Storage tank bottom plate corrosion risk prediction method based on electrochemical test

Abstract

The invention discloses a storage tank bottom plate corrosion risk prediction method based on electrochemical testing, and relates to the technical field of oil and gas storage and transportation monitoring, and the method comprises the steps: obtaining an external soil sample and an internal deposited water sample of a to-be-predicted storage tank, and obtaining the original environment parameters of the samples; the method comprises the following steps: respectively constructing a soil simulation solution and a deposition water simulation solution based on an out-tank soil sample and an in-tank deposition water sample, and respectively adding a standard metal sample made of the same material as a bottom plate of a to-be-predicted storage tank into the soil simulation solution and the deposition water simulation solution so as to realize an external corrosion test and an internal corrosion test, obtaining a corresponding corrosion test result; and inputting the original environmental parameters, the corrosion test result and the bottom plate information of the to-be-predicted storage tank into a preset corrosion risk assessment model to perform corrosion risk level prediction on the bottom plate of the to-be-predicted storage tank, and outputting the corrosion leakage risk level of the bottom plate of the to-be-predicted storage tank. And the bottom plate can be effectively inspected and predicted without periodically opening the tank.

Description

Technical Field

[0001] This invention relates to the field of oil and gas storage and transportation monitoring technology, and in particular to a method for predicting corrosion risk of storage tank bottom plates based on electrochemical testing. Background Technology

[0002] The long-term, trouble-free operation of port storage tanks is the cornerstone of ensuring economic and ecological security. As a load-bearing component directly in contact with the foundation soil and the stored medium, the tank bottom plate is subjected to a complex electrochemical, chemical, and microbial corrosion environment over a long period, making it one of the weakest links in structural integrity. Numerous domestic and international cases of storage tank failure demonstrate that bottom plate corrosion and perforation are the primary causes of medium leakage, fires, explosions, and environmental pollution.

[0003] Currently, the industry's assessment and monitoring of the condition of storage tank bottoms heavily relies on periodic full-scale inspections. This traditional approach has inherent limitations, including significant economic costs and losses in production efficiency, as well as severe lag and static nature in risk assessment. Therefore, overcoming these technological shortcomings and achieving effective bottom plate inspections without periodic opening, thereby improving the safety management of storage tanks, is a crucial technical problem that needs to be solved in this field. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method for predicting the corrosion risk of storage tank bottom plates based on electrochemical testing, which can effectively inspect the bottom plates without periodically opening the tank. The specific solution is as follows:

[0005] In a first aspect, this application discloses a method for predicting the corrosion risk of tank bottom plates based on electrochemical testing, including:

[0006] Obtain soil samples from outside the tank and sediment water samples from inside the tank of the tank to be predicted, and obtain the original environmental parameters of the samples;

[0007] Based on the soil sample outside the tank and the sediment water sample inside the tank, a soil simulation solution and a sediment water simulation solution are constructed respectively. A standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the soil simulation solution and the sediment water simulation solution respectively to achieve external corrosion test and internal corrosion test, and obtain the corresponding corrosion test results.

[0008] The original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted are input into the preset corrosion risk assessment model to predict the corrosion risk level of the bottom plate of the tank to be predicted and output the corrosion leakage risk level of the bottom plate of the tank to be predicted.

[0009] The preset corrosion risk assessment model is a base plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system consisting of internal corrosion parameters, external corrosion parameters, and base plate quality.

[0010] Optionally, the step of adding a standard metal sample of the same material as the bottom plate of the tank to be predicted to the soil simulation solution and the sediment water simulation solution respectively, to achieve external corrosion testing and internal corrosion testing, and to obtain the corresponding corrosion test results, includes:

[0011] A standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the soil simulation liquid to perform potentiodynamic polarization curve testing and electrochemical impedance spectroscopy testing to obtain the first electrochemical parameter in order to realize external corrosion testing.

[0012] A standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the sedimentation water simulation solution to perform potentiodynamic polarization curve testing and electrochemical impedance spectroscopy testing to obtain a second electrochemical parameter for internal corrosion testing.

[0013] The internal corrosion rate and external corrosion rate are calculated based on the first electrochemical parameter and the second electrochemical parameter, respectively, and are used as corrosion test results.

[0014] Optionally, before inputting the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted into the preset corrosion risk assessment model, the method further includes:

[0015] A risk indicator system is constructed based on internal corrosion parameters, external corrosion parameters, and base plate quality.

[0016] Create corresponding network nodes for each layer of risk indicators in the aforementioned risk indicator system;

[0017] Define a discrete state for each network node that corresponds to a preset risk state in the risk indicator system;

[0018] Based on the causal relationships in the risk indicator system, directed edges are used to connect network nodes representing lower-level risk indicators to network nodes representing higher-level risk indicators, thereby constructing a directed acyclic graph structure.

[0019] Optionally, the method for predicting corrosion risk of tank bottom plates based on electrochemical testing further includes:

[0020] A prior probability distribution is defined for network nodes in the directed acyclic graph structure corresponding to underlying risk indicators without parent nodes, and a conditional probability table is defined for network nodes in the directed acyclic graph structure corresponding to non-underlying risk indicators with parent nodes, so as to realize the network parameter definition process of the preset corrosion risk assessment model.

[0021] Optionally, the conditional probability table for defining network nodes in the directed acyclic graph structure corresponding to non-bottom-level risk indicators with parent nodes includes:

[0022] Based on the first electrochemical parameter and the second electrochemical parameter, the weight parameters of each corrosion risk factor on the corrosion rate are calculated by the single-factor variable electrochemical test method.

[0023] Based on the weight parameters, a conditional probability table is defined for network nodes in the directed acyclic graph structure corresponding to non-bottom-level risk indicators with parent nodes.

[0024] Optionally, the step of predicting the corrosion risk level of the bottom plate of the tank to be predicted and outputting the corrosion leakage risk level of the bottom plate of the tank to be predicted includes:

[0025] The original environmental parameters, corrosion test results, and bottom plate information of the tank to be predicted are processed in a state-based manner according to the judgment criteria of the corresponding underlying risk indicators in the risk indicator system by the preset corrosion risk assessment model, so as to input the state-based processing results into the corresponding network node in the preset corrosion risk assessment model.

