Method and device for calculating corrosion safety classification of coated metal support structures

Abstract

The present application relates to the technical field of metal structure corrosion monitoring and evaluation, in particular to a kind of coating-containing metal support structure corrosion safety grading calculation method, device, equipment and computer storage medium.The present application obtains multi-modal monitoring data, which includes coating aging parameter, substrate corrosion rate and structure surface image;Based on the surface image, a three-dimensional digital mirror image model is constructed, the deterioration degree value is obtained by identifying and quantifying the coating damage area;Fusion deterioration degree value, aging parameter and corrosion rate, calculate the intrinsic corrosion reliability of metal substrate and determine the grading index by combining load effect;According to the grading index, the safety level is divided, and visual rendering is carried out on the three-dimensional model.The present application realizes the accurate quantitative grading and visualization of corrosion safety risk, solves the "one-size-fits-all" operation and maintenance defects of prior art, and reduces the operation and maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of metal structure corrosion monitoring and assessment technology, and in particular to a method, apparatus, equipment and computer storage medium for calculating corrosion safety classification of coated metal support structures. Background Technology

[0002] Coated metal support structures, as core load-bearing components of critical equipment such as energy facilities and offshore platforms, are widely used in highly corrosive environments with high humidity and high salinity. To ensure their structural safety, existing technologies typically employ single or combined monitoring methods such as electrochemical impedance spectroscopy, optical imaging, and digital image modeling. These methods collect coating aging parameters or substrate corrosion rates through sensors and combine image recognition technology for qualitative or semi-quantitative analysis of coating damage. However, most of these existing corrosion safety assessment methods rely directly on single-dimensional detection data or empirical formulas for judgment, failing to establish an effective mechanism for systematically correlating and deeply integrating macroscopic coating damage (such as area and cracks), microscopic coating aging parameters (such as dielectric constant, moisture content, and porosity), and the actual corrosion rate of the substrate. Due to the lack of joint mining and collaborative reasoning of such multimodal, cross-scale, and heterogeneous data, it is impossible to accurately quantify the remaining reliability and dynamic corrosion safety risks of different regions of the structure, resulting in a "one-size-fits-all" approach to corrosion prevention and maintenance strategies. This technological limitation not only makes it difficult to adapt to the differentiated protection needs in complex service environments, but also easily leads to resource waste due to over-maintenance or safety hazards due to insufficient maintenance, seriously restricting the economic benefits and safe service level of the facility throughout its entire life cycle. Summary of the Invention

[0003] Therefore, the technical problem to be solved by the present invention is to overcome the problem in the prior art that the lack of multimodal in-depth fusion analysis of macroscopic damage to coatings, microscopic aging parameters and substrate corrosion rate leads to the inability to accurately quantify structural corrosion safety risks and the only option to adopt a "one-size-fits-all" extensive operation and maintenance strategy.

[0004] To address the aforementioned technical problems, this invention provides a method for calculating the corrosion safety classification of coated metal support structures, comprising:

[0005] Acquire multimodal monitoring data of coated metal support structures, including coating aging parameters and substrate corrosion rate data collected by sensors, and structural surface image data acquired by image acquisition equipment; A three-dimensional digital mirror model is constructed based on the surface image data of the structure, and the damaged areas of the coating in the three-dimensional digital mirror model are located and quantitatively analyzed using image recognition algorithms to obtain the coating deterioration degree value. By integrating the coating degradation degree value, coating aging parameters and substrate corrosion rate data, the intrinsic corrosion reliability of the metal substrate at different times is calculated, and the corrosion safety classification index is determined in combination with the load effect. Corrosion safety levels are classified according to the corrosion safety grading index, and the evaluation results of different levels are mapped to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering.

[0006] Preferably, the acquisition of multimodal monitoring data of the coated metal support structure includes: Corrosion monitoring sensors are deployed at representative locations of the coated metal support structure. The metal thinning is determined by measuring the change in resistance of the monitored sample, thereby obtaining data on the amount of substrate corrosion and the substrate corrosion rate. The resistance value is linearly related to the change in the thickness of the metal sample. An electrochemical impedance spectroscopy-based coating aging monitoring sensor is deployed at the representative location. By measuring the response of an alternating current signal applied to the coating surface, the impedance spectrum of the coating is obtained. Based on the impedance spectrum, the dielectric constant, water content, porosity, and aging coefficient of the coating are inferred as coating aging parameters. The coated metal support structure is scanned from all directions using an unmanned aerial vehicle equipped with a camera, and continuous frame images are obtained as structural surface image data.

