Method and device for dynamically evaluating failure risk of liquid cargo tank of chemical tanker

Abstract

The invention discloses a chemical tanker liquid cargo tank failure risk dynamic assessment method and device, and relates to the technical field of ships, and the method comprises the steps: obtaining a first critical pitting corrosion-free characteristic curve of a liquid cargo tank; collecting real-time characteristic data; identifying a spot corrosion area according to the curve; linearly fitting the spot corrosion area by adopting a spatial adjacency algorithm to obtain a linear corrosion area; diffusing the simulated line corrosion region by adopting a spatial transmissibility algorithm to obtain a surface corrosion region; and performing failure risk weighted evaluation on the surface corrosion area to obtain a failure risk evaluation index. The technical problems that an existing corrosion failure risk assessment method cannot effectively combine real-time environment parameters and material characteristics for dynamic analysis, consequently, the overall failure risk cannot be quantified in real time, and risk identification lags behind are solved. The technical effects that point-to-surface dynamic evolution simulation of the corrosion risk is achieved based on the space adjacency and transmissibility algorithm, and the safety early warning comprehensiveness and accuracy of the liquid cargo tank structure are improved are achieved.

Description

Technical Field

[0001] This application relates to the field of marine technology, specifically to a method and apparatus for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers. Background Technology

[0002] With the increasing global demand for chemical transportation, chemical tankers play a crucial role in the process. As the core component of a ship, the safety of the liquid cargo tanks directly impacts the reliability of transportation and environmental protection. While seawater washing, a common ship cleaning method, effectively removes residues, the presence of chloride ions in seawater easily triggers corrosion of the cargo tank's metallic materials, particularly pitting corrosion. As corrosion accumulates, the cargo tank may experience structural failure, posing safety risks and potentially leading to chemical leaks, causing severe environmental pollution and personal injury. Existing risk assessment technologies largely rely on periodic testing data and empirical thresholds, making it difficult to dynamically capture the spatiotemporal evolution of corrosion initiation and propagation. Especially under the combined effects of chloride ion concentration and temperature, traditional static models cannot correlate material critical corrosion resistance with service environment parameters in real time, resulting in delayed or misjudged risk warnings. Summary of the Invention

[0003] This application provides a method and apparatus for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers, which solves the technical problem that existing corrosion failure risk assessment methods cannot effectively combine real-time environmental parameters and material properties for dynamic analysis, resulting in the inability to quantify the overall failure risk in real time and the lag in risk identification.

[0004] This application provides a method for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers. The method includes: acquiring a first critical non-pitting corrosion characteristic curve of the liquid cargo tank, wherein the first critical non-pitting corrosion characteristic curve characterizes the critical non-pitting corrosion temperature curve of the tank material at different chloride ion concentrations; collecting real-time characteristic data of the liquid cargo tank, including temperature data and chloride ion concentration data of the tank area; identifying pitting corrosion regions using the first critical non-pitting corrosion characteristic curve to obtain multiple pitting corrosion risk regions; linearly fitting the multiple pitting corrosion risk regions using a spatial adjacency algorithm to obtain multiple linear corrosion risk regions; and performing diffusion simulation on the multiple linear corrosion risk regions using a spatial propagation algorithm to obtain multiple surface corrosion risk regions; and performing a weighted assessment of failure risk on the multiple surface corrosion risk regions to obtain a failure risk assessment index.

[0005] This application also provides a device for dynamic assessment of failure risk in chemical tanker cargo tanks, comprising: a critical curve acquisition module for acquiring a first critical non-pitting corrosion characteristic curve of the chemical tanker cargo tank, wherein the first critical non-pitting corrosion characteristic curve characterizes the critical non-pitting corrosion temperature curve of the cargo tank material at different chloride ion concentrations; a real-time feature acquisition module for acquiring real-time feature data of the chemical tanker cargo tank, the real-time feature data including temperature data and chloride ion concentration data of the cargo tank area; a pitting corrosion identification module for identifying pitting corrosion regions based on the first critical non-pitting corrosion characteristic curve of the real-time feature data, thereby obtaining multiple pitting corrosion risk regions; a line-surface analysis module for linearly fitting the multiple pitting corrosion risk regions using a spatial adjacency algorithm, thereby obtaining multiple line corrosion risk regions, and for performing diffusion simulation of the multiple line corrosion risk regions using a spatial propagation algorithm, thereby obtaining multiple surface corrosion risk regions; and a failure assessment module for weighted assessment of failure risk of the multiple surface corrosion risk regions, thereby obtaining failure risk assessment indices.

