Method for preventing corrosion of the inner wall of a water tank of a power plant

Abstract

The application relates to the technical field of corrosion prevention treatment, and particularly discloses a power plant water tank inner wall corrosion prevention method applied to a water tank inner wall corrosion prevention process, which comprises the following steps: obtaining image data of the water tank inner wall, performing inner wall corrosion degree detection on the image data based on deep learning; performing corrosion prevention treatment on the water tank inner wall based on the water tank inner wall corrosion prevention process according to the inner wall corrosion degree detection result; obtaining the corrosion depth value of the water tank after the corrosion prevention treatment, and performing life assessment on the water tank according to the corrosion depth value of the water tank; obtaining the water quality degradation degree in the water tank, setting a corrosion inspection period according to the water quality degradation degree and the water tank life assessment value; when the corrosion inspection period is reached, obtaining the corrosion degree change curve of the water tank inner wall, and inspecting the corrosion prevention effect according to the corrosion degree change curve. The corrosion degree of the water tank inner wall is detected based on deep learning, the detection result is more accurate and fast, periodic detection is performed on the water tank inner wall after the corrosion prevention treatment, corrosion and damage of the water tank inner wall are effectively prevented, and the safety of power plant operation is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of corrosion prevention treatment, and more particularly to a method for preventing corrosion of the inner wall of a water tank in a power plant. BACKGROUND

[0002] The desalted water tank in a power plant is a metal tank that stores a large amount of desalted water, which is then distributed to various areas of the power plant. Since desalted water is the most widely used water source in the production process of a power plant, the water quality is required to be high. The water quality of the desalted water tank in a power plant directly affects the quality of subsequent work, and some corrosion prevention products also affect the normal operation of related production equipment. During the normal operation of the power plant, the desalted water tank needs to have a high service life because of the small maintenance window of the desalted water tank. Therefore, the corrosion prevention treatment of the inner wall of the water tank cannot be ignored during the normal production and operation of the power plant.

[0003] In the prior art, the corrosion prevention treatment of the inner wall of the water tank is only limited to the observation of the inspection personnel, and the measures are taken only when serious corrosion problems occur, which pollutes the water quality in the water tank and affects the production efficiency of the entire power plant. SUMMARY

[0004] The present application provides a method for preventing corrosion of the inner wall of a water tank in a power plant, which is applied to a corrosion prevention process of the inner wall of the water tank to solve the technical problem that the corrosion of the inner wall of the water tank in the power plant cannot be found in time in the prior art. The method comprises:

[0005] obtaining image data of the inner wall of the water tank, and detecting the corrosion degree of the inner wall based on deep learning;

[0006] performing corrosion prevention treatment on the inner wall of the water tank based on the corrosion prevention process of the inner wall of the water tank according to the corrosion degree detection result;

[0007] obtaining the corrosion depth value of the water tank after the corrosion prevention treatment, and evaluating the service life of the water tank according to the corrosion depth value of the water tank;

[0008] obtaining the water quality degradation degree in the water tank, and setting a corrosion inspection period according to the water quality degradation degree and the service life evaluation value of the water tank;

[0009] when the corrosion inspection period is reached, obtaining the corrosion degree change curve of the inner wall of the water tank, and inspecting the corrosion effect according to the corrosion degree change curve.

[0010] Further, the corrosion degree detection of the inner wall based on deep learning comprises:

[0011] obtaining sample data of the inner wall image of the water tank, converting the sample data of the inner wall image of the water tank into a gray image, and setting a gray threshold value based on a k-means clustering algorithm;

[0012] According to the gray threshold, the corrosion gray image is binarized;

[0013] The rust pixel points of the binarized corrosion gray image are extracted, and the corrosion gray image is labeled according to the rust pixel points;

[0014] The corrosion gray image after the label is subjected to data enhancement, and a training set is constructed according to the enhanced corrosion gray image;

[0015] According to the image data in the training set, a water tank inner wall corrosion degree detection model is trained, and the real-time data of the water tank inner wall image is detected based on the water tank inner wall corrosion degree detection model.