[0007] Preferably, the step of using image recognition algorithms to locate and quantify the damaged areas of the coating in the three-dimensional digital mirror model to obtain the coating degradation degree value includes: A deep learning model is used as the backbone network to identify the damaged areas of the coating in the three-dimensional digital mirror model, and the spatial coordinates and area of ​​the damaged areas are extracted. The image of the damaged area is converted into a color space format and the color component values ​​are extracted as color indicators. At the same time, a semantic segmentation algorithm is used to perform pixel-level detection of cracks in the damaged area to calculate the crack area. A multimodal data fusion classification model is constructed, which takes the area of ​​the region, color index, crack area and coating aging parameters as input features and outputs the coating deterioration degree value.

[0008] Preferably, the step of integrating the coating degradation degree value, coating aging parameters, and substrate corrosion rate data to calculate the intrinsic corrosion reliability of the metal substrate at different times, and determining the corrosion safety classification index in conjunction with the load effect, includes: The coating protection level is determined based on the coating degradation level, and the corrosion rate of the supporting structure over time is calculated by combining environmental corrosion parameters. Based on the factory reliability value of the metal substrate and the corrosion rate of the support structure, the intrinsic corrosion reliability of the metal substrate at different times is calculated. Corrosion safety classification index is calculated based on the intrinsic corrosion reliability of the metal substrate and the statistical characteristics of the load effect.

[0009] Preferably, the step of determining the coating protection level based on the coating degradation degree value and calculating the corrosion rate of the support structure over time in conjunction with environmental corrosion parameters includes: Using a time-series prediction model, online monitoring data and environmental parameters are used as input variables to predict the environmental corrosion and coating degradation at different times. Based on the predicted environmental corrosivity and coating degradation values, the corrosion rate calculation model is substituted into the model. The corrosion rate calculation model is used to characterize the dynamic relationship between the corrosion rate of the supporting structure and the environmental corrosivity and coating degradation level under the condition that the coating protection level decreases with time and degradation.

[0010] Preferably, the construction of the multimodal data fusion classification model uses the region area, color index, crack area, and coating aging parameters as input features, and outputs coating degradation degree values ​​including: The image feature data consisting of the area, color index and crack area of ​​the region is processed by a first feature extraction network, and the time-series feature data represented by the coating aging parameters is processed by a second feature extraction network. A cross-modal attention fusion mechanism is introduced to perform feature-level fusion of extracted image features and temporal features; The fused feature vector is input into the classification layer, and the coating degradation level value is output according to the preset anti-corrosion standard.

[0011] Preferably, the step of classifying corrosion safety levels based on the corrosion safety grading index and mapping the evaluation results of different levels to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering includes: The calculated corrosion safety classification index is divided into multiple corrosion safety levels based on a preset threshold range, where different corrosion safety levels represent different risk levels. Call the graphics rendering library and set the visualization rendering parameters according to the spatial coordinates and time ranges corresponding to different corrosion safety levels; In browser- or server-based application software, the rendering results are dynamically overlaid onto the surface of a 3D digital mirror model, and the evolution of corrosion safety grading assessment results over time is dynamically visualized by setting a time control.

[0012] The present invention also provides a corrosion safety classification calculation device for coated metal support structures, comprising: The data acquisition module is used to acquire multimodal monitoring data of the coated metal support structure. The multimodal monitoring data includes coating aging parameters and substrate corrosion rate data collected by sensors, as well as structural surface image data acquired by image acquisition equipment. The model construction and image analysis module is used to construct a three-dimensional digital mirror model based on the image data of the structure surface, and to use image recognition algorithms to locate and quantify the coating damage area in the three-dimensional digital mirror model to obtain the coating deterioration degree value. The classification index determination module is used to integrate the coating deterioration degree value, coating aging parameters and substrate corrosion rate data to calculate the intrinsic corrosion reliability of the metal substrate at different times, and determine the corrosion safety classification index in combination with the load effect. The assessment and visualization module is used to classify corrosion safety levels based on the corrosion safety grading index, and to map the assessment results of different levels to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering.

[0013] The present invention also provides a corrosion safety classification calculation device for coated metal support structures, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the above-described method for calculating the corrosion safety classification of coated metal support structures.