[0026] Probabilistic inference is performed using the preset corrosion risk assessment model to update the posterior probability distribution of all network nodes in the Bayesian network.

[0027] The risk level of bottom plate corrosion leakage of the tank to be predicted is determined based on the posterior probability distribution of the network nodes representing the top-level risk indicators.

[0028] Optionally, after outputting the bottom plate corrosion leakage risk level of the tank to be predicted, the method further includes:

[0029] The corrosion risk assessment model is used to analyze the contribution of the bottom plate of the tank to be predicted to corrosion leakage risk. The contribution of each underlying risk indicator to the corrosion leakage risk level of the bottom plate is calculated and ranked to determine the target corrosion risk factor.

[0030] Secondly, this application discloses a tank bottom plate corrosion risk prediction device based on electrochemical testing, comprising:

[0031] The sample and parameter acquisition module is used to acquire soil samples outside the tank and sediment water samples inside the tank of the tank to be predicted, and to acquire the original environmental parameters of the samples.

[0032] The corrosion testing module is used to construct soil simulation solution and sedimentation water simulation solution based on the soil sample outside the tank and the sedimentation water sample inside the tank, respectively, and to add a standard metal sample of the same material as the bottom plate of the tank to be tested to the soil simulation solution and the sedimentation water simulation solution, respectively, so as to realize external corrosion test and internal corrosion test and obtain the corresponding corrosion test results.

[0033] The risk level prediction module is used to input the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted into a preset corrosion risk assessment model, so as to predict the corrosion risk level of the bottom plate of the tank to be predicted and output the corrosion leakage risk level of the bottom plate of the tank to be predicted.

[0034] The preset corrosion risk assessment model is a base plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system consisting of internal corrosion parameters, external corrosion parameters, and base plate quality.

[0035] Thirdly, this application discloses an electronic device, including:

[0036] Memory, used to store computer programs;

[0037] A processor is used to execute the computer program to implement the steps of the aforementioned disclosed method for predicting corrosion risk of tank bottom plates based on electrochemical testing.

[0038] Fourthly, this application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the aforementioned disclosed method for predicting corrosion risk of tank bottom plates based on electrochemical testing.

[0039] As can be seen, this application discloses the acquisition of external soil samples and internal sediment water samples of the tank to be predicted, and the acquisition of the original environmental parameters of the samples; based on the external soil samples and the internal sediment water samples, soil simulation liquid and sediment water simulation liquid are constructed respectively, and standard metal samples of the same material as the bottom plate of the tank to be predicted are added to the soil simulation liquid and the sediment water simulation liquid respectively to realize external corrosion testing and internal corrosion testing, and obtain the corresponding corrosion test results; the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted are input into a preset corrosion risk assessment model to predict the corrosion risk level of the bottom plate of the tank to be predicted, and output the corrosion leakage risk level of the bottom plate of the tank to be predicted; wherein, the preset corrosion risk assessment model is a bottom plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system composed of internal corrosion parameters, external corrosion parameters, and bottom plate quality. Therefore, by collecting soil and sediment water samples representing the internal and external corrosion environments of the tank without opening it, and conducting direct testing, the production interruption caused by opening the tank is completely avoided, reducing the economic and time costs of assessment. Furthermore, corrosion risk prediction can be performed frequently and on demand, without waiting for periodic testing, obtaining corrosion test results reflecting the current corrosion activity, providing a data foundation for dynamic assessment. Since the network structure of the pre-set corrosion risk assessment model (Bayesian network) is based on a risk indicator system composed of internal corrosion parameters, external corrosion parameters, and base plate quality, using the pre-set corrosion risk assessment model for corrosion risk prediction can solve the problems of incomplete assessment and insufficient systematicity; that is, traditional methods only focus on a single factor. This solution limits the model structure to a system covering three dimensions: internal corrosion, external corrosion, and base plate quality, ensuring that the model considers all known key risk factors at the architectural level, avoiding assessment blind spots, and making the assessment conclusions more comprehensive and reliable. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0041] Figure 1 This application discloses a flowchart of a method for predicting corrosion risk of tank bottom plates based on electrochemical testing.

[0042] Figure 2 This application discloses a specific method for predicting corrosion risk of tank bottom plates based on electrochemical testing.

[0043] Figure 3 This is a schematic diagram of the connection and testing principle of a three-electrode system disclosed in this application;

[0044] Figure 4 This is a partial schematic diagram of a Bayesian network topology constructed in the Genie software environment disclosed in this application;

[0045] Figure 5 This is a schematic diagram of a tank bottom plate corrosion risk prediction device based on electrochemical testing disclosed in this application;

[0046] Figure 6 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0048] The bottom plate of port storage tanks, as a load-bearing component in direct contact with the foundation soil and stored medium, is subjected to a complex electrochemical, chemical, and microbial corrosion environment over a long period of time, making it one of the weakest links in structural integrity. Numerous storage tank failure cases both domestically and internationally demonstrate that bottom plate corrosion and perforation are the primary causes of medium leakage, fires, explosions, and environmental pollution.

[0049] Currently, the industry's assessment and monitoring of the condition of storage tank bottoms relies heavily on periodic full-scale inspections. This traditional approach has inherent limitations, including significant economic costs and losses in production efficiency, as well as a severe lag and static nature in risk assessment.

[0050] Therefore, this invention provides a corrosion risk prediction scheme for tank bottom plates based on electrochemical testing, which can effectively inspect the bottom plates without periodically opening the tank.

[0051] Reference Figure 1 As shown in the figure, this invention discloses a method for predicting corrosion risk of storage tank bottom plates based on electrochemical testing, including:

[0052] Step S11: Obtain soil samples outside the tank and sediment water samples inside the tank to be predicted, and obtain the original environmental parameters of the samples.

