A method and system for early warning of loss of corrosion-resistant composite coating on steel surface

By setting sampling points on the steel surface to obtain multiple data and perform linear and same-direction combination, a loss array is generated, which solves the problem of insufficient accuracy of the early warning of the coating of steel structures in marine environments, and realizes the corrosion-resistant composite coating on the steel surface Effective early warning extends the service life of steel structures and reduces resource waste.

CN118759030BActive Publication Date: 2024-11-12ANJI (JIANGSU) TECHNOLOGY MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411214705.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-12
Estimated Expiration
2044-09-02

AI Technical Summary

Technical Problem

The existing technology has insufficient accuracy in the early warning of corrosion-resistant coatings of steel structures in marine environments, and it is impossible to effectively capture the complex impact of multiple factors such as chloride ion content and porosity on coating losses, resulting in waste of resources and unsatisfactory protection effects. .

Method used

By setting multiple sampling points on the steel surface, the surface charge density, chloride ion content and porosity data are obtained, and the surface charge density probability amplitude time series, the surface charge density probability ratio time series, the chloride ion content probability ratio time series, and the porosity probability ratio time series are calculated, and linearly Combining in the same direction, a corrosion particle loss array is generated, the characteristic vectors and correlations of the corrosion particle loss matrix are calculated, and loss warning is performed.

Benefits of technology

It has achieved timely loss warning for corrosion-resistant composite coatings on the steel surface, extend the service life of the steel structure, reduce resource waste, and improve overall safety and economic benefits.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118759030B_ABST
    Figure CN118759030B_ABST
Patent Text Reader

Abstract

The present invention provides a method and system for early warning of the loss of corrosion-resistant composite coatings on steel surfaces, which belongs to the field of calculation and measurement of physical data. By setting multiple sampling points on the coating surface, surface charge density, chloride ion content and porosity data are obtained at different sampling times. Based on these data, the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series of each sampling point are calculated, and they are linearly combined in the same direction to generate a corrosion electron loss array. By calculating the eigenvector of the corrosion electron loss matrix and the correlation of the corrosion electron loss of each sampling point, early warning of the loss of the coating is given. Potential problems can be discovered and dealt with in a timely manner in the early stage of corrosion of the composite coating, thereby extending the service life of the steel structure protected by the coating and reducing the waste of resources of seawater steel structures.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of physical data calculation and measurement, and specifically relates to a loss warning method and system for a corrosion-resistant composite coating on a steel surface. Background Art

[0002] A large number of oil platforms, cross-sea projects, ocean ports, artificial islands and reefs are being built in major sea areas. However, due to the strong corrosive effect of a large amount of chloride ions in the marine environment, the steel structures in marine projects suffer from serious corrosion damage, which limits the safety, reliability and durability of the entire marine project. As a simple corrosion protection technology, the demand for corrosion-resistant coatings has increased significantly. Chloride ions are the main controlling factor causing steel corrosion in the marine environment. Chlorine-induced corrosion has local inhomogeneity and random characteristics, which causes traditional corrosion-resistant coating technology to have to be used excessively during application, resulting in a large amount of resource waste. Therefore, layered double hydroxides (LDH) can be applied to the coating, so that LDH can be used as a trap to capture environmental chloride ions to reduce the corrosion rate of steel. The corrosion-resistant composite coating used in marine engineering is a four-layer composite film layer grown on the surface of steel, which may include a Mg-Al-LDH film layer attached to the surface of steel, a protein film layer adsorbed on the LDH film by hydrogen bonding, and a tungsten trioxide layer adsorbed on the protein film by hydrogen bonding. The chemical action of the LDH film layer to absorb corrosive chloride ions, the physical barrier effect of the protein sealing the porous structure of LDH, and the electrochemical action of tungsten trioxide to hinder electron transmission. However, there are some rules between the loss of the corrosion-resistant composite coating on the surface of steel and the physical properties such as surface charge density, chloride ion content, and porosity. The rules of these data reflect the electrochemical and physical properties of the coating and will directly affect its corrosion resistance. Among them, the surface charge density reflects the distribution of charges on the surface of the coating and directly affects the progress of the electrochemical reaction. High surface charge density usually means more active sites, which may accelerate the corrosion process. For certain specific types of coatings, appropriate surface charge density can form an effective protective layer and reduce the corrosion rate. Chloride ions are one of the main factors leading to steel corrosion. The higher the chloride ion content, the faster the corrosion rate. This is because chloride ions can destroy the passive film on the surface of steel, forming corrosion pits, leading to local corrosion. Local high concentrations of chloride ions can cause local corrosion in the form of stress corrosion cracking (SCC) and pitting, accelerating coating loss. The porosity of the coating reflects the density of the coating. Low porosity means that the coating is denser and can more effectively block the penetration of corrosive media (such as water, oxygen and chloride ions), thereby improving corrosion resistance. Coatings with high porosity are prone to form penetration pathways, making it easier for corrosive media to reach the surface of steel, leading to accelerated corrosion and coating loss. There is a synergistic effect between factors such as surface charge density, chloride ion content and porosity, which jointly affect the corrosion resistance of the coating. For example, although high surface charge density may lead to accelerated corrosion, if the coating has low porosity and can effectively capture chloride ions (such as LDH coating), it may still improve the corrosion resistance of the coating as a whole. The impact of these physical properties on coating performance is dynamic and changes over time. Time series analysis can help reveal the laws of how these variables change over time, thereby predicting the loss of the coating.In its actual application, the specific relationship of these data may vary depending on the type of coating, environmental conditions and usage scenarios. In some cases, coating loss may also show a linear relationship with one or more variables. For example, the coating loss rate may be positively correlated with the chloride content and negatively correlated with the porosity. In more cases, these relationships are nonlinear and may require the use of complex mathematical models or machine learning algorithms to capture. Coating loss is usually the result of multiple factors. It is necessary to comprehensively consider data such as surface charge density, chloride content, porosity, etc. in order to accurately predict the corrosion resistance of the coating. By analyzing data such as surface charge density, chloride content, porosity, etc., the law of loss of corrosion-resistant composite coatings on steel surfaces can be revealed. There is a complex relationship between these physical property data, which can be further quantified and modeled through experimental data and statistical analysis methods. Using time series analysis and machine learning algorithms, the loss of coatings can be more accurately predicted, thereby achieving efficient protection of steel structures.