[0016] Further, the gray threshold is set based on the k-means clustering algorithm, which comprises:

[0017] Obtain the pixel point data of the corrosion gray image, and randomly select k initial clustering centers according to the pixel point data;

[0018] Calculate the Euclidean distance between each pixel point data and the clustering center, and cluster and divide the pixel point data according to the Euclidean distance between each pixel point data and the clustering center;

[0019] Obtain the average value of each partition data pixel point data after clustering and division, and update the clustering center according to the average value of the pixel point data;

[0020] Repeat the iteration and set a new clustering center until the clustering center no longer changes, and obtain the final clustering center, and set the pixel point gray average value of the final clustering center as the gray threshold.

[0021] Further, the corrosion treatment of the water tank inner wall based on the water tank inner wall corrosion degree detection result comprises:

[0022] Obtain the inner wall corrosion degree grade of the real-time data of the water tank inner wall image, if the inner wall corrosion degree grade is 70%-90%, the water tank inner wall is subjected to corrosion treatment;

[0023] If the inner wall corrosion degree grade is 10%-60%, the water tank inner wall is not subjected to corrosion treatment.

[0024] Further, the corrosion depth value of the water tank is used to evaluate the service life of the water tank, which comprises:

[0025] According to the corrosion depth after the corrosion treatment, the corrosion rate is calculated, the corrosion depth required to reach the thickness of the water tank is calculated according to the corrosion rate, and the corrosion depth required to reach the thickness of the water tank is determined as the service life evaluation value of the water tank.

[0026] Further, the corrosion inspection cycle is set according to the water quality degradation degree and the water tank service life evaluation value, and the corrosion inspection cycle comprises:

[0027] The water quality parameters of the water in the water tank are acquired, the water quality parameters comprising pH value, conductivity, total organic carbon content, and the water tank loss parameter is determined according to a water tank loss calculation formula;

[0028] A first preset threshold is set, the change curve of the water tank loss parameter in a first preset period is detected, the time required for the water tank loss parameter to reach the first preset threshold is predicted according to the change curve of the water tank loss parameter, and the time required for the water tank loss parameter to reach the first preset threshold is set as the corrosion inspection cycle.

[0029] Further, the water tank loss parameter determined according to the water tank loss calculation formula is:

[0030] F = α | p-7 | + βc + γt + δd

[0031] Wherein, F is the water tank loss parameter, α, β, γ, δ are water quality parameter weights, p is the pH value, c is the conductivity, t is the total organic carbon content, and d is the water tank service life evaluation value.

[0032] Further, the corrosion effect is inspected according to the corrosion degree change curve, and the corrosion effect comprises:

[0033] The change curve of the corrosion degree of the inner wall of the water tank in a second preset period is acquired, and the second preset period is divided into a plurality of sub-periods;

[0034] The average slope of the corrosion degree of the inner wall of the water tank in the sub-period is calculated, and whether the average slope exceeds a second preset threshold is judged;

[0035] If the average slope does not exceed the second preset threshold, it is determined that the corrosion treatment reaches the corrosion effect;

[0036] If the average slope exceeds the second preset threshold, it is determined that the corrosion treatment does not reach the corrosion effect.

[0037] The beneficial effects of the present application are:

[0038] By applying the above technical scheme, the present application adopts the image processing method to detect the corrosion degree of the image of the inner wall of the water tank collected in the power plant based on deep learning, the detection time is short, and the accuracy of the corrosion degree detection is greatly improved, the service life of the water tank is evaluated, the corrosion inspection cycle is set in combination with the water quality degradation degree, and the corrosion effect is periodically detected based on the corrosion inspection cycle, the corrosion effect is detected in time and accurately evaluated, the corrosion and damage of the inner wall of the water tank are effectively prevented, the safety of the power plant operation is ensured, and the operation efficiency of the power plant is improved. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced. Obviously, the drawings in the following description only represent some of the embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort.