[0053] In this embodiment, the on-site sampling plan was executed according to standardization. Based on the design drawings of the tank foundation, the geological survey report, and the surrounding environment (adjacent electrified railways, chemical plant drainage, etc.), sampling points were set up at the edge of the tank foundation and in areas where there might be differences in corrosion (such as water accumulation areas and the interface between backfill and native soil) using a grid method or representative point method. The number of sampling points was usually no less than 5 to ensure the representativeness of the samples. A special soil sampler was used to collect soil samples from outside the tank while preserving the original structure of the soil to the greatest extent possible. The sampling depth should reach at least 0.5 meters below the tank bottom plate. For each sample, its coordinates, sampling depth, soil type, temperature, humidity, and other apparent information were recorded in detail. The soil resistivity and pH value were measured on-site using portable equipment as a preliminary reference to record the above-mentioned external soil environmental parameters. After collection, the samples were immediately sealed in polyethylene bags or glass bottles, stored at low temperature, and sent to the laboratory as soon as possible to prevent changes in the microbial community and chemical properties. At the same time, a systematic collection of sediment water samples from inside the tank was carried out. Specifically, sampling was conducted based on the tank's operating history, tank cleaning and maintenance records, and assessment of the condition of the internal bottom plate. Using a specialized, explosion-proof sampler that does not introduce metal contamination, bottom sediment samples are extracted through the sampling port at the bottom of the tank to avoid vigorous agitation that could suspend sediment and cause secondary oxidation of the samples. For each water sample, tank environmental parameters such as temperature, pH, and conductivity are measured. The samples are then aliquoted into amber glass bottles, tightly sealed, and transported to the laboratory via a cold chain under light-protected conditions to preserve the original chemical state and microbial activity of the water samples as much as possible, ensuring the accuracy of subsequent electrochemical testing results.

[0054] Step S12: Based on the soil sample outside the tank and the sediment water sample inside the tank, construct a soil simulation solution and a sediment water simulation solution respectively. Add a standard metal sample of the same material as the bottom plate of the tank to be predicted to the soil simulation solution and the sediment water simulation solution respectively to achieve external corrosion test and internal corrosion test, and obtain the corresponding corrosion test results.

[0055] In this embodiment, the collected undisturbed soil sample (external soil sample) was added to deionized water according to relevant specifications at a certain water-to-soil ratio and placed in a constant-temperature shaker for 24 hours of extraction by shaking at a specific rate. The supernatant was separated and filtered using a centrifuge to obtain a clear soil leachate, which served as the electrolyte solution simulating the actual service environment of the tank bottom plate. This is the soil simulation solution. The basic corrosion physicochemical parameters of the leachate, such as pH and conductivity, were measured and recorded as the basis for subsequent preparation of the soil simulation solution.

[0056] In this embodiment, in order to accurately quantify the risk of internal corrosion, it is necessary to simulate the sediment water environment inside the tank and conduct systematic electrochemical tests. The specific implementation includes the following two categories.

[0057] 1. Direct testing of sediment water samples: The sediment water samples collected on-site from the bottom of the tank are filtered and clarified, and used directly as electrolyte solutions, i.e. sediment water simulation solutions, to most realistically reflect the corrosivity of the water accumulation area at the bottom of the tank.

[0058] 2. Preparation of sedimentation water simulation solution: To ensure the repeatability and comparability of the experiment, a simulation solution, i.e., sedimentation water simulation solution, is precisely prepared based on the typical corrosive chemical composition of the sedimentation water at the bottom of the tank during operation. The base solution is determined according to the main components of the medium (according to the contents included in the risk indicator system), and key corrosion factors such as ion concentration, sulfur content, and pH value are precisely controlled by adding reagents such as NaCl, Na2S, HCl, and NaOH.

[0059] In this embodiment, a standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the soil simulation solution to perform potentiodynamic polarization curve testing and electrochemical impedance spectroscopy testing to obtain a first electrochemical parameter for external corrosion testing; a standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the sediment water simulation solution to perform potentiodynamic polarization curve testing and electrochemical impedance spectroscopy testing to obtain a second electrochemical parameter for internal corrosion testing; the corresponding internal corrosion rate and external corrosion rate are calculated based on the first electrochemical parameter and the second electrochemical parameter, respectively, as the corrosion test results. Understandably, corrosion tests employ potentiodynamic polarization curve testing and electrochemical impedance spectroscopy. The core electrochemical parameters obtained (first and second electrochemical parameters) include at least corrosion current density, polarization resistance, and self-corrosion potential. External soil corrosion environment testing uses a simulated soil solution as the electrolyte; internal sedimentary water corrosion environment testing uses a simulated solution of sedimentary water samples as the electrolyte. The corrosion rate of standard metal samples is calculated based on the corrosion current density to obtain the corrosion test results. The influence of soil corrosion factors and internal medium corrosion factors on the corrosion rate is normalized into weighting factors using a single-factor variable method, which are then used to construct the conditional probability tables for the corresponding nodes in the Bayesian network model.