[0003] The strong corrosive effect of a large amount of chloride ions in the marine environment has caused serious corrosion damage to the steel structures in marine engineering, limiting the safety, reliability and durability of the overall project. Traditional corrosion-resistant coating technology mainly relies on excessive use of anti-corrosion coatings to ensure the life of steel structures in harsh environments. However, this method has brought about a lot of resource waste, and it is impossible to accurately predict and prevent the loss of coatings, resulting in unsatisfactory protection effects. Among the existing related technologies, the layered double hydroxide (LDH) corrosion-resistant layer has been proven to significantly improve the corrosion resistance of steel, but its protection efficiency still needs to be optimized. In addition, the existing technology has great limitations in dealing with chloride ion corrosion, and cannot fully capture the complex effects of multiple factors such as chloride ion content and porosity on coating loss, resulting in insufficient accuracy in coating loss prediction and early warning. For example, the coating thickness gauge and coating thickness detection method described in the patent document with publication number CN111649662B only focus on the thickness, which indirectly promotes excessive use of coatings and leads to waste of resources; and the method for evaluating the corrosion resistance of coated steel plates described in the patent document with publication number CN112268850A cannot monitor and predict the loss of coatings in real time and accurately. Furthermore, a steel microstructure recognition algorithm based on a deep learning network described in the patent document with publication number CN116311237A fails to effectively integrate multiple anti-corrosion mechanisms, and the protection efficiency is not high. The above shortcomings limit the application scope and effect of existing anti-corrosion technologies, and there is an urgent need for a more efficient and accurate monitoring and early warning method to improve the corrosion resistance identification and early warning of steel materials in marine environments. Summary of the invention

[0004] The purpose of the present invention is to propose a loss warning method and system for corrosion-resistant composite coatings on steel surfaces to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0005] The present invention provides a method for early warning of the loss of corrosion-resistant composite coatings on steel surfaces, aiming to optimize the use efficiency of the coatings to improve the corrosion resistance of steel structures in marine environments. The method can be applied to the anti-corrosion coatings on the surfaces of various metal structures immersed in seawater. By setting multiple sampling points on the coating surface, surface charge density, chloride ion content and porosity data are obtained at different sampling times. Based on these data, the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series of each sampling point are calculated, and they are linearly combined in the same direction to generate a corrosion electron loss array. By calculating the eigenvector of the corrosion electron loss matrix and the correlation of the corrosion electron loss of each sampling point, early warning of the loss of the coating is given. By quantitatively analyzing and early warning the correlation of the corrosion electron loss of each sampling point of the corrosion-resistant composite coating on the steel surface, potential problems can be discovered and dealt with in time at the early stage of corrosion of the composite coating, thereby extending the service life of the steel structure protected by the coating, reducing resource waste, and improving overall safety and economic benefits.

[0006] In order to achieve the above object, according to one aspect of the present disclosure, a method for early warning of loss of a corrosion-resistant composite coating on a steel surface is provided, the method comprising the following steps:

[0007] A plurality of different sampling points are set on the surface of the coating, and at a plurality of different sampling moments, the surface charge density, chloride ion content and porosity data at each sampling point are sampled and obtained respectively; wherein the number of the sampling points is consistent with the number of the sampling moments;

[0008] According to the data of surface charge density, chloride ion content and porosity corresponding to each sampling point on the surface of the coating at each sampling time, the probability amplitude time series of the surface charge density, the probability ratio time series of the chloride ion content and the probability ratio time series of the porosity corresponding to each sampling point are calculated;

[0009] The surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point are linearly combined in the same direction to obtain the corrosion electron particle loss array of the sampling point;

[0010] The corrosion electron loss array corresponding to each sampling point is formed into a corrosion electron loss matrix, a corrosion electron loss characteristic vector is calculated, and the corrosion electron loss equivalence between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector is calculated;

[0011] Loss warning is carried out according to the corrosion particle loss phase corresponding to each sampling point.

[0012] Furthermore, the method of setting a plurality of different sampling points on the surface of the coating is as follows: homogeneous sampling is performed on the surface of the coating to obtain a plurality of non-overlapping sampling points, and the sequence number of each sampling point remains consistent at each sampling moment.

[0013] Furthermore, in the step of linearly combining the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point in the same direction to obtain the corrosion electron particle loss array of the sampling point:

[0014] The linear and unidirectional time series combination means combining the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point in the same direction of the time series in a manner that maintains a consistent linear relationship and outputs them into an array of constant dimensional size. The corrosion particle loss array is an array that can be used to represent the linear relationship between the degree of change of the surface charge density probability amplitude, chloride ion content probability ratio and porosity probability ratio of each sampling point in the time direction at the sampling moment.

[0015] Furthermore, according to the data of the surface charge density, chloride ion content and porosity corresponding to each sampling point on the surface of the coating at each sampling time, the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point are calculated, specifically:

[0016] At the same sampling time, the values ​​of the surface charge density, the chloride ion content and the porosity corresponding to each sampling point are obtained;

[0017] Extracting the probability distribution of the value of the surface charge density of each sampling point at each sampling moment relative to the value of the surface charge density of the sampling point at all sampling moments, thereby obtaining the probability density of the surface charge density of each sampling point at each sampling moment;

[0018] Extract the probability distribution of the porosity value of each sampling point at each sampling time relative to the porosity values ​​of the sampling point at all sampling times, so as to obtain the porosity probability density of each sampling point at each sampling time;

[0019] Calculate the fluctuation amplitude of the surface charge density probability density of each sampling point at each sampling moment relative to the surface charge density probability density at the preceding and following adjacent sampling moments as the surface charge density probability amplitude of each sampling point at each sampling moment, and use the array composed of the numerical values ​​of the surface charge density probability amplitude of each sampling point at each sampling moment as the surface charge density probability amplitude time series of the sampling point;

[0020] Calculate the ratio of the probability density of chloride ion content at each sampling point at each sampling time to the probability density of chloride ion content at the first sampling time as the probability ratio of chloride ion content at each sampling point at each sampling time, and use the array composed of the values ​​of the probability ratio of chloride ion content at each sampling point at each sampling time as the time series of the probability ratio of chloride ion content at the sampling point;

[0021] Calculate the ratio of the porosity probability density of each sampling point at each sampling moment to the porosity probability density of the sampling point at the first sampling moment as the porosity probability ratio of each sampling point at each sampling moment, and use the array consisting of the values ​​of the porosity probability ratio of each sampling point at each sampling moment as the porosity probability ratio time series of the sampling point.

