Method and device for predicting corrosion risk of steel structural member of coastal photovoltaic device

Abstract

This application belongs to the field of photovoltaic technology and discloses a method and device for predicting corrosion risk of steel structural components of coastal photovoltaic equipment. By acquiring multi-dimensional quality inspection data of pre-coating and environmental stress parameters, the corrosion risk level is predicted using a model, and then the period of low corrosion risk is predicted. This method can predict the corrosion risk in the later stages of installation of steel structural components of coastal photovoltaic equipment in the early stages, which is conducive to ensuring the long-term stable operation of coastal photovoltaic equipment.

Description

Technical Field

[0001] This application relates to the field of photovoltaic technology, and more specifically, to a method and apparatus for predicting corrosion risk of steel structural components of coastal photovoltaic equipment. Background Technology

[0002] Coastal photovoltaic (PV) systems, as crucial clean energy infrastructure, are essential for ensuring reliable energy supply through long-term stable operation. However, the harsh corrosiveness of the marine environment poses a significant challenge to the steel structures of PV systems. To address corrosion, anti-corrosion coatings are typically applied to the prefabricated steel structures during the initial construction phase. However, during transportation, hoisting, and installation, these coatings inevitably suffer mechanical damage, such as scratches and abrasions. This damage significantly reduces the coating's anti-corrosion performance, threatening the long-term safe operation of the PV system. Therefore, a comprehensive assessment of the pre-coating quality after installation and accurate prediction of future corrosion risks are crucial for ensuring the long-term stable operation of PV systems.

[0003] Currently, traditional methods for inspecting the quality of pre-coatings mainly rely on manual visual inspection. This method is not only inefficient, but the results are also easily influenced by the subjective factors of the inspectors, lacking objectivity and quantitative standards. Therefore, it is difficult to accurately assess the true quality of the pre-coating, let alone use it to predict future corrosion risks. Furthermore, in the early stages of coastal photovoltaic equipment construction, due to the lack of long-term operational and corrosion data accumulation, directly applying traditional big data-driven machine learning algorithms for corrosion risk prediction faces the challenge of a severe shortage of data.

[0004] Therefore, in the initial, crucial stage of installing prefabricated steel structures for coastal photovoltaic (PV) equipment, under the extreme condition of lacking long-term operational and corrosion data, accurately predicting future corrosion risks based solely on limited multi-dimensional quality inspection data of the pre-coating obtained during this initial phase has become a critical technical challenge to ensure the construction quality and long-term stable operation of coastal PV equipment. Solving this problem has significant practical implications and application value for improving the corrosion resistance of coastal PV equipment, reducing operation and maintenance costs, and ensuring energy security.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] The purpose of this application is to provide a method and device for predicting corrosion risks of steel structural components of coastal photovoltaic equipment, which can predict the corrosion risks in the later stages of installation of steel structural components of coastal photovoltaic equipment, and is conducive to ensuring the long-term stable operation of coastal photovoltaic equipment.

[0007] Firstly, this application provides a method for predicting the corrosion risk of steel structural components in coastal photovoltaic equipment. This method includes the following steps:

[0008] A1. Obtain multi-dimensional quality inspection data of the pre-coating after the steel structure components are installed and environmental stress parameters of the coastal photovoltaic equipment location; the multi-dimensional quality inspection data of the pre-coating includes coating thickness, coating porosity and coating adhesion; environmental stress parameters include salt spray concentration, ultraviolet radiation intensity and temperature change range;

[0009] A2. Based on the multi-dimensional quality inspection data and environmental stress parameters of the pre-coating, a multi-factor coupled environmental stress accelerated aging model is used to predict the quality status of the pre-coating at different service time points; the quality status includes corrosion depth, UV degradation degree and adhesion.

[0010] A3. Based on the quality status of the pre-coating at different service time points, the corrosion risk level of the steel structure at different service time points is determined by the corrosion risk level assessment model;

[0011] A4. Compare the corrosion risk level of steel structural components at different service time points with the preset risk level threshold to predict the period of low corrosion risk for steel structural components.

[0012] This method acquires multi-dimensional quality inspection data and environmental stress parameters of the pre-coating, uses a model to predict the corrosion risk level, and then predicts the period of low corrosion risk. It can predict the corrosion risk in the early stage of the installation of steel structure components of coastal photovoltaic equipment, which is conducive to ensuring the long-term stable operation of coastal photovoltaic equipment.

[0013] Preferably, step A1 includes:

[0014] A101. For the surface of steel structural components, multiple inspection areas are divided, and corrosion-sensitive areas are selected from multiple inspection areas based on corrosion sensitivity criteria. Among them, corrosion sensitivity criteria include weld point proximity, edge exposure, and water accumulation risk.

[0015] A102. Obtain surface images of corrosion-sensitive areas and use image processing algorithms to identify the coating defect type and defect area of ​​each defect feature in the surface image; wherein, the coating defect types include scratches, cracks and bubbles;

[0016] A103. Based on the type and area of ​​coating defects, calculate the degree of coating damage in corrosion-sensitive areas, and determine the coating thickness, coating porosity, and coating adhesion of steel structural components based on the degree of coating damage in corrosion-sensitive areas.

[0017] A104. Based on historical environmental monitoring data of the coastal photovoltaic equipment locations, environmental stress parameters are extracted and combined with real-time environmental monitoring data to correct the environmental stress parameters, thus obtaining the corrected environmental stress parameters.

[0018] This enables the acquisition of more effective and accurate multi-dimensional quality inspection data and environmental stress parameters for pre-coating, providing a data foundation for subsequent corrosion risk prediction.

[0019] Preferably, step A101 includes:

[0020] Acquire 3D point cloud data of the surface of steel structural components, and divide the point cloud data into multiple detection areas based on a preset segmentation algorithm;

[0021] For each inspection area, the proximity of the welding point, the edge exposure, and the risk of water accumulation are obtained. The proximity of the welding point is obtained by calculating the reciprocal of the distance from the center point of the inspection area to the nearest welding point. The edge exposure is obtained by calculating the ratio of the boundary length of the inspection area to the area of ​​the area. The risk of water accumulation is obtained by simulating a rainfall process and calculating the water depth in the inspection area.

[0022] The proximity of weld points, edge exposure, and risk of water accumulation in each detection area are weighted and summed to obtain a comprehensive corrosion sensitivity score. Based on the comprehensive corrosion sensitivity score, the detection areas are sorted from high to low, and a preset number of detection areas are selected as corrosion sensitive areas.

[0023] Therefore, by calculating and ranking the comprehensive corrosion sensitivity score, automatic screening and efficient identification of corrosion-sensitive areas are achieved, overcoming the problems of low efficiency and strong subjectivity in manual screening of corrosion-sensitive areas. This ensures the accuracy and efficiency of corrosion-sensitive area identification and lays the foundation for subsequent pre-coating quality inspection and corrosion risk prediction.

[0024] Preferably, step A103 includes:

[0025] Based on the identified coating defect types and defect areas, the total defect area for each coating defect type is calculated.

[0026] Calculate the weighted sum of the total defect areas for each type of coating defect to obtain the degree of coating damage in corrosion-sensitive areas;

[0027] Using a nonlinear relationship model between coating damage degree and coating thickness, coating porosity and coating adhesion, the coating thickness, coating porosity and coating adhesion of the corrosion-sensitive area are calculated based on the coating damage degree of the corrosion-sensitive area.

[0028] The coating thickness, porosity, and adhesion of steel structural components are determined based on the coating thickness, porosity, and adhesion of the corrosion-sensitive areas.