[0060] Specifically, a steel plate identical in material to the bottom plate of the tank to be tested is selected and processed into a standard-sized standard metal sample. The working surface of the standard metal sample is polished to a mirror finish using metallographic sandpaper of different grits, then ultrasonically cleaned and degreased with acetone and anhydrous ethanol, and dried with cold air before use. This treatment aims to simulate the relatively stable state of the surface oxide film after a certain period of service on an actual bottom plate. A large-area platinum sheet or graphite rod is used as the auxiliary electrode (CE) to ensure uniform distribution of polarization current. Depending on the solution properties, either a saturated calomel electrode (SCE) or a copper sulfate electrode (CSE) is selected as the reference electrode (RE), and all potentials are relative to this reference electrode. A standard three-electrode electrolytic cell is used, and the test is conducted in a constant-temperature (e.g., 25.0 ± 0.5°C) water bath to eliminate the influence of temperature fluctuations. Then, the electrochemical measurement sequence and parameters are extracted to obtain the electrochemical parameters. Open circuit potential (OCP) monitoring involves immersing the working electrode in the solution and continuously monitoring its open circuit potential over time until it stabilizes (usually 30-60 minutes). The stable OCP value is recorded, reflecting the metal's tendency to self-corrode. After OCP stabilization, a scan is performed at a specific rate (typically 0.166 mV / s or 0.5 mV / s) from -250 mV (relative to OCP) to +250 mV (relative to OCP). The obtained polarization curve is then subjected to Tafel extrapolation fitting using specialized electrochemical data analysis software (such as Gamry Framework) to accurately calculate the corrosion current density. The corrosion rate is the most critical parameter for quantifying corrosion rate. Using Faraday's law, it can be determined that... This is converted to annual corrosion depth (mm / a). At open-circuit potential, a sinusoidal perturbation signal with an amplitude of 10 mV is applied, with a frequency sweep typically ranging from 100 kHz to 10 m / Hz. By analyzing the Nyquist or Bode plot and fitting it with a suitable equivalent circuit model, information such as polarization resistance (Rp) and solution resistance (Rs) can be obtained.

[0061] In this embodiment, before inputting the original environmental parameters, corrosion test results, and bottom plate information of the tank to be predicted into the preset corrosion risk assessment model, the method further includes: constructing a risk indicator system based on internal corrosion parameters, external corrosion parameters, and bottom plate quality. It can be understood that, based on a thorough review and integration of authoritative domestic and international standards such as API 653, API 581, NACE SP0508, and SY / T0088, as well as engineering practice experience, a detailed and logically clear five-level risk indicator system has been constructed. Specific risk indicators are as follows:

[0062] L1: The target layer is the risk of corrosion and leakage of the bottom plate of the port storage tank. It is the final output and decision point of the entire assessment system.

[0063] L2: Criterion layer, comprising three independent yet interconnected core dimensions: internal corrosion, external corrosion, and base plate quality.

[0064] C1: Internal corrosion, a corrosion risk determined by the chemical properties of the storage medium itself, the operating conditions inside the tank, and the effectiveness of internal protective measures. The internal corrosion criterion layer includes at least "internal environment," "coating protection," "cathodic protection," and "static discharge system" as its secondary indicators. The secondary indicators of the internal environment further include "medium corrosivity," "microbial corrosion," "medium transmission and reception frequency," "medium temperature," "tank drainage effect," and "corrosion inhibitors" as tertiary indicators. Among these, the tertiary indicator of medium corrosivity is comprehensively characterized and judged by eleven quaternary indicators: "medium salt content," "medium sulfur content," "medium pH value," "medium nitrogen content," "medium oxygen content," "medium water content," "medium carbon dioxide content," "medium hydrogen sulfide content," "medium iron ion content," "medium sulfate content," and "medium conductivity." The tank drainage effect is characterized and judged by two quaternary indicators: "drainage frequency" and "tank agitator effect." The medium temperature is characterized and judged by two quaternary indicators: "medium operating temperature" and "type of tank heating device."

[0065] C2: External corrosion, a corrosion risk jointly determined by the external soil environment of the tank bottom plate, stray current interference, and the performance of external protective measures (coatings, cathodic protection). The external corrosion criterion layer includes at least "external environment," "coating protection," and "cathodic protection" as its secondary indicators. The secondary indicators of the external environment include "soil corrosivity," "water content outside the bottom plate," and "stray current" as tertiary indicators. The tertiary indicators of soil corrosion are comprehensively characterized and determined by eight quaternary indicators: "soil resistivity," "soil pH," "soil water content," "soil chloride ion content," "soil oxygen content," "soil redox potential," "soil sulfate-reducing bacteria content," and "soil side temperature."

[0066] C3: Base plate quality, determined by the inherent properties of the base plate material, structural design, manufacturing process, and installation quality, refers to its inherent ability to resist corrosion and withstand loads. The base plate quality standard layer includes at least "tank base plate" and "tank base foundation" as its secondary indicators; the secondary indicators of the tank base plate further include "tank material type," "steel plate thickness," "soil side base plate type," and "construction quality" as tertiary indicators; "construction quality" is comprehensively characterized and judged through four quaternary indicators: "layout method," "joint method," "whether there is weld damage and surface leakage traces," and "whether the base plate has obvious deformation."

[0067] L3-L5: The indicator layer refines the criteria layer step by step until it reaches the most basic level of measurable, observable, and describable specific indicators. Each bottom-level indicator is clearly defined with three risk states: "low," "medium," and "high," along with their corresponding quantitative thresholds or qualitative descriptions. This ensures the standardization, normalization, and unambiguity of data collection, laying a solid foundation for subsequent model building.

[0068] Furthermore, corresponding network nodes are created for each layer of risk indicators in the risk indicator system; a discrete state corresponding to a preset risk state in the risk indicator system is defined for each network node; based on the causal relationships in the risk indicator system, network nodes representing lower-level risk indicators are connected level by level to network nodes representing higher-level risk indicators through directed edges to construct a directed acyclic graph structure. It is understood that in Bayesian network visualization software, such as Genie, all network nodes are created sequentially from top to bottom according to the hierarchy of the indicator system; each network node corresponds to one indicator in the indicator system. Network nodes should be clearly named, such as "soil resistivity," "internal coating condition," and "leakage risk." Define a discrete state for each network node, strictly corresponding to the "low," "medium," and "high" risk states in the indicator system. Using the directed arrow tool in Genie software, connect the network nodes from the lowest level indicator to their parent nodes based on the causal relationships between risk indicators in the risk indicator system, until reaching the top-level network node, forming a complete directed acyclic graph model structure. Specifically, drag from child nodes (causes) to parent nodes (results) according to the causal relationships between risk indicators. For example, drag arrows from nodes such as "soil resistivity" and "soil pH" to the "soil corrosivity" node; then drag arrows from nodes such as "soil corrosivity," "external coating condition," and "cathode protection effect" to the "external corrosion" node; finally, the three nodes "internal corrosion," "external corrosion," and "bottom plate quality" all point to the top-level port storage tank bottom plate corrosion leakage risk node. This forms a complete directed acyclic graph model structure.