[0014] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for calculating the corrosion safety classification of coated metal support structures.

[0015] The technical solution of the present invention has the following advantages compared with the prior art: The corrosion safety classification calculation method for coated metal support structures described in this invention acquires and integrates multimodal monitoring data, including coating aging parameters, substrate corrosion rate data, and structural surface image data. It then combines a three-dimensional digital mirror model with image recognition technology to locate and quantify coating damage, calculates the intrinsic corrosion reliability of the metal substrate, and determines corrosion safety classification indicators. Finally, the evaluation results are mapped to a three-dimensional model for visualization rendering. This achieves accurate quantitative classification and spatiotemporal dynamic visualization of corrosion safety risks in coated metal support structures, fundamentally overcoming the shortcomings of existing "one-size-fits-all" extensive operation and maintenance methods. It supports the formulation of differentiated and efficient anti-corrosion solutions, significantly reduces operation and maintenance costs, and improves the safe service level of the structure throughout its entire life cycle. Attached Figure Description

[0016] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein: Figure 1 This is a flowchart illustrating the implementation of a corrosion safety classification calculation method for coated metal support structures provided by the present invention. Figure 2 This is a structural block diagram of a corrosion safety classification calculation device for coated metal support structures provided in an embodiment of the present invention. Detailed Implementation

[0017] The core of this invention is to provide a method, apparatus, equipment, and computer storage medium for calculating the corrosion safety classification of coated metal support structures, which effectively solves the problems in the prior art where the corrosion safety risk is difficult to quantify accurately and the anti-corrosion operation and maintenance strategies are crude due to the inability to conduct multimodal deep fusion assessment.

[0018] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please refer to Figure 1. Figure 1 The flowchart illustrates the implementation of a corrosion safety classification calculation method for coated metal support structures provided by this invention; the specific operation steps are as follows: S101: Acquire multimodal monitoring data of the coated metal support structure, the multimodal monitoring data including coating aging parameters and substrate corrosion rate data collected by sensors, and structural surface image data acquired by image acquisition equipment; S102: Construct a three-dimensional digital mirror model based on the surface image data of the structure, and use an image recognition algorithm to locate and quantify the damaged areas of the coating in the three-dimensional digital mirror model to obtain the coating deterioration degree value. S103: By integrating the coating degradation degree value, coating aging parameters and substrate corrosion rate data, the intrinsic corrosion reliability of the metal substrate at different times is calculated, and the corrosion safety classification index is determined in combination with the load effect. S104: Based on the corrosion safety grading index, the corrosion safety level is divided, and the evaluation results of different levels are mapped to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering.

[0020] Based on the above embodiments, in some embodiments, acquiring multimodal monitoring data of the coated metal support structure includes: Corrosion monitoring sensors are deployed at representative locations of the coated metal support structure. The metal thinning is determined by measuring the change in resistance of the monitored sample, thereby obtaining data on the amount of substrate corrosion and the substrate corrosion rate. The resistance value is linearly related to the change in the thickness of the metal sample. An electrochemical impedance spectroscopy-based coating aging monitoring sensor is deployed at the representative location. By measuring the response of an alternating current signal applied to the coating surface, the impedance spectrum of the coating is obtained. Based on the impedance spectrum, the dielectric constant, water content, porosity, and aging coefficient of the coating are inferred as coating aging parameters. The coated metal support structure is scanned from all directions using an unmanned aerial vehicle equipped with a camera, and continuous frame images are obtained as structural surface image data.

[0021] In one specific embodiment: For the coated carbon steel support structure of a large power generation facility of an energy company in a coastal area, manual inspections were conducted to trace all corrosion points on the structure. After statistical analysis, representative locations at different orientations and floor heights were selected as monitoring points. Various sensors were deployed at these representative locations, including corrosion monitoring sensors using carbon steel as the monitoring sample, coating aging monitoring sensors based on the principle of electrochemical impedance spectroscopy, and auxiliary environmental parameter acquisition devices such as temperature and humidity meters, salt spray sensors, nitrogen oxide and sulfide pollution emission monitors, and ultraviolet radiometers.

