Coating amount prediction optimization method for coating application on surface of steel structure

Abstract

The invention relates to the technical field of data processing, in particular to a method for predicting and optimizing the coating amount applied to the surface of a steel structure, and the method comprises the following steps: according to the surface roughness of each test steel material before coating and the surface roughness of each coating layer after coating; constructing a surface roughness fitting function between the surface roughness and the number of the coating layers; according to the paint deviation value between the actual paint amount and the theoretical paint amount of each paint layer of each test steel material and the surface roughness of each paint layer, a paint deviation value fitting function between the paint deviation value and the surface roughness is constructed; constructing an overall coating deviation value function according to the surface roughness fitting function and the coating deviation value fitting function; according to the number of the coating layers of each to-be-coated steel material, the theoretical coating amount of each coating layer, the coating loss rate and the overall coating deviation value function, the coating predicted amount of each to-be-coated steel material is obtained, and the accuracy and reliability of predicting the coating amount are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for predicting and optimizing the amount of paint applied to the surface of a steel structure. Background Art

[0002] Steel structures are widely used in large buildings such as factories and stadiums due to their excellent mechanical properties and large internal spaces. However, the basic framework of steel structures has a relatively low resistance to natural disasters such as fire, high temperature and humidity. Therefore, specialized coatings are applied to the surface of steel structures to prevent corrosion, rust, and fire to address these hazards. During the coating application process, the amount of coating used not only affects the construction cost but also has a direct impact on the protective effect of the coating: insufficient coating may result in poor protection, while excessive coating will waste resources and increase costs. Therefore, how to accurately predict the amount of coating to be used is a key issue in the application of steel structure coatings.

[0003] Traditionally, paint usage predictions rely primarily on a combination of theoretical coverage and actual construction experience. Theoretical coverage refers to the area that a unit volume of paint can cover on a completely smooth, flat, and pore-free surface. However, in actual construction, the roughness of the steel structure surface, differences in spraying methods, and the type of paint used all affect the paint's adhesion and coating quality, and thus the amount of paint used. Traditional methods for predicting paint usage fail to account for factors such as surface roughness, differences in spraying methods, and the type of paint used, resulting in inaccurate predictions.

[0004] Therefore, how to improve the accuracy and reliability of coating dosage prediction has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a method for predicting and optimizing the amount of paint applied to the surface of a steel structure, so as to solve the problem of how to improve the accuracy and reliability of predicting the amount of paint used.

[0006] An embodiment of the present invention provides a method for predicting and optimizing the amount of coating applied to a steel structure surface, the method comprising the following steps:

[0007] Using a preset number of steel materials to be coated under a target type as test steel materials, obtaining the number of coating layers required for any test steel material under a preset scenario, as well as the theoretical coating amount of each coating layer;

[0008] Obtaining the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating to construct a first variation curve between surface roughness and the number of coating layers; constructing a second variation curve between the coating deviation value and surface roughness based on a coating deviation value between an actual coating amount and a theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer;

[0009] Fitting the first change curve and the second change curve of all test steel materials respectively to obtain a surface roughness fitting function and a coating deviation value fitting function, and constructing an overall coating deviation value function based on the surface roughness fitting function and the coating deviation value fitting function;

[0010] According to the number of coating layers of each steel material to be coated under the target type in the required usage scenario, the theoretical coating amount of each coating layer, the coating loss rate and the overall coating deviation value function, the predicted coating amount of each steel material to be coated in the required usage scenario is obtained.

[0011] Preferably, obtaining the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating comprises:

[0012] Before coating any of the test steel materials, grayscale images of the test steel material under three preset different illumination values are obtained, grayscale images under illumination values other than the maximum illumination value and the minimum illumination value are recorded as target images, and all noise pixels in the target image are obtained according to the grayscale value of each pixel in each of the grayscale images;

[0013] For any noise pixel, calculate the grayscale value mean of all non-noise pixels in the eight neighborhoods of the noise pixel, update the grayscale value mean to the grayscale value of the noise pixel, and update the grayscale value of each noise pixel in the target image to obtain a new target image.

[0014] Dividing the new target image into a preset number of sub-regions, and obtaining the local uniformity of each sub-region according to the distribution of the grayscale values of the pixels in each sub-region;

[0015] Calculating the average of the linearly normalized values of the local uniformity of all sub-regions to obtain a local uniformity mean, obtaining the standard deviation of the grayscale values of all pixels in the new target image, calculating the reciprocal of the sum of a constant 1 and the standard deviation to obtain an overall grayscale difference index of the new target image, and performing a weighted summation of the local uniformity mean and the overall grayscale difference index to obtain an overall uniformity of the new target image;

[0016] Subtract the overall uniformity from the constant 1 to obtain the surface roughness of any test steel material before coating;

[0017] According to the grayscale images of each coating layer of any test steel material after coating at the three preset different illumination values, the surface roughness of each coating layer of any test steel material after coating is obtained accordingly.

[0018] Preferably, obtaining all noise pixels in the target image according to the grayscale value of each pixel in each grayscale image includes:

[0019] Calculating a difference between the maximum illumination value and the minimum illumination value to obtain an illumination value span, calculating a first difference between the illumination value corresponding to the target image and the minimum illumination value, and using a ratio between the first difference and the illumination value span as an illumination value change ratio;

[0020] With the lower left corner of the target image as the origin, construct a two-dimensional rectangular coordinate system corresponding to the target image, obtain the coordinates of any pixel point in the target image according to the two-dimensional rectangular coordinate system, and obtain the pixel point with the same coordinates as the any pixel point in the grayscale image at the maximum illumination value and the minimum illumination value, respectively, to obtain the maximum illumination pixel point and the minimum illumination pixel point of the any pixel point;

[0021] Calculating a grayscale value difference between the maximum illumination pixel and the minimum illumination pixel to obtain a grayscale value span, calculating a second difference between the grayscale value of any pixel and the grayscale value of the minimum illumination pixel, and using a ratio between the second difference and the grayscale value span as a grayscale value change ratio;

[0022] Calculating the absolute value of the difference between the illumination value change ratio and the grayscale value change ratio, taking the reciprocal of the sum of the absolute value of the difference and a preset constant as a first variable, and subtracting the first variable from the constant 1 to obtain the noise probability of any pixel;

[0023] If the noise probability of any pixel is greater than or equal to a preset noise probability threshold, the pixel is marked as a noise pixel.