[0069] Furthermore, network parameters are defined for the graph model structure. Specifically, prior probability distributions are defined for network nodes in the directed acyclic graph structure corresponding to bottom-level risk indicators without parent nodes, and conditional probability tables are defined for network nodes in the directed acyclic graph structure corresponding to non-bottom-level risk indicators with parent nodes, thus realizing the network parameter definition process of the preset corrosion risk assessment model. It can be understood that for network nodes corresponding to the bottom-level risk indicators, their prior probabilities are defined by directly inputting probability values ​​based on historical data statistics or expert surveys into the node probability table in the Genie software; for intermediate network nodes that are not at the bottom level and the target network nodes at the top level, their conditional probability tables are filled manually or through the data import function by editing the probability table interface of the node in the Genie software, combining expert knowledge scoring, historical data fitting, and finite element simulation result analysis. For example, for non-bottom-level nodes that have parent nodes, simple logical relationships can be used to simplify the construction of causal relationships using standard gate models such as Noisy-OR and Noisy-AND. For example, setting the "soil corrosivity" node as a Noisy-OR gate means that as long as one parent node is in a "high" risk state, it is very likely that "soil corrosivity" will also be "high".

[0070] Specifically, based on the first electrochemical parameter and the second electrochemical parameter, the weight parameters of each corrosion risk factor on the corrosion rate are calculated by the single-factor variable electrochemical testing method; and a conditional probability table is defined for the network nodes of the directed acyclic graph structure corresponding to the non-bottom-level risk indicators with parent nodes based on each weight parameter.

[0071] Step S13: Input the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted into the preset corrosion risk assessment model to predict the corrosion risk level of the bottom plate of the tank to be predicted, and output the corrosion leakage risk level of the bottom plate of the tank to be predicted; wherein, the preset corrosion risk assessment model is a bottom plate corrosion probability model based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system composed of internal corrosion parameters, external corrosion parameters, and bottom plate quality.

[0072] In this embodiment, the original environmental parameters, corrosion test results, and bottom plate information of the tank to be predicted are processed statefully according to the judgment criteria of the corresponding bottom-level risk indicators in the risk indicator system using the preset corrosion risk assessment model. The state processing results are then input into the corresponding network nodes in the preset corrosion risk assessment model. Probabilistic inference is performed through the preset corrosion risk assessment model to update the posterior probability distribution of all network nodes in the Bayesian network. Based on the posterior probability distribution of the network nodes representing the top-level risk indicators, the bottom plate corrosion leakage risk level of the tank to be predicted is determined. It is understood that all relevant data of the target tank are collected, including:

[0073] 1. Analysis report of corrosion factors in soil and sediment water sampled on site (such as Cl⁻ content, SO₄²⁻ content, etc.);

[0074] 2. Electrochemical test report (including corrosion current density, corrosion rate, etc.);

[0075] 3. Operation and maintenance records (including media test reports, cathodic protection potential, maintenance reports, etc.);

[0076] 4. Historical tank opening inspection reports (including bottom plate thickness measurement diagram, weld inspection results, coating condition assessment, etc.);

[0077] 5. Design documents (including base plate material, original thickness, etc.).

[0078] Based on the judgment criteria in the indicator system, all data are converted into "low," "medium," and "high" states for corresponding nodes using a preset corrosion risk assessment model. By clicking on the corresponding node, the observed state (i.e., input evidence) can be selected or set in its status window.

[0079] Once all available evidence has been input, the inference algorithm will automatically start using either an exact inference algorithm or an approximate inference algorithm. The algorithm will propagate the probability throughout the network based on the input evidence and update the posterior probability distribution of all unknown nodes.

[0080] The results will be output at the risk node of corrosion and leakage of the bottom plate of the port storage tank.

[0081] In this way, when new detection data is obtained, it can be immediately used as new evidence input, and the model can update the risk assessment results within seconds, achieving near real-time tracking of risks. Different scenarios can be set up for simulation analysis to predict the evolution trend of risks after certain factors worsen.

[0082] In this embodiment, the preset corrosion risk assessment model is used to analyze the contribution of the corrosion leakage risk of the bottom plate of the tank to be predicted. The contribution of each underlying risk indicator to the corrosion leakage risk level of the bottom plate is calculated and ranked to determine the target corrosion risk factor. It is understood that in Genie software, after probabilistic reasoning is completed and a high-risk conclusion is obtained, the sensitivity analysis or evidence influence analysis functions provided by the software are used to calculate and rank the contribution of each underlying risk factor node to the high-risk state of the corrosion leakage risk node of the top-level tank bottom plate. This identifies the set of key corrosion factors leading to high risk, providing a targeted objective for precise prevention and control.

[0083] As can be seen, this application discloses the acquisition of external soil samples and internal sediment water samples of the tank to be predicted, and the acquisition of the original environmental parameters of the samples; based on the external soil samples and the internal sediment water samples, soil simulation liquid and sediment water simulation liquid are constructed respectively, and standard metal samples of the same material as the bottom plate of the tank to be predicted are added to the soil simulation liquid and the sediment water simulation liquid respectively to realize external corrosion testing and internal corrosion testing, and obtain the corresponding corrosion test results; the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted are input into a preset corrosion risk assessment model to predict the corrosion risk level of the bottom plate of the tank to be predicted, and output the corrosion leakage risk level of the bottom plate of the tank to be predicted; wherein, the preset corrosion risk assessment model is a bottom plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system composed of internal corrosion parameters, external corrosion parameters, and bottom plate quality. Therefore, by collecting soil and sediment water samples representing the internal and external corrosion environments of the tank without opening it, and conducting direct testing, the production interruption caused by opening the tank is completely avoided, reducing the economic and time costs of assessment. Furthermore, corrosion risk prediction can be performed frequently and on demand, without waiting for periodic testing, obtaining corrosion test results reflecting the current corrosion activity, providing a data foundation for dynamic assessment. Since the network structure of the pre-set corrosion risk assessment model (Bayesian network) is based on a risk indicator system composed of internal corrosion parameters, external corrosion parameters, and base plate quality, using the pre-set corrosion risk assessment model for corrosion risk prediction can solve the problems of incomplete assessment and insufficient systematicity; that is, traditional methods only focus on a single factor. This solution limits the model structure to a system covering three dimensions: internal corrosion, external corrosion, and base plate quality, ensuring that the model considers all known key risk factors at the architectural level, avoiding assessment blind spots, and making the assessment conclusions more comprehensive and reliable.