[0006] The proposed method and apparatus for dynamic assessment of failure risk in chemical tanker cargo tanks, as described in this application, firstly acquires a first critical non-pitting corrosion characteristic curve for the chemical tanker cargo tank, wherein the first critical non-pitting corrosion characteristic curve characterizes the critical non-pitting corrosion temperature curve of the cargo tank material at different chloride ion concentrations; subsequently, real-time characteristic data of the chemical tanker cargo tank is collected, including temperature data and chloride ion concentration data of the cargo tank area; further, the first critical non-pitting corrosion characteristic curve is used to identify pitting corrosion regions in the real-time characteristic data, resulting in multiple pitting corrosion risk regions; then, a spatial adjacency algorithm is used to linearly fit the multiple pitting corrosion risk regions to obtain multiple line corrosion risk regions, and a spatial propagation algorithm is used to perform diffusion simulation on the multiple line corrosion risk regions to obtain multiple surface corrosion risk regions; finally, a failure risk weighted assessment is performed on the multiple surface corrosion risk regions to obtain a failure risk assessment index. This invention addresses the technical problem that existing corrosion failure risk assessment methods cannot effectively combine real-time environmental parameters and material properties for dynamic analysis, resulting in the inability to quantify overall failure risk in real time and the lag in risk identification. It achieves the technical effect of realizing dynamic evolution simulation of corrosion risk from point to surface based on spatial adjacency and propagation algorithms, thereby improving the comprehensiveness and accuracy of structural safety early warning for liquid cargo tanks. Attached Figure Description

[0007] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0008] Figure 1 This is a schematic diagram of the process for dynamic assessment of failure risk of liquid cargo tanks on chemical tankers, provided in an embodiment of this application.

[0009] Figure 2 This is a schematic diagram of the structure of the device for dynamic assessment of failure risk of liquid cargo tanks on chemical tankers provided in the embodiments of this application.

[0010] Figure labeling: 11 Critical curve acquisition module, 12 Real-time feature acquisition module, 13 Pitting corrosion identification module, 14 Line and surface analysis module, 15 Failure assessment module. Detailed Implementation

[0011] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0014] This application provides a method for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers, such as... Figure 1 As shown, the method includes:

[0015] Obtain the first critical pitting-free characteristic curve of the liquid cargo tank of a chemical tanker, wherein the first critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the liquid cargo tank material at different chloride ion concentrations.

[0016] In this embodiment, the materials used in the cargo tanks of chemical tankers (such as the metallic materials of the cargo tanks) are first obtained. Then, a pitting potential change response test is performed on the materials using a pre-configured simulated seawater electrolyte and simulated temperature. The change in the surface potential of the material under different conditions is measured using a three-electrode system (reference electrode, working electrode, and counter electrode). Subsequently, nonlinear fitting analysis is performed based on the measured potential response data to plot the critical pitting-free temperature curves of the cargo tank material at different chloride ion concentrations. This curve describes the critical point of corrosion resistance of the material under different environmental conditions, marking the corrosion resistance temperature limit of the material at a certain chloride ion concentration. After the curve is plotted, the first critical pitting-free characteristic curve of the cargo tank material is obtained. This first critical pitting-free characteristic curve provides basic data for subsequent dynamic evaluation, enabling the assessment of whether pitting corrosion will occur in the material under different environments, thereby preventing the failure risk of the cargo tank.

[0017] Furthermore, this application provides a method for obtaining the first critical pitting-free characteristic curve of a chemical tanker's liquid cargo tank, comprising:

[0018] The material composition of the chemical tanker's liquid cargo tank is obtained, and representative materials of the cargo tank material are obtained based on the material composition. A simulated seawater electrolyte and a simulated temperature are configured. The simulated seawater electrolyte includes multiple chloride ion concentration gradients, and the simulated temperature includes multiple temperature ranges. Multiple sets of test variables are set based on the simulated seawater electrolyte and the simulated temperature. The pitting potential change response test is performed on the representative material based on the multiple sets of test variables, and the pitting potential response data is output. Nonlinear fitting is performed based on the pitting potential response data to obtain the first critical non-pitting characteristic curve.

[0019] Preferably, the material composition of the chemical tanker's cargo tanks is first determined from the design drawings and technical specifications, i.e., the types of materials used. Common materials for cargo tanks include various alloy steels, aluminum alloys, and other metals. The corrosion resistance of these materials directly affects the safety of the ship. Materials used in quantities reaching a preset threshold are then extracted as representative samples to simplify the experiment and facilitate subsequent corrosion assessment. Next, to simulate the corrosive effects of real seawater on the cargo tank materials, a simulated seawater electrolyte is prepared. The chloride ion concentration in the simulated seawater electrolyte is set to multiple gradient values ​​(e.g., low, medium, high). These different chloride ion concentrations simulate the impact of chloride ion variations in seawater on corrosion. Simulated temperatures are also set, with multiple temperature ranges typically covering the temperature range the ship might encounter. Then, based on the configured simulated seawater electrolyte and simulated temperatures, multiple sets of test variables are set. These variables include chloride ion concentration, temperature range, and test time points. These test variables are organized into a three-dimensional test matrix, covering different chloride ion concentrations, temperatures, and time conditions. Based on multiple pre-configured test variables, electrochemical testing techniques are used to test the pitting potential change response. Specifically, the representative material is exposed to the simulated seawater electrolyte corresponding to each test variable, and the potential change of the representative material is measured using a three-electrode system at a set temperature. The change in the material surface potential is recorded for each test, forming pitting potential response data, until obvious pitting corrosion occurs (the current density increases sharply, exceeding a set threshold, such as 10 μA / cm). 2 Then, initialize an exponential model in the form of: Among them, T c It is the critical temperature for no pitting, C ClHere, A, B, and n are the chloride ion concentrations, A, B, and n are the parameters to be fitted, and e is the base of the natural logarithm. By inputting the temperature and concentration from the pitting potential response data into the exponential model, the least squares method is used to minimize the sum of squared errors between the fitted function and the actual data. Parameters are adjusted using gradient descent, Levenberg-Marquardt algorithm, etc., until the error is minimized. After fitting, residual analysis is used to check the difference between the fitted curve and the experimental data. If the residuals are small and randomly distributed, the fitting effect is good. In this case, the first critical pitting-free characteristic curve is plotted based on the fitted exponential model. Conversely, if the residuals are small, the fitting model is adjusted (e.g., power law model, polynomial fitting, etc.) or the initial parameters are reselected, and fitting is performed again until a better fit is obtained. The final generated first critical pitting-free characteristic curve characterizes the critical pitting-free temperature of the material at different chloride ion concentrations, providing basic data for subsequent pitting corrosion area identification and risk assessment, ensuring that ships can better cope with corrosion problems in seawater tank washing environments during transportation.