[0022] Specifically, the corrosion monitoring sensor determines the metal thinning situation by measuring the change in resistance of the carbon steel sample in real time. According to the law of resistance, the relationship between resistance R and the geometric dimensions and material properties of the metal sample satisfies the formula... Where R is resistance, ρ is resistivity, L is length, w is width, and d is thickness varying with corrosion. The system monitors the real-time change in probe resistance R through high-precision circuitry, calculates the change in metal sample thickness d, converts it into substrate corrosion, and calculates the substrate corrosion rate using time-series data. The coating aging monitoring sensor operates based on electrochemical impedance spectroscopy. The system sprays an organic coating onto the probe surface, applies small-amplitude AC signals of different frequencies for excitation, measures the current signal passing through the coating, and calculates the impedance and impedance phase angle as a function of AC signal frequency, thereby obtaining a wideband coating impedance spectrum. Based on this impedance spectrum, the dielectric constant, water content, porosity, and aging coefficient of the coating are further inferred. The unmanned high-definition camera continuously captures images of the structure from multiple angles along a preset flight trajectory, acquiring a continuous sequence of high-definition images.

[0023] It should be noted that this embodiment integrates multiple sensing mechanisms and imaging technologies to achieve multi-dimensional perception of the corrosion status of coated metal support structures, effectively overcoming the one-sidedness and lag of traditional single monitoring methods, and providing rich and complementary data support for the subsequent construction of a high-precision corrosion safety classification model.

[0024] Based on the above embodiments, in some embodiments, the step of using image recognition algorithms to locate and quantify the damaged areas of the coating in the three-dimensional digital mirror model to obtain coating degradation values ​​includes: A deep learning model is used as the backbone network to identify the damaged areas of the coating in the three-dimensional digital mirror model, and the spatial coordinates and area of ​​the damaged areas are extracted. The image of the damaged area is converted into a color space format and the color component values ​​are extracted as color indicators. At the same time, a semantic segmentation algorithm is used to perform pixel-level detection of cracks in the damaged area to calculate the crack area. A multimodal data fusion classification model is constructed, which takes the area of ​​the region, color index, crack area and coating aging parameters as input features and outputs the coating deterioration degree value.

[0025] In one specific embodiment: an improved YOLOv8n deep learning model, namely the YOLOv8-RM model, is used as the backbone network. This model introduces the RepNCSPELAN4 module, CSPNet module, ELAN module, and MC Attention module on top of the standard YOLOv8n, significantly enhancing the ability to extract features of minute coating damage against complex backgrounds. The YOLOv8-RM model is used to extract features and detect targets from structural surface image data, outputting the precise spatial coordinate location of the coating damage area in a 3D digital mirror model and the corresponding area quantization result. Furthermore, the RGB image of the identified damaged area is converted to HSV format, and the H channel component value is extracted as a color index characterizing the degree of coating discoloration. For microcracks within the damaged area, a semantic segmentation algorithm based on the U-net lightweight deep learning model is used for pixel-level detection. This model utilizes pre-trained weights from a public crack dataset and is fine-tuned on a locally collected support structure crack dataset, enabling accurate segmentation of crack regions and calculation of crack area.

[0026] Specifically, a multimodal data fusion classification model is constructed, employing either an encoder-decoder network, a transformer, or convolutional neural networks as the basic backbone models to process the aforementioned image feature data (region area, color, crack area) and temporal impedance data (coating impedance monitoring data), respectively. The model introduces a cross-modal attention fusion mechanism to perform feature-level conditional probability weighted fusion of the extracted multi-source heterogeneous features, constructing a feature fusion-based classification model. The fused feature vector is input into the classification layer, and compared with reference standards such as anti-corrosion manuals, outputting a coating degradation degree value C. The smaller the value C, the lower the coating degradation degree; conversely, the larger the value C, the more severe the degradation.

[0027] It should be noted that this embodiment, by constructing a high-precision three-dimensional digital mirror model and combining multimodal intelligent image recognition and data fusion analysis, realizes the transformation from qualitative observation to quantitative calculation of coating damage, providing objective and reliable data support for subsequent corrosion safety classification.

[0028] Based on the above embodiments, in some embodiments, the step of integrating the coating degradation degree value, coating aging parameters, and substrate corrosion rate data to calculate the intrinsic corrosion reliability of the metal substrate at different times, and determining the corrosion safety classification index in conjunction with the load effect, includes: The coating protection level is determined based on the coating degradation level, and the corrosion rate of the supporting structure over time is calculated by combining environmental corrosion parameters. Based on the factory reliability value of the metal substrate and the corrosion rate of the support structure, the intrinsic corrosion reliability of the metal substrate at different times is calculated. Corrosion safety classification index is calculated based on the intrinsic corrosion reliability of the metal substrate and the statistical characteristics of the load effect.