[0084] like Figure 2The diagram shows the overall flowchart of the dynamic risk assessment method for corrosion leakage of the bottom plate of large storage tanks in ports. This diagram systematically illustrates the complete assessment process from field data collection, laboratory testing, model building to dynamic inference. Its specific implementation can be divided into the following steps:

[0085] Systematic on-site data acquisition. The specific implementation process includes: conducting detailed on-site sampling planning. Based on the tank foundation design drawings and geological survey report, sampling points are systematically laid out using a grid method. Sampling points need to comprehensively cover the edge area, central area, and potentially corroded areas of the tank foundation to ensure representativeness of the samples. Based on this, the appropriate sampling depth and quantity are determined. Standardized on-site sampling operations are implemented. Uncirculated soil samples are obtained using professional soil sampling equipment, strictly maintaining the integrity of the soil structure. During sampling, the geographical location information and soil surface characteristics of each sampling point are recorded in detail, and rapid on-site testing is conducted simultaneously to obtain preliminary data on key parameters. Subsequently, the samples are properly packaged and preserved.

[0086] A simulated laboratory solution was prepared to complete standardized laboratory electrochemical tests. The specific implementation process included: Preparing a standardized laboratory test solution. Soil samples were collected and prepared into a simulated solution according to a standardized soil-to-water ratio and extraction procedure. A clear electrolyte solution was obtained through standard operations such as isothermal shaking and centrifugation. The sediment water sample collected on-site from the tank bottom was filtered and clarified, and used directly as the electrolyte solution to most realistically reflect the corrosiveness of the water accumulation area at the tank bottom. Simultaneously, a simulated sediment water solution was prepared. To ensure the repeatability and comparability of the experiment, the simulated solution was precisely prepared based on the typical corrosive chemical components of the sediment water at the tank bottom during operation. Environmental conditions and operating parameters were strictly controlled throughout this process. After the solution preparation was completed, basic physicochemical parameters were immediately measured to establish a complete data foundation for subsequent electrochemical tests.

[0087] Construct a complete electrochemical testing system. Establish a standard-compliant three-electrode testing system, with the working electrode material being identical to that of the tank bottom plate. Before testing, perform standardized pretreatment on the electrode working surfaces to accurately simulate the actual service conditions of the bottom plate. Simultaneously, calibrate the testing instruments to ensure measurement accuracy.

[0088] Figure 3 This is a schematic diagram illustrating the connection and testing principle of the three-electrode system of the laboratory electrochemical testing system in this embodiment of the invention. The three-electrode testing system mainly consists of a working electrode (WE), a reference electrode (RE), an auxiliary electrode (GE, also known as a counter electrode), an electrolytic cell containing electrolyte, and the core control and measurement equipment, an electrochemical workstation.

[0089] The connections and functions of each component are as follows: Working Electrode (WE): As the core of the research system, it is the site where the sample or target electrochemical reaction occurs. The working electrode is connected to the "Working Electrode (WE)" interface of the electrochemical workstation via a wire. The electrochemical workstation precisely controls the potential of the working electrode and monitors its current response in real time through this pathway. Reference Electrode (RE): Its function is to provide a known and stable reference potential for accurately determining the relative potential of the working electrode. The reference electrode is connected to the "Reference Electrode (RE)" interface of the electrochemical workstation via a wire. Theoretically, no current flows through the reference electrode during the entire test, thus ensuring its potential remains constant and establishing a reliable potential reference point for the entire test system. Auxiliary Electrode (GE): Its main function is to form a current path corresponding to the working electrode circuit, allowing the current to be closed. The auxiliary electrode is connected to the "Auxiliary Electrode (GE)" interface of the electrochemical workstation via a wire. When the electrochemical workstation applies a potential to the working electrode, electrons (or current) are transmitted through the circuit formed between the auxiliary electrode and the working electrode, ensuring the continuous progress of the electrochemical reaction while preventing current from flowing through the reference electrode and affecting its potential stability. The electrolytic cell holds the soil and sediment water simulation solutions (serving as the test electrolyte in this embodiment), providing the necessary ionic conductivity environment and reaction site for the three-electrode system. The sensitive ends of the working electrode, reference electrode, and auxiliary electrode are all immersed in the same soil leachate, forming a complete electrochemical circuit through liquid-phase bonding. The electrochemical workstation, as the control and measurement center of the system, integrates a potentiostat, signal generator, and data acquisition module. According to the user-defined test program (such as cyclic voltammetry, impedance spectroscopy, etc.), the electrochemical workstation applies a precisely controlled potential signal to the working electrode through the three-electrode system and simultaneously and accurately measures the current signal flowing between the working electrode and the auxiliary electrode, thereby characterizing and analyzing specific electrochemical behaviors in the soil leachate. During testing, the electrochemical workstation controls the potential difference between the working electrode (WE) and the reference electrode (RE) according to a preset pattern; simultaneously, it measures the current generated between the working electrode (WE) and the auxiliary electrode (GE). This current-potential relationship contains rich electrochemical information, including redox reactions, mass transfer processes, and the interfacial double-layer structure at the soil leachate interface, enabling qualitative and quantitative analysis of specific pollutants or chemical components in the soil. The system performs test data acquisition and calculates electrochemical weighting factors. The core of this section lies in quantifying the contribution of each corrosion factor to the corrosion rate using a single-factor variable method combined with electrochemical response, thus providing experimental evidence for the conditional probability table of the Bayesian network. The specific implementation includes the following three stages:

[0090] Phase 1: Experimental Design Phase. First, a baseline operating condition is defined, based on field survey data to determine typical or average values ​​of various corrosion factors. Based on this, a baseline simulation solution is prepared, using analytical grade reagents and deionized water to precisely control the composition. The electrode system employs a three-electrode configuration: the working electrode is made of the same material as the tank bottom plate and undergoes standardized grinding, polishing, cleaning, and drying; a saturated calomel electrode is used as the reference electrode; and a platinum electrode is used as the auxiliary electrode. The electrochemical workstation must support open-circuit potential, linear polarization, and Tafel curve testing.