[0024] Preferably, obtaining the local uniformity of each sub-region according to the distribution of the grayscale values of the pixels in each sub-region includes:

[0025] For any sub-region, obtain the number of all pixels in the sub-region and the total grayscale value of all pixels in the sub-region;

[0026] Calculate the product of the grayscale value of any pixel point in any sub-region and the number of all pixel points, take the absolute value of the difference between the product and the total grayscale value as the grayscale value distribution characteristic value of any pixel point, calculate the average value of the grayscale value distribution characteristic values of all pixel points in any sub-region, and take the reciprocal of the result of adding the average value and the constant 1 as the local uniformity of any sub-region.

[0027] Preferably, obtaining a first variation curve between surface roughness and the number of coating layers comprises:

[0028] Obtaining an actual coating amount of each coating layer of any test steel material after coating, and obtaining a turning roughness from the surface roughness of all coating layers according to a coating deviation value between the actual coating amount and a theoretical coating amount of each coating layer of any test steel material;

[0029] The surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating are mapped into the same scatter plot, recorded as a first scatter plot, wherein the abscissa of the first scatter plot represents the number of coating layers of the any test steel material to be coated, and the ordinate represents the surface roughness. In the first scatter plot, the data between the surface roughness of the any test steel material before coating and the turning roughness are fitted to obtain a first sub-curve, and the data after the turning roughness are fitted to obtain a second sub-curve;

[0030] The first sub-curve and the second sub-curve are connected to obtain a first variation curve between surface roughness and the number of coating layers.

[0031] Preferably, the step of obtaining the turning roughness from the surface roughness of all coating layers according to the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material comprises:

[0032] Mapping the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material into the same scatter plot, wherein the abscissa of the scatter plot represents the number of coating layers of any to-be-coated steel material, and the ordinate represents the coating deviation value;

[0033] For any coating deviation value in the scatter plot, if the any coating deviation value is the first coating deviation value in the scatter plot, obtaining a slope between the any coating deviation value and its right adjacent coating deviation value, and subtracting the reciprocal of the slope from a constant 1 to obtain a turning rate of the any coating deviation value;

[0034] If any one of the coating deviation values is the last coating deviation value in the scatter plot, obtaining a slope between the any one of the coating deviation values and its left adjacent coating deviation value, and subtracting the reciprocal of the slope from a constant 1 to obtain a turning rate of the any one of the coating deviation values;

[0035] If any of the coating deviation values is not a value other than the first coating deviation value and the last coating deviation value in the scatter plot, then calculating a left slope between the any coating deviation value and its left adjacent coating deviation value, and a right slope between the any coating deviation value and its right adjacent coating deviation value, and subtracting the reciprocal of the absolute value of the difference between the left slope and the right slope from a constant 1 to obtain a turning rate of the any coating deviation value;

[0036] The inflection rates of all coating deviation values are obtained, the coating deviation value corresponding to the maximum value among all inflection rates is recorded as the inflection deviation value, and the surface roughness of the coating layer corresponding to the inflection deviation value is recorded as the inflection roughness.

[0037] Preferably, constructing a second variation curve between the coating deviation value and the surface roughness based on the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer includes:

[0038] constructing a second scatter plot based on the coating deviation value and surface roughness of each coating layer of any one of the test steel materials, wherein the abscissa of the second scatter plot represents the surface roughness of each coating layer, and the ordinate represents the coating deviation value of each coating layer;

[0039] In the second scatter plot, the data of any test steel material before the turning roughness are fitted to obtain a third sub-curve, and the data after the turning roughness are fitted to obtain a fourth sub-curve;

[0040] The third sub-curve is connected to the fourth sub-curve to obtain a second variation curve between the coating deviation value and the surface roughness.

[0041] Preferably, the first variation curve and the second variation curve of all the test steel materials are fitted respectively to obtain the surface roughness fitting function and the coating deviation value fitting function, including:

[0042] Fitting the first sub-curve in the first variation curve of all test steel materials to obtain a first sub-segmental function, fitting the second sub-curve in the first variation curve of all test steel materials to obtain a second sub-segmental function, combining the first sub-segmental function and the second sub-segmental function to form a surface roughness fitting function, wherein the independent variable of the first sub-segmental function is less than or equal to the number of coating layers corresponding to the turning roughness, and the independent variable of the second sub-segmental function is greater than the number of coating layers corresponding to the turning roughness;

[0043] The third sub-curve in the second change curve of all test steel materials is fitted to obtain a third sub-segmental function, and the fourth sub-curve in the second change curve of all test steel materials is fitted to obtain a fourth sub-segmental function. The third sub-segmental function and the fourth sub-segmental function are combined to form a coating deviation value fitting function, the independent variable of the third sub-segmental function is less than or equal to the turning roughness, and the independent variable of the fourth sub-segmental function is greater than the turning roughness.

[0044] Preferably, constructing an overall coating deviation value function based on the surface roughness fitting function and the coating deviation value fitting function comprises:

[0045] The output of the surface roughness fitting function serves as the input of the coating deviation value fitting function, and the output of the coating deviation value fitting function serves as the input of the overall coating deviation value function. The output of the overall coating deviation value function is the overall coating deviation value corresponding to the predicted coating of any steel material to be coated.

[0046] Preferably, obtaining the predicted amount of coating for each steel material to be coated in the desired usage scenario according to the number of coating layers for each steel material to be coated in the desired usage scenario under the target type, the theoretical coating amount of each coating layer, the coating loss rate, and the overall coating deviation value function includes:

[0047] For any steel material to be coated under the target type, the overall coating deviation value of any steel material to be coated is obtained according to the number of coating layers of any steel material to be coated in the required usage scenario and the overall coating deviation value function, the theoretical material amount of each coating layer is accumulated to obtain the final theoretical material amount of any steel material to be coated, the addition result of the final theoretical material amount and the overall coating deviation value is calculated, and the ratio of the addition result to the coating loss rate is used as the estimated coating amount of any steel material to be coated.

[0048] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0049] The present invention uses a preset number of steel materials to be coated under a target type as test steel materials, obtains the number of coating layers required for any test steel material under a preset scenario, and the theoretical coating amount of each coating layer; obtains the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating to construct a first change curve between surface roughness and the number of coating layers, and constructs a second change curve between the coating deviation value and surface roughness based on the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer; fits the first change curve and the second change curve of all test steel materials respectively to obtain a surface roughness fitting function and a coating deviation fitting function, and constructs an overall coating deviation function based on the surface roughness fitting function and the coating deviation fitting function; and obtains a predicted coating amount for each steel material to be coated under the required usage scenario based on the number of coating layers, the theoretical coating amount of each coating layer, the coating loss rate, and the overall coating deviation function for each steel material to be coated under the target type under the required usage scenario. Among them, the relationship between the surface roughness of each coating layer of the steel material to be coated and the number of coating layers (surface roughness fitting function), the relationship between the surface roughness of each coating layer and the coating deviation value (coating deviation fitting function), and the coating loss are obtained, so as to combine the influence of the surface roughness of the steel material on the coating amount, the influence of the coating construction method on the coating loss, and the inherent characteristics of the coating to comprehensively predict the coating amount of all steel materials to be coated under the target type, thereby improving the accuracy and reliability of the coating amount prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a flow chart of a method for predicting and optimizing the amount of paint applied to the surface of a steel structure, provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0052] The embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0053] It should be noted that the terms "first," "second," and the like in the specification of the present disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0054] In order to illustrate the technical solution of the present invention, specific embodiments are provided below.