[0020] Furthermore, this application provides a method for setting multiple sets of test variables based on the simulated seawater electrolyte and simulated temperature, which also includes:

[0021] Multiple test time points are set, where the test time point is the tank washing exposure time; based on the multiple test time points and the multiple test variables, a concentration-temperature-time three-variable test matrix is ​​constructed; the representative material is subjected to pitting potential change response test according to the concentration-temperature-time three-variable test matrix to obtain the second critical pitting-free characteristic curve; the pitting corrosion area is identified by using the second critical pitting-free characteristic curve on the real-time characteristic data.

[0022] Preferably, in practical applications, ship cargo tanks are exposed to seawater for varying periods during the tank washing process. To simulate this process, multiple test time points (i.e., tank washing exposure times) are first set. These time points can be selected based on actual conditions, such as 1 hour, 3 hours, 6 hours, and 12 hours. These time points reflect the corrosion of the material under different exposure times. Subsequently, based on the set test time points and existing test variables, a concentration-temperature-time three-variable test matrix is ​​constructed. Each set of data in this concentration-temperature-time three-variable test matrix corresponds to a specific test condition, covering different chloride ion concentrations, temperatures, and exposure times, to represent the corrosion behavior of the test material under different chloride ion concentrations, temperatures, and exposure times. Subsequently, based on the constructed concentration-temperature-time three-variable test matrix, these combinations of conditions were tested one by one in the experiment. Under each test combination, the pitting potential change response test was performed on the selected representative material. This pitting potential change response test was similar to the previous one, in which the representative material was immersed in simulated seawater electrolyte, the chloride ion concentration and temperature of the solution were adjusted to set values, and the potential change of the material was monitored at a specified exposure time to obtain pitting potential response data. Then, the chloride ion concentration and tank washing exposure time in the pitting potential response data were linearly weighted, and the weighted result was combined with the temperature in the pitting potential response data to perform the same nonlinear fitting as mentioned above to obtain the second critical pitting-free characteristic curve. Finally, based on the second critical pitting-free characteristic curve, potential pitting corrosion areas were identified by comparing and analyzing real-time characteristic data (including temperature and chloride ion concentration data in the liquid cargo tank area) with the curve, providing a scientific basis for the safe operation of ship liquid cargo tanks.

[0023] Furthermore, this application provides the second critical pitting-free characteristic curve characterizing the critical pitting-free temperature curve of the ship liquid cargo tank material under weighted variables; the weighted variables are obtained by normalizing the chloride ion concentration and tank washing exposure time in the same interval, and then linearly weighting the normalized chloride ion concentration and tank washing exposure time.

[0024] Optionally, the second critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the ship's cargo tank material under a weighted variable. This weighted variable is obtained by normalizing the chloride ion concentration and tank cleaning exposure time, followed by linear weighting calculation. Specifically, firstly, the chloride ion concentration and tank cleaning exposure time in the pitting potential response data are normalized using the maximum-minimum normalization method to eliminate differences in the magnitude and unit of different variables, allowing them to be compared and weighted on the same scale. Then, according to preset weighting factors for chloride ion concentration and tank cleaning exposure time, the normalized chloride ion concentration and tank cleaning exposure time are linearly weighted, thus combining the two normalized variables into a single weighted variable to represent the material's corrosion resistance under specific environmental conditions. This weighted variable is then combined with the temperature in the pitting potential response data to plot the second critical pitting-free characteristic curve, identifying areas where pitting corrosion may occur and improving ship safety.

[0025] Real-time characteristic data of the cargo tanker's liquid tanks are collected, including temperature data and chloride ion concentration data of the cargo tank area.

[0026] In one embodiment, temperature sensors and chloride ion concentration sensors based on ion-selective electrodes installed in the cargo tanks of chemical tankers are used to monitor different locations in the cargo tanks in real time, collecting real-time characteristic data of different locations in the cargo tanks, including temperature data and chloride ion concentration data of different areas of the cargo tanks. This real-time characteristic data provides important basic data support for subsequent corrosion risk assessment, pitting corrosion area identification and failure risk analysis, ensuring that the ship's cargo tanks can operate in the safest condition during transportation.

[0027] The first critical non-pitting corrosion characteristic curve is used to identify pitting corrosion regions in the real-time feature data, resulting in multiple pitting corrosion risk regions.