[0091] Phase Two: Single-Factor Electrochemical Experiments. For each corrosion factor (such as pH, salinity, sulfur content, etc.), the level of that factor is individually varied based on the baseline conditions. Typically, three levels are set: low, medium, and high, with the medium level corresponding to the baseline value. For each corrosion factor, a single-factor gradient experiment is conducted based on the baseline conditions. Taking pH as an example, four experimental levels are set: pH=4.0 (strongly acidic), pH=7.0 (baseline), pH=10.0 (alkaline), and pH=12.0 (strongly alkaline). Under each experimental condition, the concentrations of other factors are strictly kept constant to ensure the single-variable characteristics of the experimental results.

[0092] The test procedure includes: Open circuit potential monitoring: After immersing the working electrode in the solution, the open circuit potential change is continuously monitored. A stable state is considered reached when the potential fluctuation is less than 1 mV / min within 30 minutes; Linear polarization resistance test: Polarization is performed within the range of ±10 mV of the open circuit potential at a scan rate of 0.166 mV / s, and the polarization resistance Rp value is calculated using Gamry Framework software; Tafel curve test: The test is performed within the range of ±250 mV of the open circuit potential at a scan rate of 0.5 mV / s, and the corrosion current density is obtained by extrapolating the Tafel curves of the anodic and cathodic electrodes. Electrochemical impedance spectroscopy (EIS) testing: An AC perturbation with an amplitude of 10 mV was applied in the frequency range of 10 mHz to 100 kHz. The charge transfer resistance and double-layer capacitance parameters were obtained by equivalent circuit fitting. Each experimental condition was repeated at least three times, and the arithmetic mean was taken as the final result after removing outliers.

[0093] Phase 3: Experimental Data Processing and Weight Calculation. Based on the corrosion current density data obtained from Tafel curve testing, the annual corrosion rate is calculated according to Faraday's law:

[0094] ;

[0095] in, The annual corrosion rate is expressed as (mm / a). The value represents the corrosion current density (μA / cm²). It represents the atomic weight of the metal (g / mol). For electron transfer number, It is the Faraday constant (96485 C / mol). The density of the metal is (g / cm³).

[0096] The influence index of each corrosion factor was calculated using the range normalization method. For the first... Given a corrosion factor and a corrosion rate dataset obtained at k experimental levels, calculate:

[0097] ;

[0098] in, and These represent the maximum and minimum corrosion rates of this factor at the experimental level. The corrosion rate is the baseline operating condition.

[0099] The influence indices of each corrosion factor were normalized to obtain the weighting coefficients:

[0100] ;

[0101] The experimental weights were corrected using the Delphi method by experts, and corrosion experts were invited to independently evaluate the weight allocation based on their engineering experience. The final weight vector was obtained by combining experimental data and expert opinions using a weighted average method.

[0102] By calculating the electrochemical weighting factor as described above, the corrosion parameters measured in the laboratory are transformed into quantitative weights in the risk probability assessment, thus realizing the scientific transformation from experimental data to model parameters.

[0103] S3. Construct a multi-level, operational risk indicator system. The specific work involved in constructing this indicator system includes:

[0104] A complete indicator system framework was established. Based on industry standards and engineering practice experience, a structured indicator system comprising five levels was constructed. The top level clearly defines the overall assessment objectives, while the second level establishes the basic framework for the assessment from three dimensions: internal corrosion, external corrosion, and substrate quality. This hierarchical design ensures the systematic and comprehensive nature of the assessment.

[0105] Implement a systematic approach to refining and decomposing indicators. Each criterion level indicator is progressively broken down into multiple secondary indicators, and further refined into tertiary indicators, ultimately reaching the directly measurable lower-level indicators. During this refinement process, it is crucial to ensure the logical coherence and operational feasibility between indicators at each level, thereby forming a complete indicator logic chain.

[0106] Establish a clear system of indicator judgment standards. Set specific numerical thresholds for quantitative indicators and provide clear status descriptions for qualitative indicators. Simultaneously, establish unified judgment rules and operational guidelines to ensure that different evaluators can obtain consistent judgment results, thereby improving the repeatability of the evaluation.

[0107] Establish a standardized data collection and management system. Develop detailed data collection methods and recording requirements for each underlying indicator, including specific details such as sampling methods, testing conditions, and recording formats. Establish corresponding data quality control procedures to ensure the accuracy and reliability of the collected data, providing high-quality data input for subsequent analysis.

[0108] Improve the dynamic maintenance mechanism of the indicator system. Establish a version management and update maintenance system for the indicator system, and regularly optimize and adjust the indicator system based on feedback from actual applications and technological developments. Through this continuous improvement mechanism, maintain the advanced nature and applicability of the indicator system and adapt to the ever-changing technological environment.

[0109] The five-level risk indicator system constructed in this application, including its lowest level (directly measurable and assessable) indicators, state classifications, and complete corrosion and leakage system of the bottom plate of large port storage tanks, is shown in Table 1 below:

[0110] Table 1. Relationship between the lowest level indicators, state classification, and the complete corrosion and leakage system of the bottom plate of large port storage tanks.