[0029] Preferably, the multi-factor coupled environmental stress accelerated aging model includes a salt spray corrosion model, an ultraviolet degradation model, and a mechanical aging model. The salt spray corrosion model is based on electrochemical principles and considers the effects of chloride ion concentration, temperature, and humidity on the corrosion rate. The ultraviolet degradation model is based on photochemical reaction kinetics and considers the effects of ultraviolet radiation intensity, wavelength, and light absorption characteristics of the coating material on the degradation rate. The mechanical aging model is based on creep theory and considers the effects of stress level, temperature, and time on adhesion decay.

[0030] Step A2 includes:

[0031] A201. Based on the multi-dimensional quality inspection data of the pre-coating, initialize the parameters of the multi-factor coupled environmental stress accelerated aging model; the initialized parameters include coating thickness, porosity, adhesion, chloride ion diffusion coefficient, ultraviolet absorption coefficient, and creep coefficient;

[0032] A202. Based on the environmental stress parameters of the coastal photovoltaic equipment location, determine the environmental stress input for the multi-factor coupled environmental stress accelerated aging model; the environmental stress input includes the time-varying curves of salt spray concentration, ultraviolet radiation intensity, and temperature.

[0033] A203. Iteratively solve the multi-factor coupled environmental stress accelerated aging model with a preset time step to obtain the quality status of the pre-coated layer at different service time points; the quality status includes corrosion depth, UV degradation degree and adhesion.

[0034] Preferably, step A3 includes:

[0035] A301. Obtain data on corrosion influencing factors of steel structural components; the corrosion influencing factor data includes the quality status of the pre-coating at different service time points and environmental factor data of the coastal photovoltaic equipment location; the environmental factor data includes salt spray concentration, temperature and humidity;

[0036] A302. Based on the data of corrosion influencing factors and the corrosion risk level assessment model, calculate the corrosion risk score of steel structural components at different service time points, and map the corrosion risk score to the corresponding corrosion risk level.

[0037] Preferably, the corrosion risk level assessment model includes a corrosion risk scoring model and an environmental factor risk scoring model; the corrosion risk scoring model is used to calculate the quality status risk score based on the pre-coating quality status parameters, and the environmental factor risk scoring model is used to calculate the environmental factor risk score based on the environmental factor parameters; the pre-coating quality status parameters include corrosion depth, UV degradation degree and adhesion, and the environmental factor parameters include salt spray concentration, temperature and humidity;

[0038] Step A302 includes:

[0039] Based on the quality status of the pre-coating at different service time points, pre-coating quality status parameters are extracted, and environmental factor parameters are extracted based on environmental factor data of the coastal photovoltaic equipment location.

[0040] Based on the corrosion risk scoring model and the environmental factor risk scoring model, the quality status risk score corresponding to the extracted pre-coating quality status parameters and the environmental factor risk score corresponding to the extracted environmental factor parameters are calculated respectively.

[0041] A comprehensive corrosion risk score is obtained by integrating the quality status risk score and the environmental factor risk score.

[0042] The corresponding corrosion risk level is determined based on the comprehensive corrosion risk score.

[0043] Preferably, step A4 includes:

[0044] A401. Based on the corrosion risk level of steel structural components at different service times, construct time-series corrosion risk level curves;

[0045] A402. Based on the preset risk level threshold, determine the time point at which the corrosion risk level curve first exceeds the risk level threshold, and use it as the initial threshold exceedance time point;

[0046] A403. Starting from the initial time point exceeding the threshold, backtrack and search for the time point in the time series corrosion risk level curve that is first lower than the risk level threshold, and take it as the end point of the low corrosion risk period.

[0047] A404. Calculate the time difference between the end of the low corrosion risk period and the time when the steel structure is installed, to obtain the low corrosion risk period of the steel structure.

[0048] Preferably, after step A401 and before step A402, the following step is further included:

[0049] A401a. For each extreme weather event time point, analyze the type and intensity of the extreme weather event, and calculate the degree of impact of the corresponding extreme weather event on the corrosion risk level;

[0050] A401b. Based on the calculated impact of the corrosion risk level, correct the corrosion risk level after the corresponding extreme weather event time point in the time series corrosion risk level curve.

[0051] Secondly, this application provides a corrosion risk prediction device for steel structural components of coastal photovoltaic equipment, used to predict the corrosion risk of steel structural components of coastal photovoltaic equipment. The device includes:

[0052] The data acquisition module is used to acquire multi-dimensional quality inspection data of the pre-coating after the steel structure components are installed and environmental stress parameters of the coastal photovoltaic equipment location. The multi-dimensional quality inspection data of the pre-coating includes coating thickness, coating porosity and coating adhesion. The environmental stress parameters include salt spray concentration, ultraviolet radiation intensity and temperature change range.

[0053] The aging prediction module is used to predict the quality status of the pre-coating at different service time points based on multi-dimensional quality inspection data and environmental stress parameters of the pre-coating and using a multi-factor coupled environmental stress accelerated aging model. The quality status includes corrosion depth, UV degradation degree and adhesion.

[0054] The risk assessment module is used to determine the corrosion risk level of steel structural components at different service times based on the quality status of the pre-coating at different service times using a corrosion risk level assessment model.

[0055] The risk warning module is used to compare the corrosion risk level of steel structural components at different service points with the preset risk level threshold, and predict the period of low corrosion risk for the steel structural components.

[0056] Beneficial effects: The corrosion risk prediction method and device for steel structural components of coastal photovoltaic equipment provided in this application obtains multi-dimensional quality inspection data of pre-coating and environmental stress parameters, uses a model to predict the corrosion risk level, and then predicts the period of low corrosion risk. It can predict the corrosion risk in the later stage of the installation of steel structural components of coastal photovoltaic equipment in the early stage, which is conducive to ensuring the long-term stable operation of coastal photovoltaic equipment. Attached Figure Description

[0057] Figure 1 A flowchart illustrating the corrosion risk prediction method for steel structural components of coastal photovoltaic equipment provided in this application embodiment.

[0058] Figure 2 A schematic diagram of the corrosion risk prediction device for steel structural components of coastal photovoltaic equipment provided in this application embodiment.

[0059] Labeling Explanation: 1. Data Acquisition Module; 2. Aging Prediction Module; 3. Risk Assessment Module; 4. Risk Warning Module. Detailed Implementation

[0060] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0061] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0062] refer to Figure 1 This application proposes a method for predicting the corrosion risk of steel structural components in coastal photovoltaic equipment. The method includes the following steps:

[0063] A1. Obtain multi-dimensional quality inspection data of the pre-coating after the steel structure components are installed and environmental stress parameters of the coastal photovoltaic equipment location; the multi-dimensional quality inspection data of the pre-coating includes coating thickness, coating porosity and coating adhesion; environmental stress parameters include salt spray concentration, ultraviolet radiation intensity and temperature change range;

[0064] A2. Based on the multi-dimensional quality inspection data and environmental stress parameters of the pre-coating, a multi-factor coupled environmental stress accelerated aging model is used to predict the quality status of the pre-coating at different service time points; the quality status includes corrosion depth, UV degradation degree and adhesion.

[0065] A3. Based on the quality status of the pre-coating at different service time points, the corrosion risk level of the steel structure at different service time points is determined by the corrosion risk level assessment model;

[0066] A4. Compare the corrosion risk level of steel structural components at different service time points with the preset risk level threshold to predict the period of low corrosion risk for steel structural components.

[0067] In step A1, multi-dimensional quality inspection data and environmental stress parameters of the pre-coating are acquired, providing a data foundation for subsequent corrosion risk prediction. The multi-dimensional quality inspection data includes coating thickness, coating porosity, and coating adhesion, reflecting the initial quality state of the pre-coating. Environmental stress parameters include salt spray concentration, ultraviolet radiation intensity, and temperature variation range, representing the severity of the corrosive environment in which the steel structure is located.