[0022] Furthermore, the corrosion electron loss array corresponding to each sampling point is formed into a corrosion electron loss matrix, the corrosion electron loss characteristic vector is calculated, and the corrosion electron loss equivalence between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector is calculated, specifically:

[0023] The corrosion electron loss arrays corresponding to each sampling point are used as the columns of the matrix to form a matrix, and the formed matrix is ​​used as the corrosion electron loss matrix;

[0024] Calculate the eigenvector of the corrosion electron loss matrix as the corrosion electron loss eigenvector corresponding to the sampling point on the surface of the coating;

[0025] The cosine similarity between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss feature vector is calculated as the corrosion electron loss similitude corresponding to each sampling point.

[0026] Furthermore, the method for early warning of loss according to the corrosion particle loss phase corresponding to each sampling point is:

[0027] The sampling point with a larger corrosion particle loss phase value indicates a greater probability of coating loss. The sampling point with a corrosion particle loss phase value exceeding the threshold is marked as a loss warning point, and the loss warning point is output as a signal.

[0028] Furthermore, the method for early warning of loss according to the corrosion electron loss phase corresponding to each sampling point also includes:

[0029] According to the values ​​of the corrosion electron loss phase corresponding to each sampling point, the surface of the coating is divided into sub-areas, and the sub-areas where the cumulative values ​​of the corrosion electron loss phase corresponding to the sampling points exceed the threshold are marked as requiring coating reinforcement.

[0030] The present disclosure also provides a loss warning system for a corrosion-resistant composite coating on a steel surface, the loss warning system for a corrosion-resistant composite coating on a steel surface comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor implementing the steps in the loss warning method for a corrosion-resistant composite coating on a steel surface when executing the computer program, the loss warning system for a corrosion-resistant composite coating on a steel surface can be run in computing devices such as desktop computers, laptop computers, mobile phones, PDAs, and cloud data centers, and the executable system may include, but is not limited to, a processor, a memory, and a server cluster, and the processor executing the computer program runs in the following system units:

[0031] A data sampling unit is used to set a plurality of different sampling points on the surface of the coating, and to sample and obtain data of surface charge density, chloride ion content and porosity at each sampling point at a plurality of different sampling moments, wherein the number of sampling points is consistent with the number of sampling moments;

[0032] A probability calculation unit is used to calculate the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point on the surface of the coating at each sampling time according to the surface charge density, chloride ion content and porosity data corresponding to each sampling point on the surface of the coating at each sampling time;

[0033] A time series combination unit is used to linearly and synchronously combine the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point to obtain the corrosion electron particle loss array of the sampling point;

[0034] A loss calculation unit, used to calculate a corrosion electron loss characteristic vector by forming a corrosion electron loss array corresponding to each sampling point into a corrosion electron loss matrix, and calculate the corrosion electron loss equivalence between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector;

[0035] The loss warning unit is used to provide loss warning according to the corrosion particle loss phase corresponding to each sampling point.

[0036] The beneficial effects of the present disclosure are as follows: the present invention provides a method for early warning of the loss of corrosion-resistant composite coatings on steel surfaces, which aims to optimize the use efficiency of the coating to improve the corrosion resistance of steel structures in marine environments. The method sets multiple sampling points on the coating surface to obtain surface charge density, chloride ion content and porosity data at different sampling times. Based on these data, the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series of each sampling point are calculated, and they are linearly combined in the same direction to generate a corrosion electron loss array. By calculating the eigenvector of the corrosion electron loss matrix and the correlation of the corrosion electron loss of each sampling point, an early warning is issued for the loss of the coating. By quantitatively analyzing and early warning the correlation of the corrosion electron loss of each sampling point of the corrosion-resistant composite coating on the steel surface, potential problems can be discovered and dealt with in a timely manner in the early stages of corrosion of the composite coating, thereby extending the service life of the steel structure protected by the coating, reducing resource waste, and improving overall safety and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and other features of the present disclosure will become more apparent by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present disclosure represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0038] Figure 1 Shown is a flow chart of a method for early warning of loss of corrosion-resistant composite coating on steel surface;

[0039] Figure 2 Shown is a system structure diagram of a loss warning system for corrosion-resistant composite coatings on steel surfaces. DETAILED DESCRIPTION

[0040] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present disclosure, so as to fully understand the purpose, scheme and effect of the present disclosure. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0041] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0042] like Figure 1 The flowchart of the method for early warning of the loss of the corrosion-resistant composite coating on the surface of steel according to the present invention is shown below. Figure 1 A method and system for early warning of loss of corrosion-resistant composite coating on a steel surface according to an embodiment of the present invention will be described.

[0043] The present disclosure provides a method for early warning of the loss of a corrosion-resistant composite coating on a steel surface, the method specifically comprising the following steps:

[0044] A plurality of different sampling points are set on the surface of the coating, and at a plurality of different sampling moments, the surface charge density, chloride ion content and porosity data at each sampling point are sampled and obtained respectively; wherein the number of the sampling points is consistent with the number of the sampling moments;

[0045] According to the data of surface charge density, chloride ion content and porosity corresponding to each sampling point on the surface of the coating at each sampling time, the probability amplitude time series of the surface charge density, the probability ratio time series of the chloride ion content and the probability ratio time series of the porosity corresponding to each sampling point are calculated;

[0046] The surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point are linearly combined in the same direction to obtain the corrosion electron particle loss array of the sampling point;

[0047] The corrosion electron loss array corresponding to each sampling point is formed into a corrosion electron loss matrix, a corrosion electron loss characteristic vector is calculated, and the corrosion electron loss equivalence between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector is calculated;

[0048] Loss warning is carried out according to the corrosion particle loss phase corresponding to each sampling point.