[0055] See also Figure 1 , is a method flow chart of a method for predicting and optimizing the amount of coating applied to a steel structure surface provided by the first embodiment of the present invention, such as Figure 1 As shown, the method may include:

[0056] Step S101 : using a preset number of steel materials to be coated under a target type as test steel materials, obtaining the number of coating layers required for any test steel material under a preset scenario, and the theoretical coating amount of each coating layer.

[0057] Steel surfaces often require specialized coatings for corrosion, rust, and fire protection. The amount of coating applied not only affects construction costs but also directly impacts the protective effectiveness of the coating: insufficient coating can result in poor protection, while excessive coating wastes resources and increases costs. Therefore, accurately predicting coating application volume is a key issue in steel coating applications.

[0058] Traditionally, paint usage predictions rely primarily on a combination of theoretical coating coverage and actual application experience. Theoretical coating coverage refers to the area that a unit volume of paint can cover on a completely smooth, flat, and pore-free surface. However, in actual construction, factors such as the surface roughness of steel, variations in spraying methods, and the type of paint used can affect paint adhesion and coating quality, and thus paint usage. Traditional methods for predicting paint usage fail to account for these factors, resulting in inaccurate predictions.

[0059] Therefore, in an embodiment of the present invention, the amount of paint that may be used is comprehensively predicted based on the influence of the surface roughness of the steel material on the amount of paint, the influence of the paint application method on the loss of paint, and the characteristics of the paint itself, so as to improve the accuracy and reliability of the prediction of the paint usage.

[0060] First, take a steel material to be coated as an example, and use it as the target type. Then, 10% of the number of all steel materials to be coated under the target type that need to use the same coating is used as test steel materials. In the embodiment of the present invention, fire retardant coating is taken as an example, and there is no restriction here. The implementer can set the number of test steel materials according to the specific scenario to conduct a coating test on each test steel material. According to the test results of each test steel material, the coating amount of all steel materials to be coated under the target type that need to use the same coating is predicted. Then, the number of coating layers of fire-retardant coating required for the test steel material in a large space is obtained, and based on the dry film thickness and theoretical coating rate of the fire-retardant coating, the theoretical coating amount of each coating layer of the test steel material is obtained. Specifically: theoretical coating rate = given theoretical coating rate ÷ dry film thickness × designed dry film thickness, theoretical coating amount = area × theoretical coating rate, wherein the given theoretical coating rate is the theoretical coating rate of the fire-retardant coating, and the designed dry film thickness is the number of coating layers of fire-retardant coating required for the test steel material in a large space. According to the difference between the actual coating amount and the theoretical coating amount of each coating layer of the test steel material in the test coating process, the influence of the surface roughness of the steel material on the coating amount is analyzed to improve the accuracy and reliability of the prediction of the coating amount. Among them, the acquisition of the number of coating layers of fire-retardant coating required for the test steel material in a large space, the acquisition of the dry film thickness and the theoretical coating rate are all existing technologies and will not be repeated here.

[0061] Step S102: Obtain the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating to construct a first change curve between surface roughness and the number of coating layers; and construct a second change curve between the coating deviation value and surface roughness based on the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer.

[0062] Considering that in the process of coating steel materials, the coating of paint can fill the tiny bumps on the surface of the steel materials, making the surface of the steel materials smoother, thereby reducing the surface roughness of the steel materials, therefore, in the process of coating steel materials, the surface roughness of the steel materials will affect the adhesion of the paint, thereby resulting in a certain difference between the actual coating amount and the theoretical coating amount of each coating layer. On the other hand, the coating construction method and the coating's own characteristics will also cause a deviation between the actual coating amount and the theoretical coating amount of each coating layer: if a certain coating construction method results in uneven coating, or if there are particulate impurities in the coating itself, the surface roughness of the steel materials will be increased, resulting in a deviation between the actual coating amount and the theoretical coating amount of each coating layer. Based on the above analysis, during the process of coating steel materials, the surface roughness of the steel materials is a changing process: when the surface roughness of the steel materials is greater than the fineness of the coating, the coating will reduce the surface roughness of the steel materials; when the surface roughness of the steel materials is less than the fineness of the coating, the coating will increase the surface roughness of the steel materials. That is, as the surface roughness of each coating layer of the steel materials after coating is different, the difference between the actual coating amount and the theoretical coating amount of each coating layer is also different, that is, the surface roughness has a certain influence on the coating amount. Therefore, in an embodiment of the present invention, the relationship between the difference between the actual coating amount and the theoretical coating amount of each coating layer and the surface roughness is analyzed, and the coating amount required for all steel materials to be coated under the target type is predicted.

[0063] At the same time, since the number of coating layers required for the same type of steel materials in different scenarios is different, for example, buildings with different uses have different requirements for fire protection levels, the higher the fire protection level, the more coating layers are required, that is, the number of coating layers will also affect the accuracy of predicting the amount of coating required for all steel materials to be coated under the target type, and the difference between the actual coating amount and the theoretical coating amount of each coating layer needs to be obtained through experiments. In order to adapt to the prediction of the coating amount of steel materials in different scenarios and reduce the number of experiments, in an embodiment of the present invention, a coating test is performed on each test steel material to obtain the actual coating amount of each test steel material in each coating layer and the surface roughness of each coating layer. Then, based on the surface roughness of each test steel material in each coating layer and the difference between the actual coating amount and the theoretical coating amount of each coating layer, the relationship between the surface roughness and the number of coating layers is analyzed. , as well as the relationship between the difference between the actual coating amount and the theoretical coating amount of each coating layer and the surface roughness, are used to predict the difference between the actual coating amount and the theoretical coating amount under different numbers of coating layers. For example: assuming that the number of coating layers is 5 when the coating test is conducted on each test steel material, if the number of coating layers of a certain steel material to be coated under the target type under the required scenario is 8, then based on the relationship between the difference between the actual coating amount and the theoretical coating amount of each coating layer of the test steel material and the surface roughness, the difference between the actual coating amount and the theoretical coating amount of a certain steel material to be coated in the 6th, 7th, and 8th coating layers can be predicted respectively without conducting the test again, and then the coating usage of all steel materials to be coated under the target type can be predicted. While adapting to the needs of different scenarios, the number of experiments is reduced, and the accuracy and reliability of the coating usage prediction are improved.