[0028] In one embodiment, after obtaining real-time feature data, the data is compared and analyzed with the first critical pitting-free feature curve to dynamically identify multiple areas at risk of pitting corrosion. These areas are marked and output in graphical or data form for operators to understand the corrosion risk in a timely manner. Furthermore, as the temperature and chloride ion concentration data of the cargo tank are continuously updated, the comparison process with the first critical pitting-free feature curve is repeated to continuously identify new pitting corrosion risk areas, ensuring dynamic monitoring and timely early warning of corrosion risks, thereby improving safety during ship transportation.

[0029] Furthermore, this application provides a method for identifying pitting corrosion regions by using the first critical pitting-free feature curve to identify multiple pitting corrosion risk regions in the real-time feature data, including:

[0030] The chemical tanker's cargo tank is divided into grids, and the real-time temperature and real-time chloride ion concentration corresponding to each grid are output. Using the first critical non-pitting corrosion characteristic curve, the critical temperature threshold corresponding to the real-time chloride ion concentration of each grid is identified. Grids with real-time temperatures greater than the corresponding critical temperature thresholds are marked as pitting corrosion risk areas, and multiple pitting corrosion risk areas are output.

[0031] Preferably, to more accurately identify corrosion risks, the cargo tanker of the chemical tanker is first divided into multiple grids according to the sensor deployment locations and a preset unit grid size. Each grid represents a small area, facilitating independent assessment of the corrosion risk in local areas. Then, based on the grid where the sensor is located, the temperature and chloride ion concentration data from the real-time feature data are assigned to the corresponding grid, forming grid-based real-time feature data. Next, based on the first critical non-pitting corrosion characteristic curve, the critical temperature threshold corresponding to the current chloride ion concentration of each grid is extracted, and the real-time temperature data of each grid is compared with the critical temperature threshold. If the real-time measured temperature is higher than the corresponding critical temperature threshold, it indicates a high corrosion risk for that grid, and this grid is marked as a pitting corrosion risk area. Finally, the same processing is performed on other grids to identify multiple pitting corrosion risk areas. These pitting corrosion risk areas provide the basic data for subsequent linear fitting and diffusion simulation, ensuring the accuracy and real-time nature of the corrosion risk assessment.

[0032] The spatial adjacency algorithm is used to linearly fit the multiple point corrosion risk regions to obtain multiple line corrosion risk regions, and the spatial propagation algorithm is used to diffuse the multiple line corrosion risk regions to obtain multiple surface corrosion risk regions.

[0033] In one embodiment, a spatial adjacency algorithm is first used to analyze pitting corrosion risk areas, calculating the adjacency distances between these areas and identifying adjacent pitting corrosion areas. When the distance between two pitting corrosion risk areas is less than a preset threshold, these two areas are considered adjacent, and the relationship between them is used for linear fitting, resulting in multiple line corrosion risk areas composed of multiple pitting corrosion areas. These line corrosion risk areas reflect the possibility of corrosion spreading along certain specific paths or directions. Subsequently, a spatial propagation algorithm is used to simulate the diffusion of the line corrosion risk areas, forming larger surface corrosion risk areas. The diffusion process is based on factors such as the passage of time, temperature changes, and chloride ion concentration fluctuations, simulating the expansion trend of the corrosion area. Through simulation, multiple surface corrosion risk areas can be obtained, representing the cumulative effect of corrosion and the possible expansion range of the corrosion area. This method enables a more comprehensive assessment of the corrosion risk in ship cargo tanks, helping operators to take timely protective measures and ensure the safe operation of the ship.

[0034] Furthermore, this application provides a method for linearly fitting the multiple point corrosion risk regions using a spatial adjacency algorithm to obtain multiple line corrosion risk regions, the method comprising:

[0035] The adjacency distance of adjacent point corrosion risk areas in the multiple point corrosion risk areas is calculated, and an adjacency distance set is output. The spatial adjacency algorithm is used to mark adjacent point corrosion risk areas with a distance less than a preset adjacency distance as adjacency relationships, and an adjacency point corrosion risk area cluster is obtained. The adjacency point corrosion risk area cluster is fitted with a linear direction to obtain multiple line corrosion risk areas.

[0036] Optionally, after identifying multiple point corrosion risk areas, each pair of point corrosion risk areas is assumed to be a group of adjacent point corrosion risk areas. Using Euclidean distance, the adjacency distance between these assumed adjacent point corrosion risk areas is calculated based on the coordinates of their center points. These adjacency distances are then added sequentially to a set, forming an adjacency distance set. Subsequently, a spatial adjacency algorithm is used to analyze the calculated adjacency distance set. During this process, each adjacency distance in the set is compared with a preset adjacency distance. If an adjacency distance is less than a threshold, it indicates that the assumed adjacent point corrosion risk area corresponding to that distance is correct, and an adjacency relationship is marked for this assumed adjacent point corrosion risk area. Afterward, based on the adjacency relationships, the sets of adjacent point corrosion risk areas are aggregated to form multiple adjacent point corrosion risk area clusters. Each cluster contains a group of adjacent point corrosion risk areas, representing a potential corrosion expansion area. After aggregating pitting corrosion risk areas into multiple clusters, the least squares method is used to perform linear fitting on the center point coordinates of the pitting corrosion risk areas in each cluster to minimize the sum of squared errors between the actual coordinates and the fitted curve. The best fitted line for each cluster is calculated and intercepted according to the starting coordinates to form multiple linear corrosion risk areas. These linear corrosion risk areas can help identify the corrosion expansion trend, assess potential corrosion risk areas, and provide an accurate basis for ship maintenance and protection measures.