[0049] In some embodiments, in order to obtain data such as surface charge density, chloride ion content, porosity, etc. on the coating, multiple sampling can be performed in a seawater environment. For the surface charge density, electrochemical impedance spectroscopy (EIS) can be used. Using electrochemical impedance spectroscopy, the charge distribution and electrochemical properties of the coating surface can be measured. This method is suitable for non-destructive testing in a seawater environment.

[0050] Scanning electrochemical microscopy (SECM) can also be used to measure the local electrochemical reaction rate through high-resolution scanning probe technology, thereby inferring the surface charge density. Among them, an electrochemical workstation equipped with EIS and SECM functions can perform high-precision measurements. Portable EIS instruments are suitable for field use and are convenient for multiple sampling in marine environments.

[0051] For chloride ion content, an ion selective electrode (ISE) can be used. Through the chloride ion selective electrode, the chloride ion concentration in the coating can be directly measured. This method is simple, fast and suitable for on-site measurement. X-ray fluorescence spectroscopy (XRF) can also be used. Using XRF, the coating can be non-destructively analyzed to measure the chloride ion content. Among them, portable ion selective electrodes are suitable for on-site measurement, which is convenient and fast. The portable X-ray fluorescence spectrometer is suitable for on-site use and can provide fast and accurate measurement results. When a metal structure with a corrosion-resistant coating is immersed in a seawater environment, the chloride ion content is sampled at each sampling point on the coating surface of the metal structure.

[0052] For porosity data acquisition, a scanning electron microscope (SEM) can be used. Through high-resolution imaging, the microstructure of the coating can be directly observed and the porosity can be measured. The gas adsorption method (BET method) can use gas adsorption technology to measure the specific surface area and pore structure of the coating, and then calculate the porosity. Among them, the portable scanning electron microscope (Portable SEM) is suitable for on-site measurement and can provide high-resolution images. The portable gas adsorption instrument is also suitable for on-site use and can provide rapid porosity measurement.

[0053] In actual experiments, the corrosion-resistant composite coating can be selected from four composite film layers grown on the surface of steel: including a Mg-Al-LDH film layer attached to the surface of steel, a protein film layer adsorbed on the LDH film by hydrogen bonding, and a tungsten trioxide layer adsorbed on the protein film by hydrogen bonding. The present invention integrates a variety of traditional corrosion protection technologies, using the chemical action of the LDH film layer to absorb corrosive chloride ions, the physical barrier effect of the protein sealing the porous structure of LDH, and the electrochemical action of tungsten trioxide to hinder electron transmission. In the specific experimental steps, for the preparation of the coating surface, 60# to 2000# sandpaper can be used to polish the steel surface in sequence until the scratches on the steel surface are uniform and consistent to obtain a polished steel surface; phosphoric acid pickling is used to etch the steel surface for 2 minutes, and then rinsed with tap water for 30 seconds to obtain pickled steel; a hydrothermal reaction solution is prepared, including 0.05 M Mg(NO3)2, 0.025 M Al(NO3)3, and 0.375 The acid-washed steel and the hydrothermal reaction solution are placed in a high-pressure reactor, kept at a constant temperature of 120°C for 12 hours, and rinsed with tap water and blown dry after the reaction to obtain steel with a Mg-Al-LDH film layer attached to the surface; a protein and WO3 mixed solution is prepared, first 30 mL of protein is added to 20 mL of deionized water, vigorously stirred evenly, then 5 g of tungsten trioxide is added, and vigorously stirred for 1.5 hours until the solution is uniform; the steel with the Mg-Al-LDH film layer is placed in the protein and tungsten trioxide mixed solution, placed in a sealed beaker, and magnetically stirred at a constant temperature of 50°C for 2 hours to obtain steel with a Mg-Al-LDH film, a protein film, and a tungsten trioxide layer attached to the surface.

[0054] In its implementation steps, the coating samples soaked in seawater need to collect data at different time points to ensure that the samples are in the same environmental conditions before and after sampling to avoid human interference.

[0055] Regular sampling is required. According to actual needs, a reasonable sampling time interval should be set to ensure the continuity and integrity of the data. Instruments should be calibrated regularly to ensure the accuracy and reliability of the measured data. The consistency of the sampling environment should be controlled as much as possible to reduce the impact of external factors on the measurement results. Then a systematic data storage and management system should be established for subsequent data analysis and model training.

[0056] Furthermore, the method of setting a plurality of different sampling points on the surface of the coating is as follows: homogeneous sampling is performed on the surface of the coating to obtain a plurality of non-overlapping sampling points, and the sequence number of each sampling point remains consistent at each sampling moment.

[0057] Furthermore, according to the data of the surface charge density, chloride ion content and porosity corresponding to each sampling point on the surface of the coating at each sampling time, the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point are calculated, specifically:

[0058] At the same sampling time, the values ​​of the surface charge density, the chloride ion content and the porosity corresponding to each sampling point are obtained.

[0059] The value of the chloride ion content should be understood as a value representing the concentration or density of the chloride ion content per unit area calculated based on the corresponding surface charge density, or a value representing the concentration or density of the chloride ion content per unit volume calculated based on the corresponding porosity.

[0060] In some embodiments, the total number of sampling moments is recorded as m, wherein the arrangement sequence of each sampling moment is represented by t in chronological order, and the value range of t is 1 to m;

[0061] The total number of sampling points on the surface of the coating is n, where each sampling point can be represented by i in order according to the distance from the geometric center point of the coating surface, where the sampling point with sequence number i is recorded as point(i);

[0062] For the data of sampling point point(i) at sampling time t, point(i) has the corresponding surface charge density value eden(i,t), chloride ion content value lv(i,t) and porosity value pole(i,t). For example, when the sampling point with sequence i equals 2 is point(2), for the data at sampling time t equals 3, point(2) has the corresponding surface charge density value eden(2,3), chloride ion content value lv(2,3) and porosity value pole(2,3);

[0063] Extracting the probability distribution of the value of the surface charge density of each sampling point at each sampling moment relative to the value of the surface charge density of the sampling point at all sampling moments, thereby obtaining the probability density of the surface charge density of each sampling point at each sampling moment;

[0064] In some embodiments, for a sampling point point(i) with a serial number of i, firstly, the values ​​of the surface charge density of point(i) at all sampling moments from t being in the range of 1 to m are obtained for probability statistics, and the values ​​of the surface charge density at each sampling moment from t being in the range of 1 to m are divided into multiple numerical intervals of probability distribution, and the probability density of the values ​​of the surface charge density at each sampling moment in each numerical interval that fall within the interval is counted as the probability density of the numerical interval, and then the probability density of the numerical interval in which the values ​​of the surface charge density at each sampling moment of point(i) are located is obtained as the probability density of the surface charge density of point(i) at the sampling moment.