[0064] In an embodiment of the present invention, taking the i-th test steel material as an example, a grayscale image of the i-th test steel material before coating is first obtained, so as to obtain the surface roughness of the i-th test steel material according to the uniformity of the grayscale value distribution in the grayscale image: the more uniform the grayscale value distribution in the grayscale image, the smaller the surface roughness of the i-th test steel material. Since there may be noise interference in the process of obtaining the grayscale image of the i-th test steel material, which may lead to inaccurate results of the surface roughness obtained based on the grayscale image of the i-th test steel material, considering that the luminous flux irradiated by the light source to the surface of the test steel material is the illumination, the luminous flux displayed on the surface of the test steel material is the brightness, and the brightness is expressed as different grayscale values after grayscale processing, the grayscale value of normal pixel points will increase with the increase of the light source illumination, while the noise point will not. Therefore, in an embodiment of the present invention, three light sources with different illumination values are respectively used to vertically illuminate the i-th test steel. For the same surface area of the material, grayscale images at three different illumination values are obtained to obtain the surface roughness of the i-th test steel material based on the grayscale images at different illumination values while reducing noise interference. Regarding the setting of illumination values, low illumination is usually 50-200 lux, medium illumination is 200-500 lux, and high illumination is 500-1000 lux. Therefore, in the embodiment of the present invention, three different illumination values are set to 200 lux, 500 lux, and 1000 lux, respectively. This is not limited here, and the implementer can set it according to the specific scenario. Then, based on the grayscale images of the i-th test steel material at three different illumination values before coating, the specific method for obtaining the surface roughness of the i-th test steel material is as follows:

[0065] (1) The grayscale image under the illumination values other than the maximum illumination value and the minimum illumination value is recorded as the target image. According to the grayscale value of each pixel point in each grayscale image corresponding to the i-th test steel material, all the noise pixels in the target image are obtained, and the grayscale value of each noise pixel point in the target image is updated to obtain a new target image.

[0066] In the embodiment of the present invention, the maximum illumination value is 1000 lux, the minimum illumination value is 200 lux, and the illumination value other than the maximum and minimum illumination values is 500 lux. Therefore, the grayscale image of the i-th test steel material under the illumination value of 500 lux is recorded as the target image. According to the change of the grayscale value of the pixel point in the target image with the illumination value, the noise pixel point in the target image is obtained. Specifically:

[0067] Calculating a difference between the maximum illumination value and the minimum illumination value to obtain an illumination value span, calculating a first difference between the illumination value corresponding to the target image and the minimum illumination value, and using a ratio between the first difference and the illumination value span as an illumination value change ratio;

[0068] With the lower left corner of the target image as the origin, construct a two-dimensional rectangular coordinate system corresponding to the target image, obtain the coordinates of any pixel point in the target image according to the two-dimensional rectangular coordinate system, and obtain the pixel point with the same coordinates as the any pixel point in the grayscale image at the maximum illumination value and the minimum illumination value, respectively, to obtain the maximum illumination pixel point and the minimum illumination pixel point of the any pixel point;

[0069] Calculating a grayscale value difference between the maximum illumination pixel and the minimum illumination pixel to obtain a grayscale value span, calculating a second difference between the grayscale value of any pixel and the grayscale value of the minimum illumination pixel, and using a ratio between the second difference and the grayscale value span as a grayscale value change ratio;

[0070] Calculating the absolute value of the difference between the illumination value change ratio and the grayscale value change ratio, taking the reciprocal of the sum of the absolute value of the difference and a preset constant as a first variable, and subtracting the first variable from the constant 1 to obtain the noise probability of any pixel;

[0071] If the noise probability of any pixel is greater than or equal to a preset noise probability threshold, the pixel is marked as a noise pixel.

[0072] In one embodiment, taking the j-th pixel in the target image as an example, the calculation formula for the noise probability of the j-th pixel is:

[0073]

[0074] Among them, α j Indicates the noise probability of the j-th pixel, Z j Indicates the illumination value of the target image, Z min Indicates the minimum illumination value, Z max Indicates the maximum illumination value, X j Indicates the gray value of the j-th pixel, X min Indicates the grayscale value of the pixel with minimum illumination, X max Indicates the grayscale value of the pixel with maximum illumination, c represents a preset constant, and 1 represents a constant.

[0075] It should be noted that That is the ratio of illumination value change, X min Used to represent the gray value of the j-th pixel when the illumination value is the minimum value, X max Used to represent the grayscale value of the j-th pixel when the illumination value is maximum. is the gray value change ratio, The smaller the value of , the more the gray value of the j-th pixel changes with the change of the light source illumination, and the change ratio is similar, and thus αj The smaller it is, the smaller the probability that the j-th pixel is a noise point. Setting c=1 is used to prevent the denominator from being 0. There is no restriction here, and the implementer can set it according to the specific scenario.

[0076] According to experimental statistics, the preset noise probability threshold is set to 0.7. There is no restriction here. The implementer can set it according to the specific scene. If the noise probability α of the jth pixel j If ≥0.7, the jth pixel is marked as a noise pixel. Similarly, the noise probability of each pixel in the target image is obtained, and then all the noise pixels in the target image are obtained.

[0077] In order to reduce the interference of noise pixels in the target image on subsequent calculations, the grayscale value of each noise pixel in the target image is updated according to the grayscale value of the pixels in the neighborhood of each noise pixel in the target image to obtain a new target image. Specifically:

[0078] For any noise pixel, calculate the grayscale value mean of all non-noise pixels in the eight neighborhoods of the noise pixel, update the grayscale value mean to the grayscale value of the noise pixel, and update the grayscale value of each noise pixel in the target image to obtain a new target image.

[0079] (2) The new target image is divided into a preset number of sub-regions, and the local uniformity of each sub-region is obtained according to the distribution of the grayscale values of the pixels in each sub-region.

[0080] In an embodiment of the present invention, the surface roughness of the i-th test steel material is obtained based on the uniformity of the grayscale value distribution of the pixels in the new target image: the more uniform the grayscale value distribution in the grayscale image, the smaller the surface roughness of the i-th test steel material. In order to more accurately analyze the distribution of the grayscale values of the pixels in the new target image, in an embodiment of the present invention, the new target image is divided into a preset number of sub-regions, and the distribution of the grayscale values of the pixels in each sub-region is analyzed separately. The number of sub-regions is related to the size of the new target image. Assuming that the size of the new target image in an embodiment of the present invention is 10×10, the number of sub-regions is set to 4, and the size of each sub-region is 5×5. There is no limitation here, and the implementer can set it according to the specific scenario.