[0029] In one specific embodiment: First, the coating protection level P is calculated based on the coating degradation value C, with the relationship P = 100 / C. Then, the corrosion rate D(t) of the supporting structure at time t is calculated using the following formula:

[0030] Where P represents the coating protection level, C represents the coating degradation value, E represents the environmental corrosivity (the larger the E, the stronger the environmental corrosivity and the faster the corrosion), t represents time, and D(t) represents the corrosion rate of the supporting structure. This formula comprehensively considers the decay of the coating protection level and the dynamic changes in environmental corrosivity.

[0031] Furthermore, a factory reliability value for the metal substrate is set. (Usually set to 100), set the corrosion reduction factor k per unit corrosion depth considering coating protection (usually set to 10 / mm). Calculate the intrinsic corrosion reliability R(t) of the metal substrate at time t according to the following formula:

[0032] This indicator reflects the level of structural reliability remaining in the metal substrate after deducting corrosion damage.

[0033] Finally, the corrosion safety classification index β(t) is determined by combining the load effect. It is assumed that the load effect S is constant, and that S and R(t) follow a normal distribution, where... and These are the mean and standard deviation of the load effect, respectively. The standard deviation of intrinsic corrosion reliability for the metallic substrate is given. The formula for calculating the corrosion safety classification index β(t) is as follows:

[0034] It should be noted that this embodiment calculates intrinsic corrosion reliability by integrating multimodal monitoring data and determines the grading index by combining load effects. This overcomes the limitations of single-parameter evaluation and enables quantitative and dynamic accurate rating of the corrosion safety status of coated metal support structures.

[0035] Based on the above embodiments, in some embodiments, the step of determining the coating protection level according to the coating deterioration degree value and calculating the corrosion rate of the support structure over time in combination with environmental corrosion parameters includes: Using a time-series prediction model, online monitoring data and environmental parameters are used as input variables to predict the environmental corrosion and coating degradation at different times. Based on the predicted environmental corrosivity and coating degradation values, the corrosion rate calculation model is substituted into the model. The corrosion rate calculation model is used to characterize the dynamic relationship between the corrosion rate of the supporting structure and the environmental corrosivity and coating degradation level under the condition that the coating protection level decreases with time and degradation.

[0036] In one specific embodiment, a corrosion rate and coating impedance prediction model is constructed using a Long Short-Term Memory (LSTM) model combined with an attention mechanism (BiLSTM+Attention). This model takes historical corrosion rate data and coating impedance data from online monitoring, as well as environmental factors such as temperature, humidity, salt spray, nitrogen oxides, sulfides, and ultraviolet radiation, as input variables. After optimizing the model parameters through a training-testing-validation process, the model can accurately predict the environmental corrosivity E and coating degradation value C at different future times. The predicted E and C are then substituted into the aforementioned corrosion rate formula.

[0037] (where P = 100 / C), the corrosion rate D(t) of the support structure as a function of time is calculated.

[0038] It should be noted that this embodiment achieves dynamic prediction of corrosion rate through a time-series prediction model, providing a basis for corrosion safety trend analysis and forward-looking operation and maintenance decisions.

[0039] Based on the above embodiments, in some embodiments, the construction of the multimodal data fusion classification model, using the region area, color index, crack area, and coating aging parameters as input features, outputs a coating degradation degree value including: The image feature data consisting of the area, color index and crack area of ​​the region is processed by a first feature extraction network, and the time-series feature data represented by the coating aging parameters is processed by a second feature extraction network. A cross-modal attention fusion mechanism is introduced to perform feature-level fusion of extracted image features and temporal features; The fused feature vector is input into the classification layer, and the coating degradation level value is output according to the preset anti-corrosion standard.

[0040] In one specific embodiment: Long Short-Term Memory (LSTM) network and two-dimensional convolutional neural network (2DCNN) are used as the basic backbone models to process one-dimensional temporal impedance data and two-dimensional image feature data for feature extraction. Through cross-modal attention fusion operations such as feature-level conditional probability weighted fusion, the degree of coating degradation of the support structure is graded, and the coating degradation degree value C is output.