[0040] Figure 1 A flowchart of a method for preventing corrosion of the inner wall of a water tank in a power plant is shown. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments only represent some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort fall within the scope of protection of the present application.

[0042] The embodiments of the present application provide a method for preventing corrosion of the inner wall of a water tank in a power plant, as shown in Figure 1 The method is applied to a corrosion prevention process for the inner wall of a water tank and includes the following steps.

[0043] S101, obtaining image data of the inner wall of the water tank, and detecting the corrosion degree of the inner wall based on deep learning.

[0044] In the present embodiment, a high-definition camera is arranged above the water tank in the power plant to collect image data of the inner wall of the water tank, and the image collection data of the inner wall of the historical water tank is used as sample data.

[0045] In some embodiments of the present application, the detection of the corrosion degree of the inner wall based on deep learning includes: obtaining sample data of the image of the inner wall of the water tank, converting the sample data of the image of the inner wall of the water tank into a grayscale image, and setting a grayscale threshold based on a k-means clustering algorithm; performing binaryzation processing on the corrosion grayscale image according to the grayscale threshold; extracting rust pixel points of the corrosion grayscale image after the binaryzation processing, and marking the corrosion grayscale image according to the rust pixel points; performing data enhancement on the corrosion grayscale image after the marking, constructing a training set according to the enhanced corrosion grayscale image; training a corrosion degree detection model of the inner wall of the water tank according to the image data in the training set, and detecting real-time data of the image of the inner wall of the water tank based on the corrosion degree detection model of the inner wall of the water tank.

[0046] In this embodiment, the gray value of the pixel point greater than the gray threshold value in the gray image is set to 255, the gray value of the pixel point less than or equal to the gray threshold value in the gray image is set to 0, the corrosion gray image is binarized according to the gray threshold value, the pixel point with the gray value of 0 is determined as a rust pixel point, the corrosion gray image is labeled according to the proportion of the rust pixel point in the corrosion gray image, and the corrosion gray image is divided into 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90% nine categories according to the proportion of the rust pixel point in the corrosion gray image.

[0047] In some embodiments of the present application, the k-means clustering algorithm is used to set the gray threshold value, including: obtaining pixel point data of the corrosion gray image, randomly selecting k initial clustering centers according to the pixel point data; calculating the Euclidean distance between each pixel point data and the clustering center, and clustering and dividing the pixel point data according to the Euclidean distance between each pixel point data and the clustering center; obtaining the average value of the pixel point data of each partition data after clustering and division, and updating the clustering center according to the average value of the pixel point data; repeating iteration and setting a new clustering center until the clustering center no longer changes, obtaining the final clustering center, and setting the average value of the pixel point gray of the final clustering center as the gray threshold value.

[0048] In this embodiment, the k-means clustering algorithm is used to set the gray threshold value, and the value of k is set to 8.

[0049] S102, based on the inner wall corrosion degree detection result, the inner wall of the water tank is subjected to corrosion prevention treatment based on the water tank inner wall corrosion prevention process.

[0050] In this embodiment, the water tank inner wall corrosion prevention process specifically includes:

[0051] S1021, polishing the tank body. The surface oxide scale, dust, and oil stains are cleaned, and uneven parts are trimmed. The surface treatment is dry, dust-free, oil-free, and free of other contaminants including soluble salts. The rusted parts on the surface are polished to make the surface present a metallic luster, and the roughness of the surface after polishing and rust removal reaches Ry=40-80 μm.

[0052] S1022, spraying an aluminum powder paint coating. After the polishing acceptance, the aluminum powder paint is sprayed using compressed air, and the thickness is controlled at 80-90 μm, and the dry film thickness is controlled at 40-50 μm.