[0068] In step A2, a multi-factor coupled environmental stress accelerated aging model is used, combined with the data obtained in step A1, to predict the quality status of the pre-coating at different service time points. The quality status is specifically manifested in corrosion depth, UV degradation degree, and adhesion. This step simulates the aging process of the pre-coating under the coupled effects of various environmental stresses through model calculations, thus achieving the prediction of future quality status.

[0069] In step A3, based on the quality status predicted in step A2 at different service time points, the corrosion risk level of the steel structural component at different service time points is determined using a corrosion risk level assessment model. This step quantifies the quality status of the pre-coating into a corrosion risk level, thus achieving corrosion risk assessment.

[0070] In step A4, the corrosion risk level at different service time points obtained in step A3 is compared with the preset risk level threshold to predict the low corrosion risk period of the steel structure. This step ultimately outputs the service life of the steel structure under low corrosion risk, providing a reference for operation and maintenance decisions (for example, operation and maintenance decisions may include periodically conducting on-site inspections of the steel structure at a first inspection frequency during the low corrosion risk period, and periodically conducting on-site inspections of the steel structure at a second inspection frequency after the low corrosion risk period, with the first inspection frequency being lower than the second inspection frequency, thereby reducing the maintenance workload during the low corrosion risk period while ensuring low risk).

[0071] Specifically, this technical solution obtains initial data in step A1, predicts future quality status using a model in step A2, assesses corrosion risk level in step A3, and predicts the period of low corrosion risk in step A4. By predicting future risks from the initial state, it solves the technical problem of accurately predicting corrosion risks in the early stages of installation of steel structural components for coastal photovoltaic equipment and provides a reference for operation and maintenance decisions. The coordinated work of each step enables effective prediction of corrosion risks for steel structural components of coastal photovoltaic equipment.

[0072] In some preferred embodiments, step A1 includes:

[0073] A101. For the surface of steel structural components, multiple inspection areas are divided, and corrosion-sensitive areas are selected from multiple inspection areas based on corrosion sensitivity criteria. Among them, corrosion sensitivity criteria include weld point proximity, edge exposure, and water accumulation risk.

[0074] A102. Obtain surface images of corrosion-sensitive areas and use image processing algorithms to identify the coating defect type and defect area of ​​each defect feature in the surface image; wherein, the coating defect types include scratches, cracks and bubbles;

[0075] A103. Based on the type and area of ​​coating defects, calculate the degree of coating damage in corrosion-sensitive areas, and determine the coating thickness, coating porosity, and coating adhesion of steel structural components based on the degree of coating damage in corrosion-sensitive areas.

[0076] A104. Based on historical environmental monitoring data of the coastal photovoltaic equipment locations, extract environmental stress parameters (i.e., environmental monitoring data includes various environmental stress parameters), and combine them with real-time environmental monitoring data to correct the environmental stress parameters, thus obtaining the corrected environmental stress parameters.

[0077] Specifically, for step A101, the detection area can be divided using a mesh generation method, uniformly dividing the surface of the steel structure into multiple rectangular regions, or an irregular region division method can be used, for example, dividing the surface into multiple polygonal regions based on the geometry and structural characteristics of the steel structure. The weld point proximity criterion for corrosion sensitivity can be quantified as the reciprocal of the distance from the center point of the detection area to the nearest weld point; the smaller the distance, the higher the weld point proximity, and the higher the corrosion sensitivity. Edge exposure can be quantified as the ratio of the boundary length of the detection area to the area of ​​the region; the larger the ratio, the higher the edge exposure, and the higher the corrosion sensitivity. Water accumulation risk can be calculated by simulating rainfall processes, such as using finite element analysis or computational fluid dynamics simulations. The greater the water accumulation depth, the higher the water accumulation risk (the calculated water accumulation depth can be directly used as the water accumulation risk, or a conversion formula can be used to convert the water accumulation depth into the water accumulation risk, in which the water accumulation risk is proportional to the water accumulation depth), and the higher the corrosion sensitivity. Each corrosion sensitivity criterion can be assigned a weight, and then a weighted sum is performed to obtain a comprehensive corrosion sensitivity score. Corrosion-sensitive areas are then screened based on the comprehensive corrosion sensitivity score.

[0078] For step A102, the surface image can be acquired by capturing images with an industrial camera. Image processing algorithms can include image segmentation, defect detection, and feature extraction. Image segmentation algorithms divide the coating surface image into different regions, such as defective and non-defective regions. Defect detection algorithms detect defect features in the image, such as scratches, cracks, and bubbles. Feature extraction algorithms extract parameters such as the type and area of ​​the defect features.

[0079] Specifically, for step A103, the degree of coating damage can be defined as the weighted sum of the total defect areas for each type of coating defect. The weights can be determined based on the degree of influence of different defect types on corrosion risk; for example, scratches have a weight of 0.5, cracks have a weight of 0.8, and bubbles have a weight of 0.3. The coating thickness, coating porosity, and coating adhesion of the steel structural components can be calculated using a nonlinear relationship model between the degree of coating damage and these parameters. This nonlinear relationship model can be obtained through fitting or training experimental data, for example, using a neural network model or a support vector machine model.

[0080] Regarding step A104, historical environmental monitoring data can be obtained from meteorological departments or environmental monitoring stations, while real-time environmental monitoring data can be acquired in real time through environmental sensors installed at the locations of coastal photovoltaic equipment. The correction of environmental stress parameters can employ Kalman filtering or particle filtering algorithms, fusing historical and real-time data to improve the accuracy of environmental stress parameters.

[0081] Specifically, step A101 enables targeted selection of the detection area. Corrosion-sensitive areas are screened using corrosion sensitivity criteria, concentrating detection resources on areas with high corrosion risk and improving detection efficiency and targeting. Step A102 enables automatic identification and quantification of coating defects. Image processing algorithms automatically identify coating defect types and areas, avoiding the subjectivity and inefficiency of manual visual inspection and improving the objectivity and accuracy of detection results. Step A103 enables the quantification of coating quality detection data. The degree of coating damage is calculated based on defect type and area, and quantitative indicators such as coating thickness, porosity, and adhesion are determined, providing a data foundation for subsequent corrosion risk prediction. Step A104 enables accurate acquisition of environmental stress parameters. Environmental stress parameters are corrected based on historical and real-time data, improving their accuracy and real-time performance, providing accurate environmental input for subsequent corrosion risk prediction. Therefore, the technical solution in step A1 enables more effective and accurate acquisition of multi-dimensional pre-coating quality detection data and environmental stress parameters, providing a data foundation for subsequent corrosion risk prediction.

[0082] In some possible implementations, step A101 includes:

[0083] Acquire 3D point cloud data of the surface of steel structural components, and divide the point cloud data into multiple detection areas based on a preset segmentation algorithm;

[0084] For each inspection area, the proximity of the welding point, the edge exposure, and the risk of water accumulation are obtained. The proximity of the welding point is obtained by calculating the reciprocal of the distance from the center point of the inspection area to the nearest welding point. The edge exposure is obtained by calculating the ratio of the boundary length of the inspection area to the area of ​​the area. The risk of water accumulation is obtained by simulating a rainfall process and calculating the water depth in the inspection area.

[0085] The proximity of weld points, edge exposure, and risk of water accumulation in each detection area are weighted and summed to obtain a comprehensive corrosion sensitivity score. Based on the comprehensive corrosion sensitivity score, the detection areas are sorted from high to low, and a preset number of detection areas are selected as corrosion sensitive areas.

