater the probability of abnormality at the cth position, the greater the degree of abnormality.

[0051] In this embodiment, the gradients of the pixels in the target connected domain in different directions are calculated to evaluate the degree of disorder of the temperature distribution in the target connected domain. The degree of abnormality of each position on the surface of the venturi tube of the steam booster is calculated based on the degree of disorder of the temperature distribution in the target connected domain and the degree of area fluctuation in the target connected domain at each moment, so that the abnormality of the steam booster production efficiency can be identified according to the degree of abnormality.

[0052] Step S4: If the abnormality level is greater than a threshold, the production efficiency of the steam booster is abnormal.

[0053] Specifically, when the degree of abnormality at a position on the surface of the steam booster venturi tube is greater than a preset threshold, it indicates that blockage or leakage occurs at that position and the steam booster production efficiency is abnormal, and relevant personnel are arranged to perform inspection and maintenance; otherwise, the steam booster production efficiency is normal.

[0054] Optionally, after determining whether the production efficiency of the steam booster is abnormal, an abnormality report may be generated, wherein the abnormality report includes the abnormality location, the degree of abnormality and recommended repair and maintenance measures, so that the user can repair or regularly maintain the abnormal location.

[0055] Optionally, after determining whether the production efficiency of the steam turbine is abnormal, a database is established and maintained to store historical infrared thermal images, channel importance calculation results, abnormality analysis data, and inspection and maintenance records to facilitate subsequent analysis and improvement.

[0056] In this embodiment, the abnormality degree of each position on the surface of the venturi tube of the steam booster is obtained and compared with a preset threshold. When the abnormality degree of a position is greater than the preset threshold, an abnormal warning is issued to determine whether the production efficiency of the steam booster is normal.

Claims

1. A method for identifying the production efficiency of a steam turbine, characterized in that: include: Obtain infrared thermal images of any position on the surface of the venturi tube of the steam booster at multiple angles at different times; The probability that a pixel point in each infrared thermogram belongs to an edge at each moment is calculated, and the probability is positively correlated with the gradient amplitude of any pixel point in the corresponding infrared thermogram in any of the three channels R, G, and B, and the importance of any channel; the importance represents the difference change between the pixel values ​​of the pixel points in the target area and the pixel points in the background area in any channel in the historical infrared thermogram of each historical abnormal production; If the probability is greater than a preset threshold, the corresponding pixel point belongs to an edge pixel point, and a connected domain of the corresponding infrared thermal map is obtained based on all edge pixels, and the overlapping area of ​​the connected domains at all angles at each moment is taken as the target connected domain at the corresponding moment; Calculate the degree of abnormality at any position; the degree of abnormality is positively correlated with the degree of temperature distribution disorder of the target connected domain at each moment, and negatively correlated with the area of ​​the corresponding target connected domain, and the degree of temperature distribution disorder represents the degree of fluctuation of the temperature gradient of each pixel point in the target connected domain; If the abnormality degree is greater than a threshold value, the production efficiency of the steam booster is abnormal.

2. A method for identifying the production efficiency of a steam booster according to claim 1, characterized in that: The degree of abnormality is specifically: ; Among them, Q c Indicates the abnormality of the cth position, M c represents the total number of times the infrared thermal image is taken at the cth position, S k,c represents the area of ​​the target connected domain at the c-th position at the k-th time, represents the average area of ​​the target connected domain at the cth position at all times, H k It indicates the degree of disorder of temperature distribution of the target connected domain at the kth moment. exp( ) is an exponential function with the natural constant e as the base.

3. A method for identifying the production efficiency of a steam booster according to claim 1, characterized in that: The degree of importance specifically includes: ; Among them, D i represents the importance of channel i; m represents the number of historical infrared thermal images of historical abnormal production collected, a i,j The average pixel value of all pixels in the target area on channel i in the jth historical infrared thermal map representing historical abnormal production, b i,j The target area is the abnormal temperature distribution area obtained by segmenting the j-th historical infrared thermogram, and the background area is the normal temperature distribution area obtained by segmenting the j-th historical infrared thermogram.

4. A method for identifying steam booster production efficiency according to claim 1, characterized in that: The probability is specifically: ; Among them, P is the probability that the pixel belongs to the edge point, D i Indicates the importance of channel i, F i is the gradient amplitude of the pixel in channel i, max(F) is the maximum value of the gradient amplitude, and channel i is any of the three channels R, G, and B.

5. A method for identifying production efficiency of a steam booster according to claim 1, characterized in that: The degree of disorder of the temperature distribution is specifically: ; Among them, H k represents the degree of temperature distribution disorder of the target connected domain at the kth moment, N represents the collection of different directions, It represents the standard deviation of the temperature gradient in the lth direction of any pixel in the target connected domain, and norm() is the standardization function.

6. A method for identifying the production efficiency of a steam booster according to claim 1, characterized in that: The degree of disorder of the temperature distribution is obtained by calculating the entropy of the temperature gradient of each pixel in the target connected domain.

7. A method for identifying steam booster production efficiency according to claim 1, characterized in that: Also includes: The steps of using Gaussian filtering to denoise each infrared thermal image.

8. A method for identifying steam booster production efficiency according to claim 1, characterized in that: Also includes: After determining that the production efficiency of the steam booster is abnormal, an abnormality report is generated, wherein the abnormality report includes the abnormality location, the abnormality degree and the recommended repair and maintenance measures.

9. A method for identifying steam booster production efficiency according to claim 1, characterized in that: After judging that the production efficiency of the steam booster is abnormal, it also includes: A database is established and maintained for storing the infrared thermal map, the calculation results of the importance of the RGB channels, the abnormality analysis data and the inspection and maintenance records.

10. A method for identifying production efficiency of a steam booster according to claim 1, characterized in that: The historical infrared thermogram of historical abnormal production is an infrared thermogram of the historical steam booster turbine produced when the steam volume per unit time was abnormal.

Images