[0053] S1023, three cloth and five oil coating. After the aluminum powder paint coating is dry, glass cloth + epoxy resin + asphalt paint is used to construct according to the three cloth and five oil construction standard.

[0054] S1024, rubber paint coating. After the three cloth five oil coating is dry, spray the chlorinated rubber anticorrosive finish. The rubber paint layer dry film thickness is controlled at 100-120 μm.

[0055] In some embodiments of the present application, the anticorrosion treatment of the inner wall of the water tank based on the corrosion degree detection result of the inner wall includes: obtaining the corrosion degree grade of the real-time data of the inner wall image of the water tank, if the corrosion degree grade is 70%-90%, the inner wall of the water tank is treated for corrosion protection; if the corrosion degree grade is 10%-60%, the inner wall of the water tank is not treated for corrosion protection.

[0056] In the present embodiment, when the corrosion degree of the inner wall of the water tank is the first type, it indicates that the corrosion condition of the image acquisition point of the inner wall of the water tank has reached 80%, and needs to be treated for corrosion protection, when the corrosion degree of the inner wall of the water tank is the second type or the third type, it indicates that the corrosion condition of the image acquisition point of the inner wall of the water tank is general, and does not need to be treated for corrosion protection.

[0057] S103, obtaining the corrosion depth value of the water tank after the anticorrosion treatment, and evaluating the service life of the water tank according to the corrosion depth value of the water tank;

[0058] In the present embodiment, the corrosion depth is the difference between the initial thickness and the current remaining thickness of the water tank, and the corrosion depth is used to represent the corrosion degree of the water tank.

[0059] In some embodiments of the present application, the service life evaluation of the water tank according to the corrosion depth value of the water tank includes: calculating the corrosion rate according to the corrosion depth after the anticorrosion treatment, calculating the time required for the corrosion depth to reach the thickness of the water tank according to the corrosion rate, and determining the time required for the corrosion depth to reach the thickness of the water tank as the service life evaluation value of the water tank.

[0060] In the present embodiment, the corrosion rate is calculated by recording the time used by the water tank from the initial thickness to the current remaining thickness, and the time required for the corrosion depth to reach the thickness of the water tank is obtained by the corrosion rate.

[0061] S104, obtaining the water quality degradation degree in the water tank, and setting the anticorrosion inspection period according to the water quality degradation degree and the service life evaluation value of the water tank;

[0062] In some embodiments of the present application, the anticorrosion inspection period is set according to the water quality degradation degree and the service life evaluation value of the water tank, which includes: obtaining the water quality parameters of the water in the water tank, the water quality parameters including pH value, conductivity, and total organic carbon content, determining the water loss parameter according to the water tank loss calculation formula; setting a first preset threshold, detecting the change curve of the water loss parameter in the first preset period, predicting the time required for the water loss parameter to reach the first preset threshold according to the change curve of the water loss parameter, and setting the time required for the water loss parameter to reach the first preset threshold as the anticorrosion inspection period.

[0063] In the embodiment, the water quality parameter is taken as the water quality deterioration degree, and the water quality parameter is combined with the water tank life evaluation value to set the water tank loss calculation formula. The first preset threshold is set according to the historical operation data of the power plant in combination with the minimum water quality parameter of the water tank, the first preset period is 90 days, the change curve of the water tank loss parameter in the first preset period is fitted as a linear regression model, and the time required for the water tank loss parameter to reach the first preset threshold is calculated through the linear regression model.

[0064] In some embodiments of the present application, the water tank loss parameter is determined according to the water tank loss calculation formula as follows:

[0065] F = a | p - 7 | + b c + g t + d d

[0066] Wherein, F is the water tank loss parameter, a, b, g, d are the water quality parameter weights, p is the pH value, c is the conductivity, t is the total organic carbon content, and d is the water tank life evaluation value.