[0041] It should be noted that this embodiment effectively integrates image spatial features and temporal electrochemical features through a cross-modal attention fusion mechanism, thereby improving the accuracy and robustness of coating degradation determination.

[0042] Based on the above embodiments, in some embodiments, the step of classifying corrosion safety levels according to the corrosion safety grading index and mapping the evaluation results of different levels to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering includes: The calculated corrosion safety classification index is divided into multiple corrosion safety levels based on a preset threshold range, where different corrosion safety levels represent different risk levels. Call the graphics rendering library and set the visualization rendering parameters according to the spatial coordinates and time ranges corresponding to different corrosion safety levels; In browser- or server-based application software, the rendering results are dynamically overlaid onto the surface of a 3D digital mirror model, and the evolution of corrosion safety grading assessment results over time is dynamically visualized by setting a time control.

[0043] In one specific embodiment: based on the calculated β(t) value and referring to the corrosion prevention manual standard, corrosion safety is divided into four levels, denoted as Level I, II, III, and IV. The higher the level value, the greater the corrosion safety risk, requiring priority for operational and maintenance measures. The Cesium function library is used to set heatmap rendering parameters according to the spatial coordinates and time range corresponding to different corrosion safety levels. The calculated corrosion safety levels I to IV are mapped to different color gradients and dynamically superimposed onto the surface of the 3D digital mirror model in the form of a heatmap. A B / S architecture-based application software is developed, allowing users to stretch, scale, rotate, and translate the 3D model through the software interface. Users can also view the dynamic evolution of the support structure corrosion safety classification assessment results over time by setting a time progress bar.

[0044] It should be noted that this embodiment transforms abstract corrosion safety classification indicators into intuitive spatial visualization information, realizing the spatiotemporal dynamic visualization of corrosion risk. This makes the differences in corrosion safety risks in different areas immediately apparent, providing a direct and reliable decision-making basis for formulating differentiated and efficient anti-corrosion operation and maintenance solutions.

[0045] Please refer to Figure 2 , Figure 2 A structural block diagram of a corrosion safety classification calculation device for coated metal support structures provided in this embodiment of the invention; the specific device may include: The data acquisition module 100 is used to acquire multimodal monitoring data of the coated metal support structure. The multimodal monitoring data includes coating aging parameters and substrate corrosion rate data collected by sensors, as well as structural surface image data acquired by image acquisition equipment. The model construction and image analysis module 200 is used to construct a three-dimensional digital mirror model based on the image data of the structure surface, and to use image recognition algorithms to locate and quantify the coating damage area in the three-dimensional digital mirror model to obtain the coating deterioration degree value. The grading index determination module 300 is used to integrate the coating deterioration degree value, coating aging parameters and substrate corrosion rate data to calculate the intrinsic corrosion reliability of the metal substrate at different times, and determine the corrosion safety grading index in combination with the load effect. The assessment and visualization module 400 is used to classify corrosion safety levels according to the corrosion safety grading index, and to map the assessment results of different levels to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering.

[0046] In one specific embodiment, the data acquisition module 100 is specifically used to: deploy corrosion monitoring sensors with carbon steel as the monitoring sample at representative locations of the coated metal support structure; determine the metal thinning situation by measuring the change in resistance value of the carbon steel sample in real time; and determine the metal thinning situation according to the formula. A correlation model between resistance and thickness was established to obtain data on substrate corrosion amount and corrosion rate. A coating aging monitoring sensor based on the principle of electrochemical impedance was deployed at the representative location. Small-amplitude AC signals of different frequencies were applied to the probe surface coated with organic coating to obtain a wide-band coating impedance spectrum and infer the dielectric constant, water content, porosity and aging coefficient of the coating. A high-definition camera mounted on an unmanned vehicle was used to perform an all-round scan of the coated metal support structure, and oblique photography was used to obtain continuous high-definition images as structural surface image data.

[0047] In one specific embodiment, the model construction and image analysis module 200 is specifically used to: process the continuous frame high-definition image sequence into an osgb data file, and construct a full-size three-dimensional digital mirror model through 3D TILES format conversion; identify the damaged areas of the coating using an improved YOLOv8n deep learning model (YOLOv8-RM), and extract the spatial coordinates and area of ​​the damaged areas; convert the RGB image of the damaged area into an HSV format image and extract the H channel component value as a color index, while using a U-net-based semantic segmentation algorithm to perform pixel-level detection of cracks in the damaged areas to calculate the crack area; construct a multimodal data fusion classification model that introduces a cross-modal attention fusion mechanism, and output the coating degradation degree value.