[0086] The acquisition of 3D point cloud data can be achieved using 3D scanners, such as laser scanners or structured light scanners. These devices can quickly acquire the 3D geometric information of the surface of steel structural components. The preset segmentation algorithm can be a region-growing algorithm or a clustering algorithm, such as the K-means algorithm. These algorithms can divide the point cloud data into multiple spatially continuous regions, each of which serves as a detection area, facilitating subsequent corrosion sensitivity analysis of each region.

[0087] Specifically, the calculation of weld point proximity involves first determining the three-dimensional coordinates of all weld points on the steel structure, then calculating the distance from the center point of each inspection area to all weld points, and selecting the reciprocal of the minimum distance as the weld point proximity. The calculation of edge exposure involves first extracting the boundary points of each inspection area, then calculating the total length of the boundary points, and finally calculating the area of ​​the inspection area. The ratio of the boundary length to the area is the edge exposure. The calculation of water accumulation risk can be achieved by simulating rainfall processes, such as using computer simulations to simulate the flow and accumulation of rainfall on the surface of the steel structure, calculating the water depth in each inspection area, and quantifying the water depth as water accumulation risk (either directly using the calculated water depth as the water accumulation risk, or converting the water depth to water accumulation risk using a conversion formula where the water accumulation risk is proportional to the water depth). In the weighted summation of weld point proximity, edge exposure, and water accumulation risk for each inspection area, weights can be preset according to the actual situation to obtain a comprehensive corrosion sensitivity score.

[0088] The preset number can be an absolute number or a percentage of the total number of detection areas. Selecting a preset number of detection areas as corrosion-sensitive areas, for example, can select the top 10 or top 20% of detection areas in the comprehensive corrosion sensitivity score as corrosion-sensitive areas, so as to facilitate subsequent key detection and risk assessment.

[0089] Specifically, through the application of 3D point cloud data acquisition and segmentation algorithms, the surface of steel structural components was effectively divided into multiple inspection areas, achieving automatic division and refined management of these areas. By quantitatively calculating three corrosion sensitivity criteria—weld point proximity, edge exposure, and water accumulation risk—an objective assessment of the corrosion sensitivity of each inspection area was achieved. Thus, the calculation and ranking of comprehensive corrosion sensitivity scores enabled automatic screening and efficient identification of corrosion-sensitive areas, overcoming the inefficiency and subjectivity inherent in manual screening. This ensured the accuracy and efficiency of corrosion-sensitive area identification, laying the foundation for subsequent pre-coating quality inspection and corrosion risk prediction.

[0090] In some possible implementations, step A103 includes:

[0091] Based on the identified coating defect types and defect areas, the total defect area for each coating defect type is calculated.

[0092] Calculate the weighted sum of the total defect areas for each type of coating defect to obtain the degree of coating damage in corrosion-sensitive areas;

[0093] Using a nonlinear relationship model between coating damage degree and coating thickness, coating porosity and coating adhesion, the coating thickness, coating porosity and coating adhesion of the corrosion-sensitive area are calculated based on the coating damage degree of the corrosion-sensitive area.

[0094] The coating thickness, porosity, and adhesion of steel structural components are determined based on the coating thickness, porosity, and adhesion of the corrosion-sensitive areas.

[0095] The calculation of the weighted sum of the total defect areas for each type of coating defect can be performed by pre-setting weighting coefficients for different coating defect types such as scratches, cracks, and bubbles. The weighting coefficient for scratches can be set to 0.3, for cracks to 0.7, and for bubbles to 0.1, but is not limited to these values. Therefore, by multiplying the total defect area of ​​each type of coating defect by its corresponding weighting coefficient and summing the results, a quantitative value of the coating damage degree in corrosion-sensitive areas can be obtained.

[0096] The nonlinear relationship model between coating damage degree and coating thickness, coating porosity, and coating adhesion can include three neural network models. The input to each of the three models is set to the coating damage degree, and the outputs are coating thickness, coating porosity, and coating adhesion, respectively. By training each neural network model with pre-collected experimental data, the nonlinear mapping relationship between coating damage degree and coating thickness, coating porosity, and coating adhesion can be obtained.

[0097] When determining the coating thickness, porosity, and adhesion of steel structural components based on the coating thickness, porosity, and adhesion of corrosion-sensitive areas, the coating thickness, porosity, and adhesion of the corrosion-sensitive areas can be directly used as the coating thickness, porosity, and adhesion of the steel structural components. Alternatively, redundant calculations can be performed based on the coating thickness, porosity, and adhesion of the corrosion-sensitive areas to obtain the coating thickness, porosity, and adhesion of the steel structural components.

[0098] Specifically, step A103 first analyzes the surface image of the corrosion-sensitive area, identifying various types of coating defects such as scratches, cracks, and bubbles, and quantifies the total area of ​​each defect type. Then, based on a preset weighting scheme, for example, assigning a higher weight to crack defects, a numerical value comprehensively reflecting the degree of coating damage is calculated. Subsequently, this value is input into a pre-trained nonlinear relationship model, which outputs predicted values ​​for coating thickness, coating porosity, and coating adhesion corresponding to the degree of coating damage. Finally, these predicted values ​​are used to determine key parameters characterizing the overall pre-coating quality of the steel structure component. Through the above process, an accurate assessment of the pre-coating quality status of the steel structure component can be achieved.

[0099] Specifically, the multi-factor coupled environmental stress accelerated aging model includes a salt spray corrosion model, an ultraviolet degradation model, and a mechanical aging model. The salt spray corrosion model is based on electrochemical principles and considers the effects of chloride ion concentration, temperature, and humidity on the corrosion rate. The ultraviolet degradation model is based on photochemical reaction kinetics and considers the effects of ultraviolet radiation intensity, wavelength, and light absorption characteristics of the coating material on the degradation rate. The mechanical aging model is based on creep theory and considers the effects of stress level, temperature, and time on adhesion decay.

[0100] Step A2 includes:

[0101] A201. Based on the multi-dimensional quality inspection data of the pre-coating, initialize the parameters of the multi-factor coupled environmental stress accelerated aging model; the initialized parameters include coating thickness, porosity, adhesion, chloride ion diffusion coefficient, ultraviolet absorption coefficient, and creep coefficient;

[0102] A202. Based on the environmental stress parameters of the coastal photovoltaic equipment location, determine the environmental stress input for the multi-factor coupled environmental stress accelerated aging model; the environmental stress input includes the time-varying curves of salt spray concentration, ultraviolet radiation intensity, and temperature.

[0103] A203. Iteratively solve the multi-factor coupled environmental stress accelerated aging model with a preset time step to obtain the quality status of the pre-coated layer at different service time points; the quality status includes corrosion depth, UV degradation degree and adhesion.

[0104] The multi-factor coupled environmental stress accelerated aging model consists of three sub-models, simulating the aging behavior of pre-coated materials under the coupled effects of various environmental stresses. The salt spray corrosion model, as one sub-model, can be constructed using electrochemical corrosion kinetics equations. The corrosion rate is correlated with parameters such as chloride ion concentration, temperature, and humidity. The chloride ion concentration can be obtained from the salt spray concentration, and temperature and humidity can be directly derived from environmental stress parameters (i.e., environmental stress parameters can also include humidity). The ultraviolet degradation model, as another sub-model, can be constructed based on photochemical reaction kinetics. The ultraviolet degradation rate can be correlated with ultraviolet radiation intensity, wavelength, and the light absorption coefficient of the coating material. Ultraviolet radiation intensity and wavelength can be obtained from environmental stress parameters (i.e., environmental stress parameters can also include ultraviolet wavelength), while the light absorption coefficient of the coating material is determined by the properties of the pre-coated material itself. The mechanical aging model, as the third sub-model, can be constructed based on creep theory. The coating adhesion decay rate can be correlated with stress level, temperature, and time. The stress level can be obtained from the stress analysis of the steel structure, and temperature can also be derived from environmental stress parameters.