[0065] Extract the probability distribution of the value of the chloride ion content of each sampling point at each sampling time relative to the value of the chloride ion content of the sampling point at all sampling times, so as to obtain the probability density of the chloride ion content of each sampling point at each sampling time;

[0066] In some embodiments, for a sampling point point(i) with a serial number of i, firstly, the values ​​of the chloride ion content of point(i) at all sampling moments from t being in the range of 1 to m are obtained for probability statistics, and the values ​​of the chloride ion content at each sampling moment from t being in the range of 1 to m are divided into a plurality of numerical intervals of probability distribution, and the probability density of the values ​​of the chloride ion content at each sampling moment in each numerical interval falling within the interval is counted as the probability density of the numerical interval, and then the probability density of the numerical interval in which the values ​​of the chloride ion content at each sampling moment of point(i) are located is obtained as the probability density of the chloride ion content of point(i) at the sampling moment.

[0067] Extract the probability distribution of the porosity value of each sampling point at each sampling time relative to the porosity values ​​of the sampling point at all sampling times, so as to obtain the porosity probability density of each sampling point at each sampling time;

[0068] In some embodiments, for a sampling point point(i) with a serial number of i, firstly, the porosity values ​​of point(i) at all sampling moments from t being in the range of 1 to m are obtained for probability statistics, and the porosity values ​​at each sampling moment from t being in the range of 1 to m are divided into multiple numerical intervals of probability distribution, and the probability density of the porosity values ​​at each sampling moment in each numerical interval that fall within the interval is counted as the probability density of the numerical interval, and then the probability density of the numerical interval in which the porosity values ​​of point(i) at each sampling moment are located is obtained as the porosity probability density of point(i) at the sampling moment.

[0069] Calculate the fluctuation amplitude of the surface charge density probability density of each sampling point at each sampling moment relative to the surface charge density probability density at the previous and next adjacent sampling moments as the surface charge density probability amplitude of each sampling point at each sampling moment, and use the array composed of the numerical values ​​of the surface charge density probability amplitude of each sampling point at each sampling moment as the surface charge density probability amplitude time series of the sampling point.

[0070] In some embodiments, the surface charge density probability density of the sampling point point(i) at each sampling moment is obtained by traversal. For the traversal of the sampling moment number t: when the sequence number t is in the range of 2 to m minus 1, each sampling moment has a previous sampling moment and a next sampling moment, that is, the adjacent sampling moments before and after it. The ratio of the charge density probability density Deden(i, t) of point(i) at the sampling moment number t to the charge density probability density Deden(i, t-1) corresponding to the previous sampling moment t-1 is calculated. The ratio of the charge density probability density Deden(i, t) of point(i) at the sampling moment number t to the charge density probability density Deden(i, t+1) corresponding to the next sampling moment t+1 is calculated. The arithmetic average of the ratio of Deden(i, t) to Deden(i, t-1) and the ratio of Deden(i, t) to Deden(i, t+1) is used as the surface charge density probability amplitude Weden(i, t) of the sampling point point(i) at the sampling moment t.

[0071] When the sequence number t is 1, the charge density probability density of the sampling point point(i) at the first sampling time is Deden(i,1), and the charge density probability density Deden(i,2) corresponding to the sampling time next to the first sampling time, that is, the sampling time with sequence number 2, is obtained. The ratio of Deden(i,1) to Deden(i,2) is calculated as the surface charge density probability amplitude Weden(i,1) of the sampling point point(i) at the first sampling time.

[0072] When the sequence number t is m, the charge density probability density Deden(i,m) of the sampling point point(i) at the last sampling moment is obtained, and the charge density probability density Deden(i,m-1) corresponding to the sampling moment before the last sampling moment, that is, the sampling moment with sequence number m-1, is obtained, and the ratio of Deden(i,m) to Deden(i,m-1) is calculated as the surface charge density probability amplitude Weden(i,m) of the sampling point point(i) at the last sampling moment.

[0073] Among them, when calculating the ratio between the surface charge density probability density of two adjacent sampling moments, the surface charge density probability density of the two adjacent sampling moments can be first processed with an activation function, which is conducive to making the surface charge density probability amplitude of each sampling point at each sampling moment more prominent. The fluctuation amplitude of the surface charge density probability density of the sampling point at the current moment relative to the adjacent sampling moments before and after. Because the surface charge density cannot be used to observe the degree of loss of the composite coating on the steel surface by simply judging its amount, but the surface charge density is more likely to fluctuate in value during the progression of coating corrosion, so it is necessary to compare the surface charge density probability amplitudes of the adjacent sampling moments before and after.

[0074] Calculate the ratio of the probability density of chloride ion content at each sampling point at each sampling time to the probability density of chloride ion content at the first sampling time as the probability ratio of chloride ion content at each sampling point at each sampling time, and use the array composed of the numerical values ​​of the probability ratio of chloride ion content at each sampling point at each sampling time as the time series of the probability ratio of chloride ion content at the sampling point.

[0075] Calculate the ratio of the porosity probability density of each sampling point at each sampling moment to the porosity probability density of the sampling point at the first sampling moment as the porosity probability ratio of each sampling point at each sampling moment, and use the array consisting of the values ​​of the porosity probability ratio of each sampling point at each sampling moment as the porosity probability ratio time series of the sampling point.