[0067] In the embodiment, the water quality parameter weights are set according to historical operation experience, and the water quality parameters of the water in the water tank are tested by the water quality testing device.

[0068] S105, when the anticorrosion inspection period is reached, the corrosion degree change curve of the inner wall of the water tank is obtained, and the anticorrosion effect is inspected according to the corrosion degree change curve.

[0069] In the embodiment, the anticorrosion inspection period is counted after the anticorrosion treatment is completed,

[0070] In some embodiments of the present application, the anticorrosion effect is inspected according to the corrosion degree change curve, including: obtaining the change curve of the corrosion degree of the inner wall of the water tank in a second preset period, dividing the second preset period into a plurality of sub-periods; calculating the average slope of the corrosion degree of the inner wall of the water tank in the sub-periods, and judging whether the average slope exceeds a second preset threshold; if the average slope does not exceed the second preset threshold, it is determined that the anticorrosion treatment reaches the anticorrosion effect; if the average slope exceeds the second preset threshold, it is determined that the anticorrosion treatment does not reach the anticorrosion effect.

[0071] In the embodiment, the change curve of the corrosion degree category of the inner wall of the water tank in the second preset period is drawn, the second preset period is set to 100 days, the second preset period is evenly divided into 10 sub-periods, the average slope represents the corrosion mutation value of the inner wall of the water tank, if the average slope exceeds the second preset threshold, it means that there is a corrosion mutation value, and the anticorrosion treatment has defects, if not, it means that the corrosion degree of the inner wall of the water tank changes smoothly, and the anticorrosion effect is good.

[0072] By applying the above technical solutions, the present application realizes the following effects: the gray scale image of the inner wall of the water tank is obtained according to image processing, the gray scale threshold is set through the k-means clustering algorithm, and the image is binarized, the gray scale image sample data after binarization is trained based on deep learning, the water tank inner wall corrosion degree detection model is obtained, the real-time image data is subjected to inner wall corrosion degree detection, the inner wall of the water tank is subjected to corrosion prevention treatment based on the detection result, the corrosion rate is calculated through the corrosion depth value of the water tank after the corrosion prevention treatment, and the time required for the corrosion depth value to reach the thickness of the water tank is calculated according to the corrosion rate, the time is set as the water tank life evaluation value, the corrosion prevention inspection cycle is calculated according to the water tank life evaluation value combined with the water quality parameter, and the corrosion prevention effect is inspected according to the corrosion degree change curve of the inner wall of the water tank after the corrosion prevention inspection cycle is reached. Through the method adopted in the embodiment, the corrosion degree detection result of the inner wall of the water tank of the power plant is more accurate and fast, the corrosion and damage of the inner wall of the water tank are effectively prevented, the corrosion prevention treatment effect is accurately evaluated, and the safety of the operation of the power plant is ensured.

[0073] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, or can be implemented by means of software and a necessary general hardware platform. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.), and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0074] 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 they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for preventing corrosion of the inner wall of a water tank of a power plant, applied to a process for preventing corrosion of the inner wall of a water tank, characterized in that, The method comprises: acquiring image data of the inner wall of the water tank, and detecting the corrosion degree of the inner wall of the water tank based on deep learning; performing corrosion protection treatment on the inner wall of the water tank based on the corrosion protection process of the inner wall of the water tank according to the corrosion degree detection result of the inner wall; acquiring the corrosion depth value of the water tank after the corrosion protection treatment, and evaluating the service life of the water tank according to the corrosion depth value of the water tank; acquiring the water quality degradation degree in the water tank, and setting a corrosion inspection period according to the water quality degradation degree and the service life evaluation value of the water tank; when the corrosion inspection period is reached, acquiring the corrosion degree change curve of the inner wall of the water tank, and inspecting the corrosion effect according to the corrosion degree change curve; the setting of the corrosion inspection period according to the water quality degradation degree and the service life evaluation value of the water tank comprises: acquiring water quality parameters of the water in the water tank, the water quality parameters comprising pH value, conductivity, and total organic carbon content, and determining the water tank loss parameter according to a water tank loss calculation formula; setting a first preset threshold, detecting the change curve of the water tank loss parameter in a first preset period, predicting the time required for the water tank loss parameter to reach the first preset threshold according to the change curve of the water tank loss parameter, and setting the time required for the water tank loss parameter to reach the first preset threshold as the corrosion inspection period; the determination of the water tank loss parameter according to the water tank loss calculation formula is: F = α |p-7| + βc + γt + δd wherein F is the water tank loss parameter, α, β, γ, and δ are water quality parameter weights, p is the pH value, c is the conductivity, t is the total organic carbon content, and d is the service life evaluation value of the water tank.