[0081] Furthermore, taking the kth sub-region as an example, the local uniformity of the kth sub-region is obtained according to the distribution of the grayscale values of the pixels in the kth sub-region. Specifically:

[0082] Get the number of all pixels in the kth subregion and the total grayscale value of all pixels in the kth subregion;

[0083] Calculate the product of the grayscale value of any pixel point in the kth sub-region and the number of all pixels, take the absolute value of the difference between the product and the total grayscale value as the grayscale value distribution characteristic value of any pixel point, calculate the average value of the grayscale value distribution characteristic values of all pixels in the kth sub-region, and take the reciprocal of the result of adding the average value and the constant 1 as the local uniformity of the kth sub-region.

[0084] In one embodiment, the calculation formula for the local uniformity of the kth sub-region is:

[0085]

[0086] Among them, β k Indicates the local uniformity of the kth sub-region, m k Represents the number of all pixels in the kth subregion, G p represents the grayscale value of the p-th pixel in the k-th sub-region, represents the total grayscale value of all pixels in the kth sub-region, 1 represents a constant, and || represents the absolute value sign.

[0087] It should be noted that m k ×G p It is used to represent the assumption that the grayscale values of all pixels in the kth subregion are G p When , the total grayscale value of all pixels in the kth sub-region is, The smaller it is, the closer the grayscale values of each pixel in the kth sub-region are, and the more uniform the grayscale value distribution of the pixels in the kth sub-region is, and thus β k The larger , the greater the local uniformity of the kth sub-region.

[0088] Similarly, the local uniformity of each sub-region in the new target image is obtained.

[0089] (3) According to the local uniformity of each sub-region in the new target image, the surface roughness of the i-th test steel material before coating is obtained.

[0090] After obtaining the local uniformity of each subregion in the new target image, the local uniformity of each subregion is linearly normalized to obtain a normalized local uniformity value of each subregion. The linear normalization is a prior art and will not be described in detail here. The average of all normalized local uniformity values is calculated to obtain a local uniformity mean, and then the overall uniformity of the new target image is obtained based on the local uniformity mean. Specifically:

[0091] Obtain the standard deviation of the grayscale values of all pixels in the new target image, calculate the reciprocal of the sum of a constant 1 and the standard deviation, obtain the overall grayscale difference index of the new target image, and perform weighted summation of the local uniformity mean and the overall grayscale difference index to obtain the overall uniformity of the new target image.

[0092] In one embodiment, the calculation formula for the overall uniformity of the new target image is:

[0093]

[0094] Among them, γ represents the overall uniformity of the new target image, n represents the number of sub-regions in the new target image, and β′ k It represents the linear normalized value of the local uniformity of the k-th sub-region in the new target image, s represents the standard deviation of the grayscale values of all pixels in the new target image, w1 represents the first weight, w2 represents the second weight, and 1 represents a constant used to prevent the denominator from being zero.

[0095] It should be noted that, since the local uniformity of each sub-region in the new target image is more important, w1=0.6 and w2=0.4 are set. There is no restriction here, and the implementer can set it according to the specific scenario. is the local uniformity mean, The larger s is, the more uniform the grayscale value distribution of the pixels in each sub-region is, and thus the larger γ is, the more uniform the grayscale value distribution of the pixels in the new target image is; since the standard deviation can be used to measure the degree of discreteness of the data, the smaller s is, the more concentrated the grayscale value distribution of the pixels in the new target image is, that is, the more uniform the grayscale value distribution of the pixels in the new target image is, and thus the larger γ is, the greater the overall uniformity of the new target image is.

[0096] After obtaining the overall uniformity of the new target image, the surface roughness of the i-th test steel material is obtained according to the overall uniformity. Specifically:

[0097] The surface roughness of the i-th test steel material before coating is obtained by subtracting the overall uniformity from the constant 1.

[0098] In one embodiment, the surface roughness of the i-th test steel material is calculated as follows:

[0099] μ=1―γ

[0100] Where μ represents the surface roughness of the i-th test steel material, γ represents the overall uniformity of the new target image, and 1 represents a constant.

[0101] It should be noted that the greater the overall uniformity of the new target image, the smaller the surface roughness of the i-th test steel material.

[0102] At this point, the surface roughness of the i-th test steel material before coating is obtained. Similarly, based on the grayscale images of each coating layer of the i-th test steel material under illumination values of 200 lux, 500 lux, and 1000 lux, the surface roughness of each coating layer of the i-th test steel material after coating is obtained.

[0103] Considering that during the coating process of the i-th test steel material, the coating can fill the minor surface irregularities on the steel material, making the surface smoother, i.e., the surface roughness of the i-th test steel material gradually decreases after coating; however, the coating itself may contain particulate impurities. After the surface roughness of the i-th test steel material decreases to a certain level, a turning point occurs, i.e., its surface roughness increases due to the particulate impurities in the coating. Therefore, as the number of coating layers increases, the surface roughness of the i-th test steel material may show a trend of first decreasing and then increasing. Because the surface roughness of the steel material affects the adhesion of the coating, i.e., the surface roughness of each coating layer of the steel material affects the difference between the actual coating amount and the theoretical coating amount of each coating layer, in this embodiment of the present invention, the actual coating amount of each coating layer of the i-th test steel material after coating is first obtained. Specifically, a certain amount of coating is obtained, a layer of coating is applied to the surface of the i-th test steel material, and the remaining coating amount is obtained. Based on the remaining coating amount, the actual coating amount used for the coating layer is obtained. Then, the absolute value of the difference between the actual coating amount and the theoretical coating amount of each coating layer of the i-th test steel material after coating is calculated to obtain the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer. Based on the coating deviation value of each coating layer, the turning roughness is obtained from the surface roughness of all coating layers of the i-th test steel material, thereby more accurately analyzing the relationship between surface roughness and the number of coating layers. The specific method for obtaining the turning roughness from the surface roughness of all coating layers of the i-th test steel material based on the coating deviation value of each coating layer is:

[0104] Mapping the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of the i-th test steel material into the same scatter plot, wherein the abscissa of the scatter plot represents the number of coating layers of any one of the steel materials to be coated, and the ordinate represents the coating deviation value;

[0105] For any coating deviation value in the scatter plot, if the any coating deviation value is the first coating deviation value in the scatter plot, the slope between the any coating deviation value and its right adjacent coating deviation value is obtained, which is recorded as K1. The inverse of the slope is subtracted from the constant 1 to obtain the turning rate of the any coating deviation value, which is recorded as Right now

[0106] If any of the paint deviation values is the last paint deviation value in the scatter plot, the slope between the paint deviation value and its left adjacent paint deviation value is obtained, which is recorded as K2. The inverse of the slope is subtracted from the constant 1 to obtain the turning rate of the paint deviation value, which is recorded as Right now

[0107] If any of the coating deviation values is not the first coating deviation value and the last coating deviation value in the scatter plot, the left slope between the any coating deviation value and its left adjacent coating deviation value is calculated, recorded as K2, and the right slope between the any coating deviation value and its right adjacent coating deviation value is recorded as K1. The inverse of the absolute value of the difference between the left slope and the right slope is subtracted from the constant 1 to obtain the turning rate of the any coating deviation value, recorded as Right now Among them, || represents the absolute value symbol;

[0108] The inflection rates of all coating deviation values are obtained, the coating deviation value corresponding to the maximum value among all inflection rates is recorded as the inflection deviation value, and the surface roughness of the coating layer corresponding to the inflection deviation value is recorded as the inflection roughness.