[0076] For the probability ratio of chloride ion content and porosity, the higher the chloride ion content, the faster the corrosion rate; the porosity of the coating reflects the density of the coating. Therefore, it is necessary to compare the probability density of the two with the first sampling moment, so that it is easier to find the increase of the probability of chloride ion content and porosity over time.

[0077] This is because chloride ions can destroy the passive film on the surface of steel, forming corrosion pits, which leads to the intensification of local corrosion. Low porosity means that the coating is denser and can more effectively block the penetration of corrosive media (such as water, oxygen and chloride ions), thereby improving corrosion resistance.

[0078] Furthermore, in the method of combining the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point linearly and synchronously to obtain the corrosion electron particle loss array of the sampling point:

[0079] The linear and unidirectional time series combination means combining the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point in the same direction of the time series in a manner that maintains a consistent linear relationship and outputs them into an array of constant dimensional size. The corrosion particle loss array is an array that can be used to represent the linear relationship between the degree of change of the surface charge density probability amplitude, chloride ion content probability ratio and porosity probability ratio of each sampling point in the time direction at the sampling moment.

[0080] The method of combining the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point in the same direction of the time series in a way that maintains a consistent linear relationship and outputting them into an array of constant dimensional size can be specifically as follows:

[0081] In some embodiments, in order to keep the three time series in the same direction of the time series to maintain a consistent linear relationship, if it is required to improve the effectiveness of measuring the linear relationship between the degree of change of the surface charge density probability amplitude, the chloride ion content probability ratio and the porosity probability ratio of each sampling point in the time direction of the sampling moment, it is preferred that the values ​​of each dimension in the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point can be aligned according to the time sequence of each sampling moment, and then the values ​​of each dimension in the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point are multiplied by dimension, and after the multiplication, they are combined and output to form an array whose dimension size remains unchanged from the original surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series, as the corrosion electron loss array of the sampling point;

[0082] In some embodiments, if the calculation speed is to be improved, it is preferred to align the values ​​of each dimension in the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point according to the time sequence of each sampling moment, and then take the average of the values ​​of each dimension in the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point by dimension, and take the array composed of the average of the values ​​of each dimension by dimension as the corrosion electron loss array of the sampling point;

[0083] It is worth noting that in the embodiment, the cross product should not be used between the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series to calculate the corrosion electron loss array of the sampling point, because the nature of the cross product will lead to the collapse between different dimensions and is not conducive to measuring the linear relationship between the degree of change of the surface charge density probability amplitude, the chloride ion content probability ratio and the porosity probability ratio in the time direction at the sampling moment.

[0084] Furthermore, the corrosion electron loss array corresponding to each sampling point is formed into a corrosion electron loss matrix, the corrosion electron loss characteristic vector is calculated, and the corrosion electron loss equivalence between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector is calculated, specifically:

[0085] The corrosion electron loss arrays corresponding to each sampling point are used as the columns of the matrix to form a matrix, and the formed matrix is ​​used as the corrosion electron loss matrix;

[0086] Calculate the eigenvector of the corrosion electron loss matrix as the corrosion electron loss eigenvector corresponding to the sampling point on the surface of the coating;

[0087] The cosine similarity between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss feature vector is calculated as the corrosion electron loss similitude corresponding to each sampling point.

[0088] In some embodiments, less preferably, the value of the correlation coefficient between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss feature vector, including but not limited to the Pearson correlation coefficient, etc., can also be used as the value of the corrosion electron loss phase corresponding to each sampling point. However, when using Python code to simulate the calculation program of the corrosion electron loss phase corresponding to each sampling point, it is found that the output value obtained by running the program using the correlation coefficient is distributed relatively flat in variance, while the value distribution variance obtained by using the cosine similarity is relatively large, which has more prominent data characteristics. In the subsequent sensitivity analysis, the recall rate of the cosine group (cosine similarity control group) is also superior to that of the Pearson group (Pearson coefficient control group). In theoretical analysis, because cosine similarity measures the directional distribution in high-dimensional space, it is beneficial to the distribution of the directional differences in the time direction of the sampling time in high-dimensional space, because the linear relationship between the surface charge density, the chloride ion content and the porosity is inconsistent with the expression of the loss degree of the corrosion-resistant composite coating on the steel surface. Because coating loss may show a linear relationship with one or more variables. For example, the coating loss rate may be positively correlated with the chloride ion content and negatively correlated with the porosity. In more cases, these relationships are nonlinear and may require the use of complex mathematical models or machine learning algorithms to capture. Coating loss is usually the result of multiple factors, so it is necessary to comprehensively consider data such as surface charge density, chloride ion content, porosity, etc. to accurately predict the corrosion resistance of the coating.

[0089] Furthermore, the method for early warning of loss according to the corrosion particle loss phase corresponding to each sampling point is:

[0090] The sampling point with a larger corrosion particle loss phase value indicates a greater probability of coating loss. The sampling point with a corrosion particle loss phase value exceeding the threshold is marked as a loss warning point, and the loss warning point is output as a signal.

[0091] In some embodiments, the values ​​of the corrosion electron loss phase corresponding to each sampling point can be counted and analyzed to obtain a first threshold value. Preferably, the median of the values ​​of the corrosion electron loss phase corresponding to each sampling point is used as the first threshold value.

[0092] Furthermore, the method for early warning of loss according to the corrosion electron loss phase corresponding to each sampling point also includes:

[0093] According to the values ​​of the corrosion electron loss phase corresponding to each sampling point, the surface of the coating is divided into sub-areas, and the sub-areas where the cumulative values ​​of the corrosion electron loss phase corresponding to the sampling points exceed the threshold are marked as requiring coating reinforcement.

[0094] In some embodiments, an image segmentation algorithm can be used to segment the surface of the coating into multiple different sub-regions. The segmented sub-regions may or may not partially overlap with each other. The values ​​of the corrosion electron loss phase corresponding to the sampling points in each sub-region are accumulated as the regional loss cumulative value of each sub-region. A second threshold is calculated based on the regional loss cumulative value of the sub-region, and the sub-regions whose regional loss cumulative value exceeds the second threshold are coated with reinforcement.