[0037] Furthermore, this application provides a method for performing diffusion simulation on the multiple line corrosion risk regions using a spatial propagation algorithm to obtain multiple surface corrosion risk regions, the method comprising:

[0038] A spatial propagation algorithm is used to propagate the multiple linear corrosion risk regions to obtain a set of corrosion-affected regions; multiple candidate corrosion risk regions are obtained by combining the multiple linear corrosion risk regions with the corresponding set of corrosion-affected regions; and multiple surface corrosion risk regions are obtained by identifying the connected regions of the multiple candidate corrosion risk regions.

[0039] Optionally, after identifying multiple linear corrosion risk areas, a spatial propagation algorithm is used to analyze the propagation of these areas. Specifically, multiple linear corrosion risk areas are used as initial seeds to construct a propagation model based on the physical mechanisms of corrosion. Propagation parameters are then defined for the model, including temperature, chloride ion concentration, and exposure time. These factors directly affect the rate of corrosion spread. For example, increased temperature usually accelerates the corrosion reaction, while increased chloride ion concentration enhances the corrosion intensity. Based on these parameters, the model uses mathematical formulas (such as the Arrhenius equation) to describe the rate and extent of corrosion spread. After determining the propagation parameters, the propagation model predicts the spatial spread of corrosion based on the initial conditions and environmental factors of each linear corrosion risk area. As the time step progresses, corrosion spreads from the initial area to the surrounding area. Each simulation step calculates the expansion of the corrosion area and generates a set of corrosion-affected areas. These corrosion-affected areas represent the extent of corrosion after it spreads from the linear corrosion area. The size and shape of each corrosion-affected area depend on the calculation results of the propagation model, showing the spatial impact of corrosion and its potential expansion trend. Subsequently, multiple linear corrosion risk areas are combined with their corresponding corrosion-affected areas to form multiple candidate corrosion risk areas. These candidate corrosion risk areas may become the main areas of influence for future corrosion. Then, to identify the true corrosion risk areas, an 8-neighborhood connected component labeling algorithm is used to identify the connected components of the candidate corrosion risk areas. Specifically, all voxels are traversed in row-major order. If the current voxel is a corrosion risk area and is not labeled, a new connected component labeling is initiated. The 8 neighbors (top, bottom, left, right, and diagonal directions) of the current voxel are then checked. If neighboring voxels belong to the same candidate area and are not labeled, they are merged into the current connected component. If multiple candidate areas form a continuous voxel block through neighborhood search, they are merged into a single surface corrosion risk area. Through the identification of these connected components, multiple surface corrosion risk areas can be obtained. These surface corrosion risk areas represent extensive corrosion areas caused by corrosion diffusion within the ship's cargo tanks. This provides a comprehensive corrosion risk assessment for the ship and a scientific basis for subsequent anti-corrosion measures and maintenance decisions.

[0040] A failure risk weighted assessment was performed on the multiple surface corrosion risk areas to obtain failure risk assessment indicators.

[0041] In one embodiment, for multiple identified surface corrosion risk areas, corrosion risk factors associated with each area are extracted, including corrosion area, corrosion intensity, and corrosion rate. The corrosion area represents the extent of corrosion impact, corrosion intensity reflects the severity of corrosion, and corrosion rate reflects the rate of corrosion acceleration. To comprehensively consider the different impacts of each factor on failure risk, corresponding weighting coefficients are assigned to each factor. These coefficients are based on experimental data, historical data, and expert experience. These weighting coefficients are then applied to the corrosion risk factors and summed to obtain a failure risk assessment index for each surface corrosion risk area. This index quantifies the magnitude of failure risk for each surface corrosion risk area; a higher value indicates a greater failure risk. Finally, surface corrosion risk areas can be ranked according to these indicators to help ship managers identify high-risk areas and prioritize their maintenance and treatment, thereby improving the comprehensiveness and accuracy of early warning systems for cargo tank structures.

[0042] Furthermore, this application provides a failure risk weighted assessment of the multiple surface corrosion risk areas to obtain a failure risk assessment index, the method of which includes:

[0043] The corrosion risk factor vectors of the multiple surface corrosion risk regions are extracted. The corrosion risk factor vectors include corrosion area, corrosion intensity, and corrosion rate. The corrosion area, corrosion intensity, and corrosion rate corresponding to the multiple surface corrosion risk regions are weighted and calculated to obtain the failure risk assessment index.