[0105] In step A201, the model parameter initialization operation is fundamental to prediction accuracy. Multi-dimensional quality inspection data of the pre-coating, such as coating thickness, porosity, and adhesion, can be directly used as initial parameter settings for the model. Material parameters of the pre-coating, such as chloride ion diffusion coefficient, UV absorption coefficient, and creep coefficient, can be obtained through experimental testing or by consulting material handbooks.

[0106] In step A202, environmental stress parameters at the location of coastal photovoltaic equipment, such as salt spray concentration, ultraviolet radiation intensity, and temperature, need to be compiled into time-varying curves so that the model can simulate the dynamic changes in environmental stress during actual service. These environmental stress parameter curves can be obtained through methods such as historical environmental monitoring data analysis, extrapolation of short-term environmental monitoring data, or assumptions of typical environmental scenarios.

[0107] In step A203, the iterative model solution process enables dynamic prediction of the pre-coating quality status. The preset time step can be selected based on the requirements of prediction accuracy and computational efficiency; for example, it can be 1 day, 1 week, or 1 month. Within each time step, the three sub-models are solved separately to obtain the increments in salt spray corrosion depth, UV degradation, and adhesion attenuation. These increments are then accumulated and added to the current quality status of the pre-coating to obtain the predicted quality status value for the next time step. Through continuous iteration, the quality status change curve of the pre-coating throughout its service life can be obtained.

[0108] Specifically, the multi-factor coupled environmental stress accelerated aging model, by integrating a salt spray corrosion model, an ultraviolet degradation model, and a mechanical aging model, comprehensively simulates the aging process of pre-coated steel components of coastal photovoltaic equipment under the coupled effects of various environmental stresses. The salt spray corrosion model considers the influence of salt spray, temperature, and humidity on the corrosion rate; the ultraviolet degradation model considers the influence of ultraviolet radiation intensity, wavelength, and the light absorption characteristics of the coating material on the degradation rate; and the mechanical aging model considers the influence of stress, temperature, and time on adhesion decay. Therefore, this model can more accurately reflect the aging behavior of pre-coated components under actual service conditions.

[0109] Step A201 initializes the model parameters using multi-dimensional quality inspection data of the pre-coating, enabling the model to predict based on the measured initial quality state of the pre-coating. This eliminates prediction errors caused by uncertainties in the initial state, improving the personalization and accuracy of the prediction. Step A202 determines the environmental stress input of the model based on the environmental stress parameters of the coastal photovoltaic equipment location, allowing the model to simulate the aging process under actual service conditions. This ensures the consistency between the prediction results and the actual situation, improving the reliability of the prediction. Step A203 iteratively solves the model through a preset time step, achieving dynamic prediction of the pre-coating quality state over time. This not only predicts the quality state of the pre-coating at a specific time point but also obtains the trend of the pre-coating quality state over time, providing more comprehensive information for corrosion risk assessment and operation and maintenance decisions. Therefore, this application ensures the accuracy and reliability of corrosion risk prediction by clarifying the composition and usage method of the multi-factor coupled environmental stress accelerated aging model.

[0110] In some specific implementations, the salt spray corrosion model can be constructed using a combination of Faraday's law and Wagner's corrosion theory. Chloride ion concentration can be calculated from the salt spray deposition rate and solution concentration in the salt spray test standard. Temperature and humidity can be directly obtained from environmental monitoring data. The ultraviolet degradation model can be described using the Langmuir-Hinshelwood kinetic model; ultraviolet radiation intensity and wavelength can be calculated using a solar spectral model and geographical location parameters; and the light absorption coefficient of the coating material can be obtained through ultraviolet-visible spectrophotometry. The mechanical aging model can be described using the Findley creep model; stress levels can be obtained through finite element analysis or theoretical calculations; temperature can be obtained from environmental monitoring data; and time represents the service life. In step A201, the initial coating thickness, porosity, and adhesion can also be obtained through on-site testing using non-destructive testing equipment such as ultrasonic thickness gauges, porosity testers, and pull-out adhesion testers. Material parameters such as chloride ion diffusion coefficient, ultraviolet absorption coefficient, and creep coefficient can also be obtained through indoor accelerated aging tests and material performance tests. In step A202, the environmental stress parameter curve can also be predicted by analyzing historical meteorological data from coastal meteorological stations and combining it with short-term meteorological monitoring data from the photovoltaic power station site. In step A203, the time step can be set to one month, and the model iterative solution can be implemented using numerical calculation software such as Matlab or Python. Through the above specific implementation methods, accurate prediction of the pre-coating quality status of steel structural components of coastal photovoltaic equipment can be achieved.

[0111] In some embodiments, in step A203, within each time step, the corrosion depth increment of the salt spray corrosion model is first calculated based on the salt spray concentration change curve over time, temperature, and humidity. Then, the UV degradation degree increment of the UV degradation model is calculated based on the UV radiation intensity change curve over time, wavelength, and light absorption characteristics of the coating material. Finally, the adhesion decay of the mechanical aging model is calculated based on the stress level, temperature change curve over time, and time. These increments are then added to the current quality state of the pre-coating.

[0112] Within a given time step, the salt spray corrosion model, UV degradation model, and mechanical aging model are executed sequentially to achieve coupled calculations of the increment in quality state. Specifically, firstly, the salt spray corrosion model is used to calculate the increment in corrosion depth. The input parameters used in the calculation include the salt spray concentration, temperature, and humidity varying over time. These parameters work together in the salt spray corrosion model, which is based on electrochemical principles and considers the effects of chloride ion concentration, temperature, and humidity on the corrosion rate, thus obtaining the increment in corrosion depth within the current time step. Secondly, the UV degradation model is used to calculate the increment in UV degradation degree. The input parameters used in the calculation include the UV radiation intensity, wavelength, and light absorption characteristics of the coating material varying over time. The UV degradation model is based on photochemical reaction kinetics, considering the effects of UV radiation intensity, wavelength, and light absorption characteristics of the coating material on the degradation rate, thereby calculating the increment in UV degradation degree within the current time step. Finally, the mechanical aging model is used to calculate the amount of adhesion decay. The input parameters used in the calculation include stress level, temperature, and time varying over time. The mechanical aging model, based on creep theory, considers the effects of stress level, temperature, and time on adhesion decay, thus obtaining the amount of adhesion decay within the current time step. Finally, at the end of each time step, the quality status of the pre-coating is updated by summing the increments of salt spray corrosion depth, UV degradation degree, and adhesion decay. The quality status includes corrosion depth, UV degradation degree, and adhesion. Through the above steps, the coupling mode of each sub-model in the multi-factor coupled environmental stress accelerated aging model is clarified during the iterative solution process, ensuring the accuracy and reliability of corrosion risk prediction.

[0113] In some possible implementations, step A3 includes:

[0114] A301. Obtain data on corrosion influencing factors of steel structural components; corrosion influencing factor data includes the quality status of the pre-coating at different service time points and environmental factor data of the coastal photovoltaic equipment location; environmental factor data includes salt spray concentration, temperature and humidity;

[0115] A302. Based on the data of corrosion influencing factors and the corrosion risk level assessment model, calculate the corrosion risk score of steel structural components at different service time points, and map the corrosion risk score to the corresponding corrosion risk level.