[0095] In some embodiments, the method of the present invention may further include the following implementation code:

[0096] # Calculate the probability amplitude time series and probability ratio time series

[0097] def calculate_sequences(data):

[0098] sequences = {}

[0099] for point, values ​​in data.items():

[0100] charge_density_seq = np.array(values["surface_charge_density"])

[0101] chloride_ion_seq = np.array(values["chloride_ion_content"])

[0102] porosity_seq = np.array(values["porosity"])

[0103] # Probability amplitude time series

[0104] charge_density_prob_wave = charge_density_seq / np.max(charge_density_seq)

[0105] chloride_ion_prob_ratio = chloride_ion_seq / np.max(chloride_ion_seq)

[0106] porosity_prob_ratio = porosity_seq / np.max(porosity_seq)

[0107] sequences[point] = {

[0108] "charge_density_prob_wave": charge_density_prob_wave,

[0109] "chloride_ion_prob_ratio": chloride_ion_prob_ratio,

[0110] "porosity_prob_ratio": porosity_prob_ratio

[0111] }

[0112] return sequences

[0113] sequences = calculate_sequences(data)

[0114] # Calculate the corrosion electric particle loss array and the corrosion electric particle loss matrix

[0115] def calculate_corrosion_loss_matrix(sequences):

[0116] corrosion_loss_matrix = []

[0117] for point, seqs in sequences.items():

[0118] corrosion_loss_array = seqs["charge_density_prob_wave"] + \

[0119] seqs["chloride_ion_prob_ratio"] - \

[0120] seqs["porosity_prob_ratio"]

[0121] corrosion_loss_matrix.append(corrosion_loss_array)

[0122] return np.array(corrosion_loss_matrix).T

[0123] corrosion_loss_matrix = calculate_corrosion_loss_matrix(sequences)

[0124] # Calculate eigenvectors and correlations

[0125] def calculate_eigenvector_and_similarity(corrosion_loss_matrix):

[0126] eig_values, eig_vectors = np.linalg.eig(np.cov(corrosion_loss_matrix.T))

[0127] dominant_eig_vector = eig_vectors[0, np.argmax(eig_values)]

[0128] similarities = []

[0129] for i in range(corrosion_loss_matrix.shape[1]):

[0130] similarity = np.dot(corrosion_loss_matrix[0, i], dominant_eig_vector) / \

[0131] (np.linalg.norm(corrosion_loss_matrix[0, i]) * np.linalg.norm(dominant_eig_vector))

[0132] similarities.append(similarity)

[0133] return dominant_eig_vector, similarities

[0134] dominant_eig_vector, similarities = calculate_eigenvector_and_similarity(corrosion_loss_matrix)

[0135] The above codes are partial programs intercepted from the complete project and do not constitute a limitation on the implementation method of the method described in the present invention.

[0136] The loss warning system for the corrosion-resistant composite coating on the surface of steel runs on any computing device such as a desktop computer, a laptop computer, a mobile phone, a PDA or a cloud data center. The computing device includes: a processor, a memory and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps in the loss warning method for the corrosion-resistant composite coating on the surface of steel are implemented. The executable system may include, but is not limited to, a processor, a memory, and a server cluster.

[0137] The embodiment of the present disclosure provides a loss warning system for a corrosion-resistant composite coating on a steel surface, such as Figure 2 As shown, a loss warning system for a corrosion-resistant composite coating on a steel surface of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned loss warning method embodiment for a corrosion-resistant composite coating on a steel surface are implemented. The processor executes the computer program to run in the following system units:

[0138] A data sampling unit is used to set a plurality of different sampling points on the surface of the coating, and to sample and obtain data of surface charge density, chloride ion content and porosity at each sampling point at a plurality of different sampling moments, wherein the number of sampling points is consistent with the number of sampling moments;

[0139] A probability calculation unit is used to calculate the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point on the surface of the coating at each sampling time according to the surface charge density, chloride ion content and porosity data corresponding to each sampling point on the surface of the coating at each sampling time;

[0140] A time series combination unit is used to linearly and synchronously combine the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point to obtain the corrosion electron particle loss array of the sampling point;

[0141] A loss calculation unit, used to calculate a corrosion electron loss characteristic vector by forming a corrosion electron loss array corresponding to each sampling point into a corrosion electron loss matrix, and calculate the corrosion electron loss equivalence between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector;

[0142] The loss warning unit is used to provide a loss warning according to the corrosion particle loss phase corresponding to each sampling point. Preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.

[0143] Among them, dimensionless numerical calculations are used between physical quantities of different units.

[0144] The loss warning system for the corrosion-resistant composite coating on the surface of steel can be run in computing devices such as desktop computers, laptops, mobile phones, PDAs and cloud data centers. The loss warning system for the corrosion-resistant composite coating on the surface of steel includes, but is not limited to, processors and memories. Those skilled in the art will understand that the example is only an example of a loss warning method and system for a corrosion-resistant composite coating on the surface of steel, and does not constitute a limitation on a loss warning method and system for a corrosion-resistant composite coating on the surface of steel, and may include more or fewer components than the example, or a combination of certain components, or different components. For example, the loss warning system for the corrosion-resistant composite coating on the surface of steel may also include input and output devices, network access devices, buses, etc.

[0145] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete component gate circuits or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the loss warning system for the corrosion-resistant composite coating on the surface of steel, and uses various interfaces and lines to connect the various sub-areas of the loss warning system for the corrosion-resistant composite coating on the surface of steel.

[0146] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the loss warning method and system of the corrosion-resistant composite coating on the surface of steel by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0147] The present invention provides a method for early warning of the loss of corrosion-resistant composite coatings on steel surfaces. Multiple sampling points are set on the coating surface to obtain surface charge density, chloride ion content and porosity data at different sampling times. Based on these data, the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series of each sampling point are calculated, and they are linearly combined in the same direction to generate a corrosion electron loss array. By calculating the eigenvector of the corrosion electron loss matrix and the correlation of the corrosion electron loss of each sampling point, early warning of the loss of the coating is given. Potential problems can be discovered and dealt with in a timely manner in the early stages of corrosion of the composite coating, thereby extending the service life of the steel structure protected by the coating and reducing the waste of resources of seawater steel structures.