[0044] Optionally, after identifying multiple surface corrosion risk areas, for each surface corrosion risk area, the number of meshes is counted, and the mesh area is multiplied by the number of meshes to obtain the corrosion area of ​​that surface corrosion risk area. The instantaneous diffusion rate in the propagation model is combined with the time step to calculate the cumulative corrosion depth of the surface corrosion area at each time step, which is taken as the corrosion intensity. The average instantaneous rate of the meshes in the surface corrosion risk area is calculated to obtain the corrosion rate. By concatenating the corrosion area, corrosion intensity, and corrosion rate according to a preset vector template, multiple corrosion risk factor vectors for multiple surface corrosion risk areas are formed. Subsequently, the maximum-minimum normalization method is used to process each corrosion risk factor in the corrosion risk factor vector to ensure that these corrosion risk factors are all under the same dimension. Then, according to the weighting coefficient of each corrosion risk factor, these corrosion risk factors are weighted and summed to obtain the failure risk assessment index for each surface corrosion risk area, which represents the corrosion intensity of each area and its potential impact on the ship structure or equipment, ensuring ship safety.

[0045] Furthermore, this application provides that after obtaining the failure risk assessment index, the method also includes:

[0046] A prediction model is constructed based on the functional relationship between the corrosion risk factor vector and the failure risk assessment index; the prediction model is used to obtain the tank washing safety time corresponding to reaching the preset failure risk threshold, and the tank washing safety time is used to monitor the tank washing exposure time of the chemical tanker's liquid cargo tank.

[0047] Optionally, historical corrosion risk factor vector sequences are collected to construct a training dataset. A prediction model is then built using a Long Short-Term Memory (LSTM) network, including input layers, hidden layers, fully connected layers, and an output layer. The training dataset is then input into the prediction model, and iterative training is performed through forward propagation, loss calculation (mean squared error), backpropagation, and parameter optimization (Adam optimizer) until the maximum number of iterations is reached or the loss converges. The model is then tested using untrained data to calculate the prediction error. If the prediction error is less than or equal to a preset error value, the current prediction model is output; otherwise, hyperparameters such as the learning rate and the number of training batches are adjusted to further improve the model's prediction performance. Subsequently, based on the prediction model, future corrosion risk factor vectors are predicted, and a weighted failure risk assessment index is obtained. This prediction process continues until a corrosion risk factor vector is predicted that first satisfies the prediction failure risk assessment index being greater than or equal to a preset failure risk threshold. Then, the current predicted time step is recorded and used as the tank washing safety time. This tank washing safety time is then used to monitor the tank washing exposure time of the ship's liquid cargo tanks in real time to ensure that it does not exceed the safety limit and avoid the failure of the liquid cargo tanks due to excessive corrosion, thereby improving the safety and reliability of the ship.

[0048] In the above text, refer to Figure 1 A method for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers according to embodiments of the present invention is described in detail. Next, reference will be made to... Figure 2 This invention describes a device for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers according to embodiments of the present invention.

[0049] The dynamic assessment device for failure risk of liquid cargo tanks on chemical tankers according to embodiments of the present invention solves the technical problem that existing corrosion failure risk assessment methods cannot effectively combine real-time environmental parameters and material properties for dynamic analysis, resulting in the inability to quantify the overall failure risk in real time and the lag in risk identification. It achieves the technical effect of realizing the dynamic evolution simulation of corrosion risk from point to surface based on spatial adjacency and propagation algorithms, thereby improving the comprehensiveness and accuracy of structural safety early warning for liquid cargo tanks. The dynamic assessment device for failure risk of liquid cargo tanks on chemical tankers includes: a critical curve acquisition module 11, a real-time feature acquisition module 12, a pitting corrosion identification module 13, a line-surface analysis module 14, and a failure assessment module 15.

[0050] Critical Curve Acquisition Module 11: Acquires the first critical pitting-free characteristic curve of the chemical tanker's cargo tank, wherein the first critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the cargo tank material at different chloride ion concentrations; Real-time Feature Acquisition Module 12: Acquires real-time feature data of the chemical tanker's cargo tank, including temperature data and chloride ion concentration data of the cargo tank area; Pitting Corrosion Identification Module 13: Identifies pitting corrosion regions using the first critical pitting-free characteristic curve on the real-time feature data to obtain multiple pitting corrosion risk regions; Linear and Surface Analysis Module 14: Performs linear fitting on the multiple pitting corrosion risk regions using a spatial adjacency algorithm to obtain multiple linear corrosion risk regions, and performs diffusion simulation on the multiple linear corrosion risk regions using a spatial propagation algorithm to obtain multiple surface corrosion risk regions; Failure Assessment Module 15: Performs a weighted assessment of failure risk on the multiple surface corrosion risk regions to obtain failure risk assessment indicators.

[0051] Furthermore, the critical curve acquisition module 11 also includes:

[0052] The material composition of the chemical tanker's liquid cargo tank is obtained, and representative materials of the cargo tank material are obtained based on the material composition. A simulated seawater electrolyte and a simulated temperature are configured. The simulated seawater electrolyte includes multiple chloride ion concentration gradients, and the simulated temperature includes multiple temperature ranges. Multiple sets of test variables are set based on the simulated seawater electrolyte and the simulated temperature. The pitting potential change response test is performed on the representative material based on the multiple sets of test variables, and the pitting potential response data is output. Nonlinear fitting is performed based on the pitting potential response data to obtain the first critical non-pitting characteristic curve.