[0109] After obtaining the turning roughness of the i-th test steel material during the coating process, a relationship curve between the surface roughness of the i-th test steel material and the number of coating layers is constructed based on the turning roughness, the surface roughness of the i-th test steel material before coating, the surface roughness of each coating layer after coating, and the number of coating layers. This curve is recorded as the first change curve. Specifically:

[0110] The surface roughness of the i-th test steel material before coating and the surface roughness of each coating layer after coating are mapped into the same scatter plot, which is recorded as a first scatter plot. The abscissa of the first scatter plot represents the number of coating layers of the any steel material to be coated, and the ordinate represents the surface roughness. In the first scatter plot, the data between the surface roughness of the any test steel material before coating and the turning roughness are fitted to obtain a first sub-curve, and the data after the turning roughness are fitted to obtain a second sub-curve.

[0111] The first sub-curve and the second sub-curve are connected to obtain a first variation curve between surface roughness and the number of coating layers.

[0112] Since the surface roughness of each coating layer of the steel material will affect the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer, after obtaining the relationship between the surface roughness of the i-th test steel material and the number of coating layers, a relationship curve between the coating deviation value and the surface roughness of the i-th test steel material is constructed based on the surface roughness and coating deviation value of each coating layer. This curve is recorded as the second variation curve. Specifically:

[0113] Constructing a second scatter plot based on the coating deviation value and surface roughness of each coating layer of the i-th test steel material, wherein the abscissa of the second scatter plot represents the surface roughness of each coating layer, and the ordinate represents the coating deviation value of each coating layer;

[0114] In the second scatter plot, the data of any test steel material before the turning roughness are fitted to obtain a third sub-curve, and the data after the turning roughness are fitted to obtain a fourth sub-curve;

[0115] The third sub-curve is connected to the fourth sub-curve to obtain a second variation curve between the coating deviation value and the surface roughness.

[0116] At this point, by conducting a coating test on the i-th test steel material, combined with the characteristics of the coating itself and the influence of the surface roughness of the i-th test steel material on the amount of coating, the first change curve between the surface roughness of the i-th test steel material and the number of coating layers, and the second change curve between the coating deviation value and the surface roughness were obtained. Similarly, the first change curve between the surface roughness and the number of coating layers of each test steel material, and the second change curve between the coating deviation value and the surface roughness were obtained.

[0117] Step S103, fitting the first change curve and the second change curve of all test steel materials respectively, and obtaining the surface roughness fitting function and the coating deviation value fitting function accordingly, and constructing the overall coating deviation value function based on the surface roughness fitting function and the coating deviation value fitting function.

[0118] Through step S102, a first variation curve between the surface roughness and the number of coating layers of each test steel material and a second variation curve between the coating deviation value and the surface roughness are obtained. Then, average curve fitting is performed on the first variation curve and the second variation curve of all test steel materials, and a surface roughness fitting function and a coating deviation fitting function are obtained accordingly. Specifically:

[0119] Performing average curve fitting on the first sub-curves in the first change curves of all the test steel materials to obtain a first sub-segmental function, performing average curve fitting on the second sub-curves in the first change curves of all the test steel materials to obtain a second sub-segmental function, and combining the first sub-segmental function and the second sub-segmental function to form a surface roughness fitting function, wherein the independent variable of the first sub-segmental function is less than or equal to the number of coating layers corresponding to the turning roughness, and the independent variable of the second sub-segmental function is greater than the number of coating layers corresponding to the turning roughness;

[0120] The third sub-curves in the second variation curves for all test steel materials were averaged and fitted to obtain a third sub-segmental function. The fourth sub-curves in the second variation curves for all test steel materials were averaged and fitted to obtain a fourth sub-segmental function. The third and fourth sub-segmental functions were combined to form a coating deviation fitting function. The independent variable of the third sub-segmental function was less than or equal to the turning roughness, and the independent variable of the fourth sub-segmental function was greater than the turning roughness. Average curve fitting is a conventional technique and will not be described in detail here.

[0121] Furthermore, based on the surface roughness fitting function and the coating deviation fitting function, an overall coating deviation function is constructed to predict the coating deviation value of each coating layer according to the number of coating layers of the steel material to be coated in the required scenario. Specifically:

[0122] The output of the surface roughness fitting function serves as the input of the coating deviation value fitting function, and the output of the coating deviation value fitting function serves as the input of the overall coating deviation value function. The output of the overall coating deviation value function is the overall coating deviation value corresponding to the predicted coating of any steel material to be coated.

[0123] Among them, the calculation formula of the overall coating deviation value is:

[0124]

[0125] Among them, σ represents the overall coating deviation value, L represents the number of coating layers corresponding to the turning roughness, and R l represents the coating deviation value of the lth coating layer, H represents the number of remaining coating layers after the coating layer corresponding to the turning roughness, and R h Indicates the paint deviation value of the hth paint layer after the turning roughness.

[0126] At this point, based on the paint's own characteristics and the impact of the surface roughness of the steel material to be coated on the paint dosage, the overall paint offset value of each steel material to be coated is comprehensively predicted to reflect the difference between the actual paint amount and the theoretical paint amount. The paint dosage of each steel material to be coated is predicted based on the overall paint offset value, thereby improving the accuracy and reliability of the paint dosage prediction.

[0127] Step S104, according to the number of coating layers of each steel material to be coated under the target type in the required usage scenario, the theoretical coating amount of each coating layer, the coating loss rate and the overall coating deviation value function, obtain the predicted coating amount of each steel material to be coated under the required usage scenario.