[0148] Although the description of the present disclosure has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present disclosure. In addition, the above description of the present disclosure is based on the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present disclosure that have not yet been foreseen may still represent equivalent changes to the present disclosure.

Claims

1. A method for early warning of loss of corrosion-resistant composite coating on steel surface, characterized in that: The method comprises the following steps: A plurality of different sampling points are arranged on the surface of the coating, and at a plurality of different sampling moments, the surface charge density, chloride ion content and porosity data at each sampling point are sampled and obtained respectively; wherein the number of sampling points is consistent with the number of sampling moments; wherein the value of the chloride ion content is the value of the concentration or density of the content with chloride ions per unit area calculated corresponding to the surface charge density, or the value of the concentration or density of the content with chloride ions per unit volume calculated corresponding to the porosity; According to the data of surface charge density, chloride ion content and porosity corresponding to each sampling point on the surface of the coating at each sampling time, the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point are calculated, specifically: at the same sampling time, the values ​​of the surface charge density, chloride ion content and porosity corresponding to each sampling point are obtained; the probability distribution of the value of the surface charge density of each sampling point at each sampling time relative to the value of the surface charge density of the sampling point at all sampling times is extracted, so as to obtain the surface charge density probability density of each sampling point at each sampling time; the probability distribution of the value of the chloride ion content of each sampling point at each sampling time relative to the value of the chloride ion content of the sampling point at all sampling times is extracted, so as to obtain the chloride ion content probability density of each sampling point at each sampling time; the probability distribution of the value of the porosity of each sampling point at each sampling time relative to the value of the porosity of the sampling point at all sampling times is extracted, so as to obtain the porosity probability density of each sampling point at each sampling time; calculate The fluctuation range of the surface charge density probability density of each sampling point at each sampling moment relative to the surface charge density probability density at the adjacent sampling moments is taken as the surface charge density probability amplitude of each sampling point at each sampling moment, and the array consisting of the numerical values ​​of the surface charge density probability amplitude of each sampling point at each sampling moment is taken as the surface charge density probability amplitude time series of the sampling point; the ratio of the chloride ion content probability density of each sampling point at each sampling moment to the chloride ion content probability density of the sampling point at the first sampling moment is calculated as the chloride ion content probability ratio of each sampling point at each sampling moment, and the array consisting of the numerical values ​​of the chloride ion content probability ratio of each sampling point at each sampling moment is taken as the chloride ion content probability ratio time series of the sampling point; the ratio of the porosity probability density of each sampling point at each sampling moment to the porosity probability density of the sampling point at the first sampling moment is calculated as the porosity probability ratio of each sampling point at each sampling moment, and the array consisting of the numerical values ​​of the porosity probability ratio of each sampling point at each sampling moment is taken as the porosity probability ratio time series of the sampling point; The surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point are linearly combined in the same direction to obtain the corrosion particle loss array of the sampling point, specifically including: aligning the values ​​of each dimension in the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point according to the time sequence of each sampling moment, and then multiplying the values ​​of each dimension in the surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series corresponding to each sampling point by dimension, and combining the output after multiplication to form a dimension with the same size as the original surface charge density probability amplitude time series, chloride ion content probability ratio time series and porosity probability ratio time series. The array in which the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series remain unchanged is used as the corrosion electron particle loss array of the sampling point; or the values ​​of each dimension in the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point are aligned according to the time sequence of each sampling moment, and then the values ​​of each dimension in the surface charge density probability amplitude time series, the chloride ion content probability ratio time series and the porosity probability ratio time series corresponding to each sampling point are averaged by dimension, and the array consisting of the average values ​​of each dimension by dimension is used as the corrosion electron particle loss array of the sampling point; The corrosion electron loss arrays corresponding to each sampling point are combined into a corrosion electron loss matrix, the corrosion electron loss characteristic vector is calculated, and the corrosion electron loss equivalence between the corrosion electron loss arrays corresponding to each sampling point and the corrosion electron loss characteristic vector is calculated, specifically: the corrosion electron loss arrays corresponding to each sampling point are used as the columns of the matrix to form a matrix, and the formed matrix is ​​used as the corrosion electron loss matrix; the characteristic vector of the corrosion electron loss matrix is ​​calculated as the corrosion electron loss characteristic vector corresponding to the sampling point on the surface of the coating; the cosine similarity between the corrosion electron loss array corresponding to each sampling point and the corrosion electron loss characteristic vector is calculated as the corrosion electron loss equivalence corresponding to each sampling point; Loss warning is carried out according to the corrosion particle loss phase corresponding to each sampling point.

2. The method for early warning of loss of corrosion-resistant composite coating on steel surface according to claim 1, characterized in that: in, The method of setting a plurality of different sampling points on the surface of the coating is as follows: homogeneous sampling is performed on the surface of the coating to obtain a plurality of non-overlapping sampling points, and the serial numbers of the sampling points remain consistent at each sampling moment.

3. The method for early warning of loss of corrosion-resistant composite coating on steel surface according to claim 2, characterized in that: in, The method for early warning of loss based on the phase of corrosion particle loss corresponding to each sampling point is: The sampling point with a larger corrosion particle loss phase value indicates a greater probability of coating loss. The sampling point with a corrosion particle loss phase value exceeding the threshold is marked as a loss warning point, and the loss warning point is output as a signal.

4. A method for early warning of loss of corrosion-resistant composite coating on steel surface according to claim 1 or 2, characterized in that: in, The method for early warning of loss according to the corrosion particle loss phase corresponding to each sampling point also includes: According to the values ​​of the corrosion electron loss phase corresponding to each sampling point, the surface of the coating is divided into sub-areas, and the sub-areas where the cumulative values ​​of the corrosion electron loss phase corresponding to the sampling points exceed the threshold are marked as requiring coating reinforcement.

5. A loss warning system for corrosion-resistant composite coating on steel surface, characterized in that: The loss warning system for the corrosion-resistant composite coating on the surface of steel runs in any computing device such as a desktop computer, a laptop computer or a cloud data center. The computing device includes: a processor, a memory and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps in the loss warning method for the corrosion-resistant composite coating on the surface of steel are implemented as described in any one of claims 1 to 3.