[0048] In one specific embodiment, the grading index determination module 300 is specifically used to: determine the coating protection level P according to the formula P = 100 / C, and according to the formula... Calculate the corrosion rate D(t) of the supporting structure, where E is the environmental corrosivity and t is time; according to the formula... Calculate the intrinsic corrosion reliability R(t) of the metallic substrate, where Here, k represents the factory reliability value for the metal substrate, and k is the corrosion reduction factor per unit corrosion depth; based on the formula... Calculate the corrosion safety classification index β(t).

[0049] In one specific embodiment, the evaluation and visualization module 400 is specifically used to: classify the calculated corrosion safety classification index β(t) into four corrosion safety levels: I, II, III, and IV, referring to the corrosion prevention manual; call the Cesium function library to set the heat map rendering parameters according to the spatial coordinates and time range corresponding to different corrosion safety levels; and in the application software based on the B / S architecture, dynamically overlay the heat map onto the surface of the three-dimensional digital mirror model, and realize the dynamic visualization display of the corrosion safety classification evaluation results over time by setting a time progress bar.

[0050] The beneficial effects of this invention are: it enables accurate quantitative assessment and three-dimensional dynamic visualization of the corrosion safety level of coated metal support structures, solves the problem of lack of differentiated judgment in traditional operation and maintenance, thereby supporting the formulation of efficient anti-corrosion solutions and reducing operation and maintenance costs.

[0051] The corrosion safety classification calculation device for coated metal support structures in this embodiment is used to implement the aforementioned corrosion safety classification calculation method for coated metal support structures. Therefore, the specific implementation of the corrosion safety classification calculation device for coated metal support structures can be found in the embodiment section of the corrosion safety classification calculation method for coated metal support structures above. For example, the data acquisition module 100, the model construction and image analysis module 200, the classification index determination module 300, and the evaluation and visualization module 400 are respectively used to implement steps S101, S102, S103, and S104 in the above-mentioned corrosion safety classification calculation method for coated metal support structures. Therefore, its specific implementation can be referred to the description of the corresponding embodiments, and will not be repeated here.

[0052] A specific embodiment of the present invention also provides a corrosion safety classification calculation device for coated metal support structures, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-described corrosion safety classification calculation method for coated metal support structures.

[0053] A specific embodiment of the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for calculating the corrosion safety classification of coated metal support structures.

[0054] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0055] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0056] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0057] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0058] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for calculating the corrosion safety classification of coated metal support structures, characterized in that, Includes the following steps: Acquire multimodal monitoring data of coated metal support structures, including coating aging parameters and substrate corrosion rate data collected by sensors, and structural surface image data acquired by image acquisition equipment; A three-dimensional digital mirror model is constructed based on the surface image data of the structure, and the damaged areas of the coating in the three-dimensional digital mirror model are located and quantitatively analyzed using image recognition algorithms to obtain the coating deterioration degree value. By integrating the coating degradation degree value, coating aging parameters and substrate corrosion rate data, the intrinsic corrosion reliability of the metal substrate at different times is calculated, and the corrosion safety classification index is determined in combination with the load effect. Corrosion safety levels are classified according to the corrosion safety grading index, and the evaluation results of different levels are mapped to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering.

2. The method according to claim 1, characterized in that, The acquisition of multimodal monitoring data for coated metal support structures includes: Corrosion monitoring sensors are deployed at representative locations of the coated metal support structure. The metal thinning is determined by measuring the change in resistance of the monitored sample, thereby obtaining data on the amount of substrate corrosion and the substrate corrosion rate. The resistance value is linearly related to the change in the thickness of the metal sample. An electrochemical impedance spectroscopy-based coating aging monitoring sensor is deployed at the representative location. By measuring the response of an alternating current signal applied to the coating surface, the impedance spectrum of the coating is obtained. Based on the impedance spectrum, the dielectric constant, water content, porosity, and aging coefficient of the coating are inferred as coating aging parameters. The coated metal support structure is scanned from all directions using an unmanned aerial vehicle equipped with a camera, and continuous frame images are obtained as structural surface image data.