[0053] Furthermore, the critical curve acquisition module 11 also includes:

[0054] The material composition of the chemical tanker's liquid cargo tank is obtained, and representative materials of the cargo tank material are obtained based on the material composition. A simulated seawater electrolyte and a simulated temperature are configured. The simulated seawater electrolyte includes multiple chloride ion concentration gradients, and the simulated temperature includes multiple temperature ranges. Multiple sets of test variables are set based on the simulated seawater electrolyte and the simulated temperature. The pitting potential change response test is performed on the representative material based on the multiple sets of test variables, and the pitting potential response data is output. Nonlinear fitting is performed based on the pitting potential response data to obtain the first critical non-pitting characteristic curve.

[0055] Furthermore, the critical curve acquisition module 11 also includes:

[0056] The second critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the cargo tank material under weighted variables; the weighted variables are obtained by normalizing the chloride ion concentration and tank washing exposure time in the same interval, and then linearly weighting the normalized chloride ion concentration and tank washing exposure time.

[0057] Furthermore, the pitting corrosion identification module 13 also includes:

[0058] The chemical tanker's cargo tank is divided into grids, and the real-time temperature and real-time chloride ion concentration corresponding to each grid are output. Using the first critical non-pitting corrosion characteristic curve, the critical temperature threshold corresponding to the real-time chloride ion concentration of each grid is identified. Grids with real-time temperatures greater than the corresponding critical temperature thresholds are marked as pitting corrosion risk areas, and multiple pitting corrosion risk areas are output.

[0059] Furthermore, the line-surface analysis module 14 also includes:

[0060] The adjacency distance of adjacent point corrosion risk areas in the multiple point corrosion risk areas is calculated, and an adjacency distance set is output. The spatial adjacency algorithm is used to mark adjacent point corrosion risk areas with a distance less than a preset adjacency distance as adjacency relationships, and an adjacency point corrosion risk area cluster is obtained. The adjacency point corrosion risk area cluster is fitted with a linear direction to obtain multiple line corrosion risk areas.

[0061] Furthermore, the line-surface analysis module 14 also includes:

[0062] A spatial propagation algorithm is used to propagate the multiple linear corrosion risk regions to obtain a set of corrosion-affected regions; multiple candidate corrosion risk regions are obtained by combining the multiple linear corrosion risk regions with the corresponding set of corrosion-affected regions; and multiple surface corrosion risk regions are obtained by identifying the connected regions of the multiple candidate corrosion risk regions.

[0063] Furthermore, the failure assessment module 15 also includes:

[0064] The corrosion risk factor vectors of the multiple surface corrosion risk regions are extracted. The corrosion risk factor vectors include corrosion area, corrosion intensity, and corrosion rate. The corrosion area, corrosion intensity, and corrosion rate corresponding to the multiple surface corrosion risk regions are weighted and calculated to obtain the failure risk assessment index.

[0065] Furthermore, the failure assessment module 15 also includes:

[0066] A prediction model is constructed based on the functional relationship between the corrosion risk factor vector and the failure risk assessment index; the prediction model is used to obtain the tank washing safety time corresponding to reaching the preset failure risk threshold, and the tank washing safety time is used to monitor the tank washing exposure time of the chemical tanker's liquid cargo tank.

[0067] The device for dynamic assessment of failure risk of liquid cargo tanks for chemical tankers provided in the embodiments of the present invention can execute the method for dynamic assessment of failure risk of liquid cargo tanks for chemical tankers provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0068] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0069] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements 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 dynamic assessment of failure risk in liquid cargo tanks of chemical tankers, characterized in that, The method includes: Obtain the first critical pitting-free characteristic curve of the liquid cargo tank of a chemical tanker, wherein the first critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the liquid cargo tank material at different chloride ion concentrations. Real-time characteristic data of the chemical tanker's liquid cargo tanks are collected, including temperature data and chloride ion concentration data of the cargo tank area; Using the first critical non-pitting corrosion characteristic curve, the real-time feature data is used to identify pitting corrosion regions, resulting in multiple pitting corrosion risk regions. The spatial adjacency algorithm is used to linearly fit the multiple point corrosion risk regions to obtain multiple line corrosion risk regions, and the spatial propagation algorithm is used to diffuse the multiple line corrosion risk regions to obtain multiple surface corrosion risk regions. A failure risk weighted assessment was performed on the multiple surface corrosion risk areas to obtain failure risk assessment indicators.

2. The method for dynamic assessment of failure risk of liquid cargo tanks on chemical tankers as described in claim 1, characterized in that, The method for obtaining the first critical pitting-free characteristic curve of the liquid cargo tank of a chemical tanker includes: Obtain the material composition of the chemical tanker's liquid cargo tank, and based on the material composition, obtain representative materials of the tanker's liquid cargo tank materials; Configure a simulated seawater electrolyte and a simulated temperature. The simulated seawater electrolyte includes multiple chloride ion concentration gradients, and the simulated temperature includes multiple temperature ranges. Based on the simulated seawater electrolyte and simulated temperature, multiple sets of test variables are set, and the representative material is tested for pitting potential change response based on the multiple sets of test variables, and the pitting potential response data is output. Based on the pitting potential response data, a nonlinear fitting was performed to obtain the first critical pitting-free characteristic curve.