[0116] In step A301, the data on corrosion influencing factors are clearly defined, comprising two main components. One component is the quality status of the pre-coating at different service time points. This data originates from the prediction results of step A2 and specifically reflects quality information such as corrosion depth, UV degradation degree, and adhesion of the pre-coating over time. The other component is environmental factor data regarding the location of the coastal photovoltaic equipment. This data specifically includes salt spray concentration, temperature, and humidity; these environmental parameters are key external conditions affecting the corrosion behavior of steel structural components.

[0117] In step A302, the obtained data on corrosion influencing factors will be used as input to the corrosion risk level assessment model. The model will calculate the corrosion risk score of the steel structure at different service time points. The corrosion risk score will then be mapped to the corresponding corrosion risk level, thereby realizing the graded assessment of corrosion risk.

[0118] Specifically, to more comprehensively and accurately assess corrosion risk, step A3 focuses on two factors: the evolution of the pre-coating's own quality and the influence of the external environment. Pre-coating quality status data, such as corrosion depth, reflects how the coating's resistance to corrosion changes over time. UV degradation data reflects the performance degradation of the coating material due to UV exposure. Adhesion data quantifies the bond strength between the coating and the steel structure. These quality status parameters dynamically reflect the health status of the pre-coating at different service points. Environmental factor data, such as salt spray concentration, temperature, and humidity, represent the severity of the corrosive environment in which the steel structure is located. Salt spray concentration is directly related to chloride ion corrosion of steel, while temperature and humidity jointly affect the rate of corrosion reaction. By simultaneously collecting and utilizing pre-coating quality status data and environmental factor data, the corrosion risk assessment model can comprehensively analyze the combined effects of internal and external factors on the corrosion risk of steel structures, thereby achieving corrosion risk assessment results that more closely reflect reality.

[0119] In some specific implementations, the corrosion risk level assessment model includes a corrosion risk scoring model and an environmental factor risk scoring model. The corrosion risk scoring model is used to calculate the quality status risk score based on the pre-coating quality status parameters, and the environmental factor risk scoring model is used to calculate the environmental factor risk score based on the environmental factor parameters. The pre-coating quality status parameters include corrosion depth, UV degradation degree, and adhesion, and the environmental factor parameters include salt spray concentration, temperature, and humidity.

[0120] Step A302 includes:

[0121] A302.1. Based on the quality status of the pre-coating at different service time points, extract the pre-coating quality status parameters, and based on the environmental factor data of the coastal photovoltaic equipment location, extract the environmental factor parameters;

[0122] A302.2. Based on the corrosion risk scoring model and the environmental factor risk scoring model, calculate the quality status risk score corresponding to the extracted pre-coating quality status parameters and the environmental factor risk score corresponding to the extracted environmental factor parameters, respectively.

[0123] A302.3. The quality status risk score and the environmental factor risk score are integrated and calculated to obtain the comprehensive corrosion risk score;

[0124] A302.4. Determine the corresponding corrosion risk level based on the comprehensive corrosion risk score.

[0125] The corrosion risk assessment model is constructed as a dual-model structure, comprising a corrosion risk scoring model and an environmental factor risk scoring model. The corrosion risk scoring model quantifies the impact of the pre-coating's own quality on corrosion risk. This can be achieved, for example, by pre-setting scoring standards and weights for corrosion depth, UV degradation degree, and adhesion, and obtaining a quality status risk score through weighted summation. The environmental factor risk scoring model quantifies the impact of environmental factors on corrosion risk. This can be achieved, for example, by establishing a mapping relationship between salt spray concentration, temperature, and humidity and the environmental factor risk score. This mapping relationship can be non-linear to more accurately reflect the complex influence of environmental factors. In step A302.1, quality status parameters are extracted from the pre-coating quality status data, and environmental factor parameters are extracted from the environmental factor data. In step A302.2, the quality status risk score and environmental factor risk score are calculated, using the corrosion risk scoring model and the environmental factor risk scoring model respectively. In step A302.3, the quality status risk score and environmental factor risk score are fused together. The fusion calculation can be performed using a weighted average, multiplicative model, or other fusion algorithms to comprehensively consider the combined effects of quality status and environmental factors. In step A302.4, the comprehensive corrosion risk score is mapped to the corresponding corrosion risk level. The risk level can be a pre-defined range, such as low risk, medium risk, high risk, etc., or a level value, such as level 1, level 2, level 3, level 4, level 5, etc. Each corrosion risk level corresponds to a comprehensive corrosion risk score range, and the mapping is performed based on this correspondence.

[0126] Specifically, the corrosion risk assessment model works as follows: First, two independent sub-models—the corrosion risk scoring model and the environmental factor risk scoring model—are used to assess the contributions of pre-coating quality and environmental factors to corrosion risk, respectively. This decoupled assessment method makes the risk assessment process more refined. Then, through fusion calculation, the scores from the two sub-models are combined to obtain a comprehensive corrosion risk score, thus fully considering the combined impact of pre-coating quality and environmental factors. Finally, the comprehensive corrosion risk score is used to determine the final corrosion risk level, achieving the quantification and classification of corrosion risk. This allows for a more accurate assessment of the corrosion risk level of steel structural components, overcoming the problem of traditional methods being unable to effectively quantify and assess corrosion risk levels.

[0127] In some specific implementations, the corrosion risk scoring model is designed as a linear weighted model: Quality status risk score = w1 * corrosion depth score + w2 * UV degradation score + w3 * adhesion score, where w1, w2, and w3 are weighting coefficients. The corrosion depth score, UV degradation score, and adhesion score are obtained by mapping corrosion depth, UV degradation degree, and adhesion through preset scoring criteria. The environmental factor risk scoring model is designed as a nonlinear model, for example, using a neural network model. Inputs include salt spray concentration, temperature, and humidity, and the output is an environmental factor risk score. The parameters of the neural network model are obtained through training with historical corrosion data. The fusion calculation uses a weighted average method: Comprehensive corrosion risk score = α * quality status risk score + (1-α) * environmental factor risk score, where α is a weighting coefficient that can be adjusted according to actual conditions. Through this implementation, a corrosion risk level assessment model that can effectively and accurately evaluate corrosion risk levels can be constructed.

[0128] In some implementations, step A4 includes:

[0129] A401. Based on the corrosion risk level of steel structural components at different service times, construct time-series corrosion risk level curves;

[0130] A402. Based on the preset risk level threshold, determine the time point at which the time series corrosion risk level curve first exceeds the risk level threshold, and use it as the initial threshold exceedance time point;

[0131] A403. Starting from the initial time point exceeding the threshold, backtrack and search for the time point in the time series corrosion risk level curve that is first lower than the risk level threshold, and take it as the end point of the low corrosion risk period.

[0132] A404. Calculate the time difference between the end of the low corrosion risk period and the time when the steel structure is installed, to obtain the low corrosion risk period of the steel structure.

[0133] In step A401, the construction of the time series corrosion risk level curve can be implemented by marking the corrosion risk level data of the steel structure at each service time point on the coordinate system with the service time as the horizontal axis and the corrosion risk level as the vertical axis, and connecting the marked points in chronological order to obtain the time series corrosion risk level curve.

[0134] In step A402, the preset risk level threshold can be set based on engineering experience or relevant standards. For example, the risk level threshold can be set to level 3, representing a medium corrosion risk. Determining the initial threshold exceedance time point can be implemented by starting from the beginning of the time series corrosion risk level curve and searching along the positive direction of the time axis. When the corresponding corrosion risk level on the curve first exceeds the preset risk level threshold, that time point is recorded as the initial threshold exceedance time point.

[0135] In step A403, the backward search for the end point of the low corrosion risk period can be implemented as follows: starting from the initial time point exceeding the threshold, backtracking along the negative direction of the time axis, searching on the time series corrosion risk level curve, and when the corresponding corrosion risk level on the curve is lower than the risk level threshold for the first time, the time point is recorded as the end point of the low corrosion risk period.