3. The method according to claim 2, characterized in that, The method of using image recognition algorithms to locate and quantify the damaged areas of the coating in the three-dimensional digital mirror model, and obtaining coating degradation values, includes: A deep learning model is used as the backbone network to identify the damaged areas of the coating in the three-dimensional digital mirror model, and the spatial coordinates and area of ​​the damaged areas are extracted. The image of the damaged area is converted into a color space format and the color component values ​​are extracted as color indicators. At the same time, a semantic segmentation algorithm is used to perform pixel-level detection of cracks in the damaged area to calculate the crack area. A multimodal data fusion classification model is constructed, which takes the area of ​​the region, color index, crack area and coating aging parameters as input features and outputs the coating deterioration degree value.

4. The method according to claim 3, characterized in that, The method of integrating the coating degradation degree value, coating aging parameters, and substrate corrosion rate data to calculate the intrinsic corrosion reliability of the metal substrate at different times, and determining the corrosion safety classification index in conjunction with the load effect, includes: The coating protection level is determined based on the coating degradation level, and the corrosion rate of the supporting structure over time is calculated by combining environmental corrosion parameters. Based on the factory reliability value of the metal substrate and the corrosion rate of the support structure, the intrinsic corrosion reliability of the metal substrate at different times is calculated. Corrosion safety classification index is calculated based on the intrinsic corrosion reliability of the metal substrate and the statistical characteristics of the load effect.

5. The method according to claim 4, characterized in that, The process of determining the coating protection level based on the coating degradation degree value and calculating the corrosion rate of the supporting structure over time by combining environmental corrosion parameters includes: Using a time-series prediction model, online monitoring data and environmental parameters are used as input variables to predict the environmental corrosion and coating degradation at different times. Based on the predicted environmental corrosivity and coating degradation values, the corrosion rate calculation model is substituted into the model. The corrosion rate calculation model is used to characterize the dynamic relationship between the corrosion rate of the supporting structure and the environmental corrosivity and coating degradation level under the condition that the coating protection level decreases with time and degradation.

6. The method according to claim 1, characterized in that, The multimodal data fusion classification model is constructed by using the region area, color index, crack area, and coating aging parameters as input features, and outputting coating degradation degree values ​​including: The image feature data consisting of the area, color index and crack area of ​​the region is processed by a first feature extraction network, and the time-series feature data represented by the coating aging parameters is processed by a second feature extraction network. A cross-modal attention fusion mechanism is introduced to perform feature-level fusion of extracted image features and temporal features; The fused feature vector is input into the classification layer, and the coating degradation level value is output according to the preset anti-corrosion standard.

7. The method according to claim 1, characterized in that, The step of classifying corrosion safety levels based on the corrosion safety grading index and mapping the evaluation results of different levels to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering includes: The calculated corrosion safety classification index is divided into multiple corrosion safety levels based on a preset threshold range, where different corrosion safety levels represent different risk levels. Call the graphics rendering library and set the visualization rendering parameters according to the spatial coordinates and time ranges corresponding to different corrosion safety levels; In browser- or server-based application software, the rendering results are dynamically overlaid onto the surface of a 3D digital mirror model, and the evolution of corrosion safety grading assessment results over time is dynamically visualized by setting a time control.

8. A corrosion safety classification calculation device for coated metal support structures, characterized in that, include: The data acquisition module is used to acquire multimodal monitoring data of the coated metal support structure. The multimodal monitoring data includes coating aging parameters and substrate corrosion rate data collected by sensors, as well as structural surface image data acquired by image acquisition equipment. The model construction and image analysis module is used to construct a three-dimensional digital mirror model based on the image data of the structure surface, and to use image recognition algorithms to locate and quantify the coating damage area in the three-dimensional digital mirror model to obtain the coating deterioration degree value. The classification index determination module is used to integrate the coating deterioration degree value, coating aging parameters and substrate corrosion rate data to calculate the intrinsic corrosion reliability of the metal substrate at different times, and determine the corrosion safety classification index in combination with the load effect. The assessment and visualization module is used to classify corrosion safety levels based on the corrosion safety grading index, and to map the assessment results of different levels to the corresponding spatial coordinates of the three-dimensional digital mirror model for visualization rendering.

9. A corrosion safety classification calculation device for coated metal support structures, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the corrosion safety classification calculation method for coated metal support structures as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the corrosion safety classification calculation method for coated metal support structures as described in any one of claims 1 to 7.

Images