3. The method for dynamic assessment of failure risk of liquid cargo tanks on chemical tankers as described in claim 2, characterized in that, Based on the simulated seawater electrolyte and simulated temperature, multiple sets of test variables are set, and the method further includes: Multiple test time points are set, where the test time point is the exposure time during the washing process; Based on the multiple sets of test time points and the multiple sets of test variables, a three-variable test matrix of concentration-temperature-time is constructed. Based on the concentration-temperature-time three-variable test matrix, the representative material was tested for pitting potential change response to obtain the second critical pitting-free characteristic curve. The second critical non-pitting feature curve is used to identify the pitting corrosion region in the real-time feature data.

4. The method for dynamic assessment of failure risk of liquid cargo tanks in chemical tankers as described in claim 3, characterized in that, The second critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the ship's liquid cargo tank material under weighted variables; The weighted variables are obtained by normalizing the chloride ion concentration and the tank washing exposure time within the same interval, and then linearly weighting the normalized chloride ion concentration and the tank washing exposure time.

5. The method for dynamic assessment of failure risk of liquid cargo tanks on chemical tankers as described in claim 1, characterized in that, Using the first critical non-pitting corrosion characteristic curve to identify pitting corrosion regions in the real-time feature data, multiple pitting corrosion risk regions are obtained. The method includes: The chemical tanker's liquid cargo tank is divided into grids, and the real-time temperature and real-time chloride ion concentration corresponding to each grid are output. Using the first critical non-pitting corrosion characteristic curve, identify the critical temperature threshold of the real-time chloride ion concentration corresponding to each grid, mark the grids with real-time temperatures greater than the corresponding critical temperature thresholds as pitting corrosion risk areas, and output multiple pitting corrosion risk areas.

6. The method for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers as described in claim 5, characterized in that, The spatial adjacency algorithm is used to linearly fit the multiple point corrosion risk regions to obtain multiple line corrosion risk regions. The method includes: Calculate the adjacency distance between adjacent pitting corrosion risk areas in the plurality of pitting corrosion risk areas, and output the adjacency distance set; The spatial adjacency algorithm is used to mark the corrosion risk areas of adjacent points that are less than the preset adjacency distance as adjacency relationships, thus obtaining the cluster of corrosion risk areas of adjacent points; Linear direction fitting is performed on the clusters of adjacent corrosion risk areas to obtain multiple linear corrosion risk areas.

7. The method for dynamic assessment of failure risk of liquid cargo tanks in chemical tankers as described in claim 6, characterized in that, A spatial propagation algorithm is used to simulate the diffusion of the multiple linear corrosion risk regions to obtain multiple surface corrosion risk regions. The method includes: A spatial propagation algorithm is used to perform propagation calculations on the multiple linear corrosion risk areas to obtain a set of corrosion-affected areas. Obtain multiple candidate corrosion risk regions composed of the multiple linear corrosion risk regions and the corresponding corrosion-affected regions; Connectivity region identification is performed on the multiple candidate corrosion risk regions to obtain multiple surface corrosion risk regions.

8. The method for dynamic assessment of failure risk of liquid cargo tanks on chemical tankers as described in claim 1, characterized in that, A failure risk weighted assessment is performed on the multiple surface corrosion risk areas to obtain failure risk assessment indices. The method includes: Extract the corrosion risk factor vector of the multiple surface corrosion risk regions, wherein the corrosion risk factor vector includes corrosion area, corrosion intensity and corrosion rate; The corrosion area, corrosion intensity, and corrosion rate corresponding to the multiple surface corrosion risk areas are weighted and calculated to obtain the failure risk assessment index.

9. The method for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers as described in claim 8, characterized in that, After obtaining the failure risk assessment indicators, the method also includes: Construct a prediction model based on the functional relationship between the corrosion risk factor vector and the failure risk assessment index; The prediction model is used to obtain the tank washing safety time corresponding to reaching the preset failure risk threshold, and the tank washing safety time is used to monitor the tank washing exposure time of the chemical tanker's liquid cargo tank.

10. A device for dynamic assessment of failure risk in liquid cargo tanks of chemical tankers, characterized in that, The apparatus is used to implement the dynamic assessment method for failure risk of liquid cargo tanks in chemical tankers as described in any one of claims 1-9, and the apparatus comprises: Critical curve acquisition module: acquires the first critical pitting-free characteristic curve of the chemical tanker's liquid cargo tank, wherein the first critical pitting-free characteristic curve characterizes the critical pitting-free temperature curve of the tanker's liquid cargo tank material at different chloride ion concentrations. Real-time feature acquisition module: Acquires real-time feature data of the cargo tank of the chemical tanker, including temperature data and chloride ion concentration data of the cargo tank area; Pitting corrosion identification module: Identifies pitting corrosion regions in the real-time feature data using the first critical non-pitting corrosion feature curve to obtain multiple pitting corrosion risk regions; Line-surface analysis module: The spatial adjacency algorithm is used to linearly fit the multiple point corrosion risk areas to obtain multiple line corrosion risk areas, and the spatial propagation algorithm is used to diffuse the multiple line corrosion risk areas to obtain multiple surface corrosion risk areas. Failure assessment module: Performs a weighted assessment of failure risk for the multiple surface corrosion risk areas to obtain failure risk assessment indicators.

Images