[0136] In step A404, calculating the low corrosion risk period can be implemented by subtracting the time value corresponding to the completion time of the steel structure installation from the time value corresponding to the end of the low corrosion risk period, obtaining the time difference, which is the low corrosion risk period of the steel structure. Thus, through the above steps, the low corrosion risk period is quantitatively determined.

[0137] Specifically, this method combines time series analysis and threshold judgment to effectively predict the duration of low corrosion risk. Using a time series corrosion risk level curve clearly demonstrates the trend of corrosion risk evolution over time, laying the foundation for subsequent threshold judgment and duration definition. By setting preset risk level thresholds, acceptable and unacceptable corrosion risk levels can be quantitatively defined, providing a clear standard for predicting the duration of low corrosion risk. By using a backward-looking approach to search for the endpoint of the low corrosion risk duration, the critical point at which corrosion risk transitions from an acceptable to an unacceptable level can be accurately captured, ensuring the accuracy of the low corrosion risk duration prediction.

[0138] Furthermore, the steps following step A401 and before step A402 may also include:

[0139] A401a. For each extreme weather event time point (i.e., the time point of occurrence of each extreme weather event that occurs between the completion of the steel structure installation and the current time), analyze the type and intensity of the extreme weather event, and calculate the degree of impact of the corresponding extreme weather event on the corrosion risk level;

[0140] A401b. Based on the calculated impact of the corrosion risk level, correct the corrosion risk level after the corresponding extreme weather event time point in the time series corrosion risk level curve.

[0141] Regarding the impact of extreme weather events on corrosion risk levels, step A401a is proposed to analyze the type and intensity of each extreme weather event at each time point and calculate the degree of impact of the corresponding extreme weather event on the corrosion risk level. Extreme weather event types can include, but are not limited to, typhoons, rainstorms, high temperatures, and low temperatures. The intensity of extreme weather events can be characterized using quantitative indicators; for example, typhoon intensity can be represented by wind speed, rainstorm intensity by rainfall, and high and low temperature intensity by temperature values. The calculation of the degree of impact can be based on a pre-established mapping model between extreme weather events and corrosion risk levels. This model can be constructed based on historical data statistical analysis, corrosion mechanism research, or expert experience. Step A401b is proposed to correct the corrosion risk level after the corresponding extreme weather event time point in the time-series corrosion risk level curve based on the degree of impact calculated in step A401a. The correction method can be to directly add the degree of impact value to the original corrosion risk level or to adjust it using a weighted average or other methods. Therefore, the time-series corrosion risk level curve can more accurately reflect the sudden impact of extreme weather events on corrosion risk.

[0142] Specifically, after constructing the time-series corrosion risk level curve and before determining the initial threshold time point, an additional step of extreme weather event impact assessment and correction is added. In step A401a, for each extreme weather event recorded in the time series, the specific type of the event is first determined, such as whether it is a typhoon or a rainstorm. Then, the intensity of the event is assessed, such as the wind force of a typhoon or the rainfall amount of a rainstorm. Based on this information, a pre-defined impact assessment model is used to quantify the specific degree of impact of the extreme weather event on the corrosion risk level of the steel structure components. This impact value may be expressed as an increase in the risk level. In step A401b, the calculated degree of impact is used to correct the time-series corrosion risk level curve. Specifically, in the curve, starting from the time point of the extreme weather event, the corrosion risk level at subsequent time points is adjusted upwards, and the adjustment magnitude is the degree of impact value calculated in step A401a. Through this correction, the time-series corrosion risk level curve can reflect the sudden impact of extreme weather, thus providing a more realistic risk assessment basis for predicting subsequent low corrosion risk periods.

[0143] In some specific implementations, it is assumed that time point T in the time-series corrosion risk level curve records a strong typhoon event. In step A401a, analysis shows that the typhoon is a Category 17 strong typhoon. Calculations using the extreme weather event-corrosion risk level mapping model indicate that this strong typhoon event will increase the corrosion risk level by 2 levels. In step A401b, the corrosion risk level values ​​at time point T and all subsequent time points on the time-series corrosion risk level curve are increased by 2 from their original values. For example, if the originally predicted corrosion risk level at time point T+1 was level 3, it will be corrected to level 5. In this way, the accelerating impact of extreme weather events on corrosion risk is quantified and reflected in the corrosion risk level prediction, making the prediction results for the low corrosion risk period more reliable.

[0144] refer to Figure 2 This application also provides a corrosion risk prediction device for steel structural components of coastal photovoltaic equipment, used to predict the corrosion risk of steel structural components of coastal photovoltaic equipment. The device includes:

[0145] Data acquisition module 1 is used to acquire multi-dimensional quality inspection data of the pre-coating after the steel structure components are installed and environmental stress parameters of the coastal photovoltaic equipment location; the multi-dimensional quality inspection data of the pre-coating includes coating thickness, coating porosity and coating adhesion; the environmental stress parameters include salt spray concentration, ultraviolet radiation intensity and temperature change range (refer to step A1 above for details).

[0146] The aging prediction module 2 is used to predict the quality status of the pre-coating at different service time points based on the multi-dimensional quality inspection data and environmental stress parameters of the pre-coating and using a multi-factor coupled environmental stress accelerated aging model. The quality status includes corrosion depth, UV degradation degree and adhesion (refer to step A2 above for details).

[0147] Risk assessment module 3 is used to determine the corrosion risk level of steel structural components at different service times based on the quality status of the pre-coating at different service times using a corrosion risk level assessment model (for details, refer to step A3 above).

[0148] Risk warning module 4 is used to compare the corrosion risk level of steel structural components at different service time points with the preset risk level threshold, and predict the period of low corrosion risk of steel structural components (for details, refer to step A4 above).

[0149] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0150] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0151] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0152] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0153] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for predicting the corrosion risk of steel structural components in coastal photovoltaic equipment, used to predict the corrosion risk of steel structural components in coastal photovoltaic equipment, characterized in that, The method includes the following steps: A1. Obtain multi-dimensional quality inspection data of the pre-coating after the steel structure components are installed and environmental stress parameters of the coastal photovoltaic equipment location; the multi-dimensional quality inspection data of the pre-coating includes coating thickness, coating porosity and coating adhesion; environmental stress parameters include salt spray concentration, ultraviolet radiation intensity and temperature change range; A2. Based on the multi-dimensional quality inspection data and environmental stress parameters of the pre-coating, a multi-factor coupled environmental stress accelerated aging model is used to predict the quality status of the pre-coating at different service time points; the quality status includes corrosion depth, UV degradation degree and adhesion. A3. Based on the quality status of the pre-coating at different service time points, the corrosion risk level of the steel structure at different service time points is determined by the corrosion risk level assessment model; A4. Compare the corrosion risk level of steel structural components at different service time points with the preset risk level threshold to predict the low corrosion risk period of the steel structural components; Step A4 includes: A401. Based on the corrosion risk level of steel structural components at different service times, construct time-series corrosion risk level curves; A402. Based on the preset risk level threshold, determine the time point at which the time series corrosion risk level curve first exceeds the risk level threshold, and use it as the initial threshold exceedance time point; A403. Starting from the initial time point exceeding the threshold, backtrack and search for the time point in the time series corrosion risk level curve that is first lower than the risk level threshold, and take it as the end point of the low corrosion risk period. A404. Calculate the time difference between the end of the low corrosion risk period and the time when the steel structure is installed, to obtain the low corrosion risk period of the steel structure.

2. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 1, characterized in that, Step A1 includes: A101. For the surface of steel structural components, multiple inspection areas are divided. Based on corrosion sensitivity criteria, corrosion-sensitive areas are selected from the multiple inspection areas. Among them, corrosion sensitivity criteria include weld point proximity, edge exposure, and water accumulation risk. A102. Obtain surface images of corrosion-sensitive areas and use image processing algorithms to identify the coating defect type and defect area of ​​each defect feature in the surface image; wherein, the coating defect types include scratches, cracks and bubbles; A103. Based on the type and area of ​​coating defects, calculate the degree of coating damage in corrosion-sensitive areas, and determine the coating thickness, coating porosity, and coating adhesion of steel structural components based on the degree of coating damage in corrosion-sensitive areas. A104. Based on historical environmental monitoring data of the coastal photovoltaic equipment locations, environmental stress parameters are extracted and combined with real-time environmental monitoring data to correct the environmental stress parameters, thus obtaining the corrected environmental stress parameters.

3. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 2, characterized in that, Step A101 includes: Acquire 3D point cloud data of the surface of steel structural components, and divide the point cloud data into multiple detection areas based on a preset segmentation algorithm; For each inspection area, the proximity of the welding point, the edge exposure, and the risk of water accumulation are obtained. The proximity of the welding point is obtained by calculating the reciprocal of the distance from the center point of the inspection area to the nearest welding point. The edge exposure is obtained by calculating the ratio of the boundary length of the inspection area to the area of ​​the area. The risk of water accumulation is obtained by simulating a rainfall process and calculating the water depth in the inspection area. The proximity of weld points, edge exposure, and risk of water accumulation in each detection area are weighted and summed to obtain a comprehensive corrosion sensitivity score. Based on the comprehensive corrosion sensitivity score, the detection areas are sorted from high to low, and a preset number of detection areas are selected as corrosion sensitive areas.

4. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 2, characterized in that, Step A103 includes: Based on the identified coating defect types and defect areas, the total defect area for each coating defect type is calculated. Calculate the weighted sum of the total defect areas for each type of coating defect to obtain the degree of coating damage in corrosion-sensitive areas; Using a nonlinear relationship model between coating damage degree and coating thickness, coating porosity and coating adhesion, the coating thickness, coating porosity and coating adhesion of the corrosion-sensitive area are calculated based on the coating damage degree of the corrosion-sensitive area. The coating thickness, porosity, and adhesion of steel structural components are determined based on the coating thickness, porosity, and adhesion of the corrosion-sensitive areas.

5. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 1, characterized in that, The multi-factor coupled environmental stress accelerated aging model includes a salt spray corrosion model, an ultraviolet degradation model, and a mechanical aging model. The salt spray corrosion model is based on electrochemical principles and considers the effects of chloride ion concentration, temperature, and humidity on the corrosion rate. The ultraviolet degradation model is based on photochemical reaction kinetics and considers the effects of ultraviolet radiation intensity, wavelength, and light absorption characteristics of the coating material on the degradation rate. The mechanical aging model is based on creep theory and considers the effects of stress level, temperature, and time on adhesion decay. Step A2 includes: A201. Based on the multi-dimensional quality inspection data of the pre-coating, initialize the parameters of the multi-factor coupled environmental stress accelerated aging model; the initialized parameters include coating thickness, porosity, adhesion, chloride ion diffusion coefficient, ultraviolet absorption coefficient, and creep coefficient; A202. Based on the environmental stress parameters of the coastal photovoltaic equipment location, determine the environmental stress input for the multi-factor coupled environmental stress accelerated aging model; the environmental stress input includes the time-varying curves of salt spray concentration, ultraviolet radiation intensity, and temperature. A203. Iteratively solve the multi-factor coupled environmental stress accelerated aging model with a preset time step to obtain the quality status of the pre-coated layer at different service time points; the quality status includes corrosion depth, UV degradation degree and adhesion.

6. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 1, characterized in that, Step A3 includes: A301. Obtain data on corrosion influencing factors of steel structural components; the corrosion influencing factor data includes the quality status of the pre-coating at different service time points and environmental factor data of the coastal photovoltaic equipment location; the environmental factor data includes salt spray concentration, temperature and humidity; A302. Based on the data of corrosion influencing factors and the corrosion risk level assessment model, calculate the corrosion risk score of steel structural components at different service time points, and map the corrosion risk score to the corresponding corrosion risk level.

7. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 6, characterized in that, The corrosion risk level assessment model includes a corrosion risk scoring model and an environmental factor risk scoring model. The corrosion risk scoring model is used to calculate the quality status risk score based on the pre-coating quality status parameters, and the environmental factor risk scoring model is used to calculate the environmental factor risk score based on the environmental factor parameters. The pre-coating quality parameters include corrosion depth, UV degradation degree, and adhesion; environmental factors include salt spray concentration, temperature, and humidity. Step A302 includes: Based on the quality status of the pre-coating at different service time points, pre-coating quality status parameters are extracted, and environmental factor parameters are extracted based on environmental factor data of the coastal photovoltaic equipment location. Based on the corrosion risk scoring model and the environmental factor risk scoring model, the quality status risk score corresponding to the extracted pre-coating quality status parameters and the environmental factor risk score corresponding to the extracted environmental factor parameters are calculated respectively. A comprehensive corrosion risk score is obtained by integrating the quality status risk score and the environmental factor risk score. The corresponding corrosion risk level is determined based on the comprehensive corrosion risk score.

8. The corrosion risk prediction method for steel structural components of coastal photovoltaic equipment according to claim 1, characterized in that, The steps following step A401 and before step A402 include: A401a. For each extreme weather event time point, analyze the type and intensity of the extreme weather event, and calculate the degree of impact of the corresponding extreme weather event on the corrosion risk level; A401b. Based on the calculated impact of the corrosion risk level, correct the corrosion risk level after the corresponding extreme weather event time point in the time series corrosion risk level curve.

9. A corrosion risk prediction device for steel structural components of coastal photovoltaic equipment, used to predict the corrosion risk of steel structural components of coastal photovoltaic equipment, characterized in that, The device includes: The data acquisition module is used to acquire multi-dimensional quality inspection data of the pre-coating after the steel structure components are installed and environmental stress parameters of the coastal photovoltaic equipment location. The multi-dimensional quality inspection data of the pre-coating includes coating thickness, coating porosity and coating adhesion. The environmental stress parameters include salt spray concentration, ultraviolet radiation intensity and temperature change range. The aging prediction module is used to predict the quality status of the pre-coating at different service time points based on multi-dimensional quality inspection data and environmental stress parameters of the pre-coating and using a multi-factor coupled environmental stress accelerated aging model. The quality status includes corrosion depth, UV degradation degree and adhesion. The risk assessment module is used to determine the corrosion risk level of steel structural components at different service times based on the quality status of the pre-coating at different service times using a corrosion risk level assessment model. The risk warning module is used to compare the corrosion risk level of steel structural components at different service time points with the preset risk level threshold, and predict the period of low corrosion risk of steel structural components. When the risk warning module compares the corrosion risk level of steel structural components at different service points with preset risk level thresholds to predict the period of low corrosion risk for the steel structural components, it performs the following: A401. Based on the corrosion risk level of steel structural components at different service times, construct time-series corrosion risk level curves; A402. Based on the preset risk level threshold, determine the time point at which the time series corrosion risk level curve first exceeds the risk level threshold, and use it as the initial threshold exceedance time point; A403. Starting from the initial time point exceeding the threshold, backtrack and search for the time point in the time series corrosion risk level curve that is first lower than the risk level threshold, and take it as the end point of the low corrosion risk period. A404. Calculate the time difference between the end of the low corrosion risk period and the time when the steel structure is installed, to obtain the low corrosion risk period of the steel structure.

Images