2. The method of claim 1, wherein the water tank is a power plant water tank. the corrosion degree detection of the inner wall of the water tank based on deep learning comprises: acquiring sample data of the inner wall image of the water tank, converting the sample data of the inner wall image of the water tank into a grayscale image, and setting a grayscale threshold based on a k-means clustering algorithm; performing binaryzation processing on the corrosion grayscale image according to the grayscale threshold; extracting rust pixel points of the binaryzation-processed corrosion grayscale image, and performing category labeling on the corrosion grayscale image according to the rust pixel points; performing data enhancement on the category-labeled corrosion grayscale image, and constructing a training set according to the enhanced corrosion grayscale image; training a water tank inner wall corrosion degree detection model according to the image data in the training set, and detecting real-time data of the inner wall image of the water tank based on the water tank inner wall corrosion degree detection model.

3. The method of claim 2, wherein the water tank is a power plant water tank. the setting of the grayscale threshold based on the k-means clustering algorithm comprises: acquiring pixel point data of the corrosion grayscale image, and randomly selecting k initial clustering centers according to the pixel point data; calculating the Euclidean distance between each pixel point data and the clustering center, and performing clustering division on the pixel point data according to the Euclidean distance between each pixel point data and the clustering center; acquiring the average value of the pixel point data of each partition data after clustering division, and updating the clustering center according to the average value of the pixel point data; repeating iteration and setting a new clustering center until the clustering center no longer changes, obtaining the final clustering center, and setting the average value of the pixel point grayscale of the final clustering center as the grayscale threshold.

4. The method of claim 1, wherein the water tank is a power plant water tank. the corrosion protection treatment on the inner wall of the water tank based on the corrosion protection process of the inner wall of the water tank according to the corrosion degree detection result of the inner wall comprises: The inner wall corrosion degree grade of the real-time data of the inner wall image of the water tank is obtained, if the inner wall corrosion degree grade is 70%-90%, the inner wall of the water tank is treated by corrosion protection; If the inner wall corrosion degree grade is 10%-60%, the inner wall of the water tank is not treated by corrosion protection.

5. The method of claim 1, wherein the water tank is a power plant water tank. The service life of the water tank is evaluated according to the corrosion depth value of the water tank, which comprises: The corrosion rate is calculated according to the corrosion depth after the corrosion protection treatment, the time required for the corrosion depth to reach the thickness of the water tank is calculated according to the corrosion rate, and the time required for the corrosion depth to reach the thickness of the water tank is determined as the service life evaluation value of the water tank.

6. The method of claim 1, wherein the water tank is a power plant water tank. The corrosion effect is tested according to the corrosion degree change curve, which comprises: The change curve of the inner wall corrosion degree of the water tank in the second preset period is obtained, and the second preset period is divided into multiple sub-periods; The average slope of the inner wall corrosion degree of the water tank in the sub-period is calculated, and whether the average slope exceeds the second preset threshold is judged; If the average slope does not exceed the second preset threshold, it is determined that the corrosion protection treatment reaches the corrosion protection effect; If the average slope exceeds the second preset threshold, it is determined that the corrosion protection treatment does not reach the corrosion protection effect.

Images