[0128] Through step S103, an overall coating deviation function is obtained, which is used to reflect the difference between the actual coating amount and the theoretical coating amount of each steel material to be coated. Furthermore, considering the influence of the coating application method on coating loss, the coating amount of each steel material to be coated under the target type is predicted to obtain an estimated coating amount for each steel material to be coated. Specifically:

[0129] For any steel material to be coated under the target type, the overall coating deviation value of any steel material to be coated is obtained according to the number of coating layers of any steel material to be coated in the required usage scenario and the overall coating deviation value function, the theoretical material amount of each coating layer is accumulated to obtain the final theoretical material amount of any steel material to be coated, the addition result of the final theoretical material amount and the overall coating deviation value is calculated, and the ratio of the addition result to the coating loss rate is used as the estimated coating amount of any steel material to be coated.

[0130] In one embodiment, taking the uth steel material to be coated as an example, if the coating method is brush coating, the loss rate is generally 5%-15%. If the spraying method is used, the loss rate is about 10%-20% when airless spraying is used, and the loss rate is about 50% when air spraying is used. At the same time, due to the adhesion of the coating, some residue will remain inside the container, and the loss rate is generally 5%. In this embodiment of the present invention, the loss rate of the spraying method and the loss rate of the container residue are added together to obtain the coating loss rate. There is no limitation here, and the implementer can set the type of loss rate according to the specific scenario. The calculation formula of the estimated coating amount of the uth steel material to be coated is:

[0131]

[0132] Among them, F u represents the estimated value of the coating amount of the u-th steel material to be coated, W represents the final theoretical material amount of the u-th steel material to be coated, σ represents the overall coating deviation value of the u-th steel material to be coated, ρ1 represents the coating loss rate of the coating construction method, and ρ2 represents the loss rate of the coating residue in the coating container.

[0133] It should be noted that, in the embodiment of the present invention, ρ1=20% and ρ2=5% are set, which is not limited here and can be set by the implementer according to the specific scenario.

[0134] Similarly, an estimated value of the coating amount for each steel material to be coated under the target category is obtained, and the amount of coating required for the steel material to be coated under the target category is reserved based on the estimated value of the coating amount.

[0135] In summary, the present invention uses a preset number of steel materials to be coated under a target type as test steel materials, obtains the number of coating layers required for any test steel material under a preset scenario, and the theoretical coating amount of each coating layer; obtains the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating to construct a first change curve between surface roughness and the number of coating layers, and constructs a second change curve between coating deviation and surface roughness based on the coating deviation between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer; fits the first change curve and the second change curve of all test steel materials respectively to obtain a surface roughness fitting function and a coating deviation fitting function, and constructs an overall coating deviation function based on the surface roughness fitting function and the coating deviation fitting function; and obtains a predicted coating amount for each steel material to be coated under the required usage scenario based on the number of coating layers, the theoretical coating amount of each coating layer, the coating loss rate, and the overall coating deviation function for each steel material to be coated under the target type under the required usage scenario. Among them, the relationship between the surface roughness of each coating layer of the steel material to be coated and the number of coating layers (surface roughness fitting function), the relationship between the surface roughness of each coating layer and the coating deviation value (coating deviation fitting function), and the coating loss are obtained, so as to combine the influence of the surface roughness of the steel material on the coating amount, the influence of the coating construction method on the coating loss, and the inherent characteristics of the coating to comprehensively predict the coating amount of all steel materials to be coated under the target type, thereby improving the accuracy and reliability of the coating amount prediction.

[0136] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for predicting and optimizing the amount of coating applied to a steel structure surface, characterized in that: The coating amount prediction and optimization method for applying coating on the surface of steel structure includes: Using a preset number of steel materials to be coated under a target type as test steel materials, obtaining the number of coating layers required for any test steel material under a preset scenario, as well as the theoretical coating amount of each coating layer; Obtaining the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating to construct a first variation curve between surface roughness and the number of coating layers; constructing a second variation curve between the coating deviation value and surface roughness based on a coating deviation value between an actual coating amount and a theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer; Fitting the first change curve and the second change curve of all test steel materials respectively to obtain a surface roughness fitting function and a coating deviation value fitting function, and constructing an overall coating deviation value function based on the surface roughness fitting function and the coating deviation value fitting function; According to the number of coating layers of each steel material to be coated under the target type in the required usage scenario, the theoretical coating amount of each coating layer, the coating loss rate and the overall coating deviation value function, the predicted coating amount of each steel material to be coated in the required usage scenario is obtained.

2. A coating amount prediction and optimization method for coating application on steel structure surface according to claim 1, characterized in that: The obtaining of the surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating comprises: Before coating any of the test steel materials, grayscale images of the test steel material under three preset different illumination values are obtained, grayscale images under illumination values other than the maximum illumination value and the minimum illumination value are recorded as target images, and all noise pixels in the target image are obtained according to the grayscale value of each pixel in each of the grayscale images; For any noise pixel, calculate the grayscale value mean of all non-noise pixels in the eight neighborhoods of the noise pixel, update the grayscale value mean to the grayscale value of the noise pixel, and update the grayscale value of each noise pixel in the target image to obtain a new target image. Dividing the new target image into a preset number of sub-regions, and obtaining the local uniformity of each sub-region according to the distribution of the grayscale values of the pixels in each sub-region; Calculating the average of the linearly normalized values of the local uniformity of all sub-regions to obtain a local uniformity mean, obtaining the standard deviation of the grayscale values of all pixels in the new target image, calculating the reciprocal of the sum of a constant 1 and the standard deviation to obtain an overall grayscale difference index of the new target image, and performing a weighted summation of the local uniformity mean and the overall grayscale difference index to obtain an overall uniformity of the new target image; Subtract the overall uniformity from the constant 1 to obtain the surface roughness of any test steel material before coating; According to the grayscale images of each coating layer of any test steel material after coating at the three preset different illumination values, the surface roughness of each coating layer of any test steel material after coating is obtained accordingly.

3. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 2, characterized in that: The step of obtaining all noise pixels in the target image according to the grayscale value of each pixel in each grayscale image comprises: Calculating a difference between the maximum illumination value and the minimum illumination value to obtain an illumination value span, calculating a first difference between the illumination value corresponding to the target image and the minimum illumination value, and using a ratio between the first difference and the illumination value span as an illumination value change ratio; With the lower left corner of the target image as the origin, construct a two-dimensional rectangular coordinate system corresponding to the target image, obtain the coordinates of any pixel point in the target image according to the two-dimensional rectangular coordinate system, and obtain the pixel point with the same coordinates as the any pixel point in the grayscale image at the maximum illumination value and the minimum illumination value, respectively, to obtain the maximum illumination pixel point and the minimum illumination pixel point of the any pixel point; Calculating a grayscale value difference between the maximum illumination pixel and the minimum illumination pixel to obtain a grayscale value span, calculating a second difference between the grayscale value of any pixel and the grayscale value of the minimum illumination pixel, and using a ratio between the second difference and the grayscale value span as a grayscale value change ratio; Calculating the absolute value of the difference between the illumination value change ratio and the grayscale value change ratio, taking the reciprocal of the sum of the absolute value of the difference and a preset constant as a first variable, and subtracting the first variable from the constant 1 to obtain the noise probability of any pixel; If the noise probability of any pixel is greater than or equal to a preset noise probability threshold, the pixel is marked as a noise pixel.

4. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 2, characterized in that: The obtaining of the local uniformity of each sub-region according to the distribution of the grayscale values of the pixels in each sub-region includes: For any sub-region, obtain the number of all pixels in the sub-region and the total grayscale value of all pixels in the sub-region; Calculate the product of the grayscale value of any pixel point in any sub-region and the number of all pixel points, take the absolute value of the difference between the product and the total grayscale value as the grayscale value distribution characteristic value of any pixel point, calculate the average value of the grayscale value distribution characteristic values of all pixel points in any sub-region, and take the reciprocal of the result of adding the average value and the constant 1 as the local uniformity of any sub-region.

5. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 1, characterized in that: The step of obtaining a first variation curve between surface roughness and the number of coating layers comprises: Obtaining an actual coating amount of each coating layer of any test steel material after coating, and obtaining a turning roughness from the surface roughness of all coating layers according to a coating deviation value between the actual coating amount and a theoretical coating amount of each coating layer of any test steel material; The surface roughness of any test steel material before coating and the surface roughness of each coating layer after coating are mapped into the same scatter plot, recorded as a first scatter plot, wherein the abscissa of the first scatter plot represents the number of coating layers of the any test steel material to be coated, and the ordinate represents the surface roughness. In the first scatter plot, the data between the surface roughness of the any test steel material before coating and the turning roughness are fitted to obtain a first sub-curve, and the data after the turning roughness are fitted to obtain a second sub-curve; The first sub-curve and the second sub-curve are connected to obtain a first variation curve between surface roughness and the number of coating layers.

6. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 5, characterized in that: The step of obtaining the turning roughness from the surface roughness of all coating layers according to the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material comprises: Mapping the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material into the same scatter plot, wherein the abscissa of the scatter plot represents the number of coating layers of any to-be-coated steel material, and the ordinate represents the coating deviation value; For any coating deviation value in the scatter plot, if the any coating deviation value is the first coating deviation value in the scatter plot, obtaining a slope between the any coating deviation value and its right adjacent coating deviation value, and subtracting the reciprocal of the slope from a constant 1 to obtain a turning rate of the any coating deviation value; If any one of the coating deviation values is the last coating deviation value in the scatter plot, obtaining a slope between the any one of the coating deviation values and its left adjacent coating deviation value, and subtracting the reciprocal of the slope from a constant 1 to obtain a turning rate of the any one of the coating deviation values; If any of the coating deviation values is not a value other than the first coating deviation value and the last coating deviation value in the scatter plot, then calculating a left slope between the any coating deviation value and its left adjacent coating deviation value, and a right slope between the any coating deviation value and its right adjacent coating deviation value, and subtracting the reciprocal of the absolute value of the difference between the left slope and the right slope from a constant 1 to obtain a turning rate of the any coating deviation value; The inflection rates of all coating deviation values are obtained, the coating deviation value corresponding to the maximum value among all inflection rates is recorded as the inflection deviation value, and the surface roughness of the coating layer corresponding to the inflection deviation value is recorded as the inflection roughness.

7. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 5, characterized in that: The method of constructing a second variation curve between the coating deviation value and the surface roughness based on the coating deviation value between the actual coating amount and the theoretical coating amount of each coating layer of any test steel material and the surface roughness of each coating layer comprises: constructing a second scatter plot based on the coating deviation value and surface roughness of each coating layer of any one of the test steel materials, wherein the abscissa of the second scatter plot represents the surface roughness of each coating layer, and the ordinate represents the coating deviation value of each coating layer; In the second scatter plot, the data of any test steel material before the turning roughness are fitted to obtain a third sub-curve, and the data after the turning roughness are fitted to obtain a fourth sub-curve; The third sub-curve is connected to the fourth sub-curve to obtain a second variation curve between the coating deviation value and the surface roughness.

8. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 7, characterized in that: The first variation curve and the second variation curve of all the test steel materials are fitted respectively to obtain the surface roughness fitting function and the coating deviation value fitting function, including: Fitting the first sub-curve in the first variation curve of all test steel materials to obtain a first sub-segmental function, fitting the second sub-curve in the first variation curve of all test steel materials to obtain a second sub-segmental function, combining the first sub-segmental function and the second sub-segmental function to form a surface roughness fitting function, wherein the independent variable of the first sub-segmental function is less than or equal to the number of coating layers corresponding to the turning roughness, and the independent variable of the second sub-segmental function is greater than the number of coating layers corresponding to the turning roughness; The third sub-curve in the second change curve of all test steel materials is fitted to obtain a third sub-segmental function, and the fourth sub-curve in the second change curve of all test steel materials is fitted to obtain a fourth sub-segmental function. The third sub-segmental function and the fourth sub-segmental function are combined to form a coating deviation value fitting function, the independent variable of the third sub-segmental function is less than or equal to the turning roughness, and the independent variable of the fourth sub-segmental function is greater than the turning roughness.

9. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 1, characterized in that: The step of constructing an overall coating deviation value function based on the surface roughness fitting function and the coating deviation value fitting function comprises: The output of the surface roughness fitting function serves as the input of the coating deviation value fitting function, and the output of the coating deviation value fitting function serves as the input of the overall coating deviation value function. The output of the overall coating deviation value function is the overall coating deviation value corresponding to the predicted coating of any steel material to be coated.

10. The method for predicting and optimizing the amount of coating applied to a steel structure surface according to claim 9, characterized in that: Obtaining the predicted amount of coating for each steel material to be coated in the desired usage scenario based on the number of coating layers for each steel material to be coated under the target type in the desired usage scenario, the theoretical coating amount of each coating layer, the coating loss rate, and the overall coating deviation value function, includes: For any steel material to be coated under the target type, the overall coating deviation value of any steel material to be coated is obtained according to the number of coating layers of any steel material to be coated in the required usage scenario and the overall coating deviation value function, the theoretical material amount of each coating layer is accumulated to obtain the final theoretical material amount of any steel material to be coated, the addition result of the final theoretical material amount and the overall coating deviation value is calculated, and the ratio of the addition result to the coating loss rate is used as the estimated coating amount of any steel material to be coated.

Images