[0111]

[0112]

[0113]

[0114] Furthermore, the establishment of a Bayesian network model based on Genie software specifically includes the following technical solutions:

[0115] In the Genie software environment, corresponding network nodes are created based on the hierarchical structure of the indicator system. It is ensured that the naming of each node accurately reflects its technical meaning, and that the definition of the node's state fully corresponds to the risk level in the indicator system. Specifically, in the soil corrosivity node, the corrosion factor weights obtained from electrochemical testing are used as key inputs to its conditional probability table definition, establishing a direct link from experimental data to corrosivity assessment. This step lays the structural foundation for subsequent probabilistic inference.

[0116] Specifically, a complete network topology is constructed. By analyzing the causal relationships between indicators, directed edges are used to establish logical connections from child nodes to parent nodes. After the network is constructed, a rigorous topology check is performed to ensure that a compliant directed acyclic graph structure is formed. Based on this, the rationality and completeness of the network structure are verified.

[0117] Figure 4 The diagram shows a partial schematic of the Bayesian network topology constructed in the Genie software environment according to an embodiment of this application. The diagram graphically illustrates the causal path and network connectivity from the bottom soil corrosivity indicators (such as soil resistivity and pH value) to the "external corrosion" criterion layer, and finally to the "leakage risk" target layer.

[0118] By deeply analyzing the soil corrosion mechanism and the interactions between various environmental parameters, a directed connection rule was established from child nodes (direct or indirect influencing factors) to parent nodes (affected objects). Based on this rule, a complete set of directed edges was constructed. The illustrated local topology clearly demonstrates some key causal relationships, for example:

[0119] Multiple parameters, including "soil moisture content", "sulfate-reducing bacteria content", "soil resistivity", "soil nitrogen ion content", and "soil pH value", serve as a parent node, directly pointing to the comprehensive assessment node of "soil corrosivity".

[0120] External environmental factors (such as stray currents and climatic conditions) and the "soil corrosivity" node, which characterizes the internal properties of the soil, work together as parent nodes, pointing to the more macroscopic "external corrosion" node.

[0121] Ultimately, the risks of "external corrosion," "internal corrosion," and "bottom plate quality" together determine the final "risk of corrosion and leakage of the bottom plate of large port storage tanks."

[0122] After the initial network structure is constructed, a rigorous topological logic check is performed. This process ensures that there are no loops of any kind in the entire network (i.e., starting from any node, it is impossible to return to that node along a directed edge), thus forming a correctly structured directed acyclic graph. This check is a prerequisite for ensuring that Bayesian networks can perform accurate probabilistic reasoning.

[0123] After passing the topology check, the rationality and integrity of the network structure need to be finally verified.

[0124] The established causal relationships are reviewed to ensure that all connections conform to the understanding of corrosion science and engineering practice, and to avoid erroneous associations that violate the mechanism.

[0125] Verify that all identified key risk factors have been incorporated into the network and that all critical direct causal relationships have been effectively expressed through directed edges, ensuring that the model can comprehensively cover the main risk paths of corrosion leakage from the tank bottom plate.

[0126] A prior probability distribution based on historical statistical data is established for the bottom-level nodes, while a conditional probability table combining expert knowledge and engineering experience is constructed for non-bottom-level nodes. When defining the conditional probability table for soil corrosivity nodes, the results of electrochemical weight factor calculations are integrated. For example, for a soil corrosivity node, its parent nodes include corrosion factors such as pH value and salinity. Each combination of parent node states corresponds to a conditional probability distribution of soil corrosivity, where the weight of each parent node is adjusted by Wi obtained from electrochemical experiments, thus reflecting the relative contribution of each factor to corrosivity. In this way, the conditional probability table not only integrates expert experience but also is based on experimental data, improving the objectivity and accuracy of the model.

[0127] Complex probabilistic relationships are reasonably simplified, and the model is validated and tested using historical case data. Based on the test results, the model parameters are adjusted and optimized as necessary. The focus is on validating the effectiveness of electrochemical weights in soil corrosivity assessment and optimizing the fusion method of these weights in the conditional probability table. This process ensures that the model has good predictive accuracy and practicality.

[0128] Establish version management and update / maintenance procedures for the model, and continuously improve model performance based on actual application results. Pay special attention to the degree of conformity between the electrochemical weighting factors and the actual situation, and adjust their integration strategy in the conditional probability table as needed. At the same time, establish model usage standards, and ensure the consistency of the evaluation process and the comparability of results through standardized operating procedures, providing institutional guarantees for long-term application.

[0129] The specific implementation process of dynamic risk reasoning, assessment, and in-depth analysis includes:

[0130] The first step is to conduct systematic evidence data collection and organization. This involves comprehensively collecting relevant data from the target storage tank, including on-site test results, operation and maintenance records, historical inspection reports, and other multi-source information. Particular emphasis is placed on the collection and organization of electrochemical test data, converting continuous parameters such as corrosion current density and polarization resistance into discrete state observation values ​​according to a pre-defined corrosion level classification standard. This step provides sufficient evidence to support the risk assessment, with electrochemical evidence providing crucial quantitative basis for judging soil corrosivity.

[0131] The second step involves implementing a standardized probabilistic reasoning calculation process. The processed state data is input as evidence into the established Bayesian network model, with electrochemical test results serving as a key piece of evidence for soil corrosivity assessment. The software's built-in probabilistic reasoning algorithm dynamically updates the probabilistic states of all nodes in the network. The Bayesian-based reasoning mechanism effectively integrates electrochemical data with other environmental parameters, handles uncertainties in multi-source evidence, and yields a comprehensive assessment of soil corrosivity.

[0132] The third step involves a comprehensive analysis and interpretation of the results. The focus is on the posterior probability distribution of the top-level risk nodes, combined with pre-defined risk level classification standards, to arrive at accurate risk assessment conclusions. Simultaneously, the contribution of electrochemical data to the final risk assessment is analyzed to verify its effectiveness in corrosivity judgment. By analyzing the development trends and changing patterns of the risk state, an intuitive risk situation awareness is provided for decision-making.

[0133] The fourth step is to conduct in-depth contribution analysis. Sensitivity analysis is used to calculate the contribution of each underlying factor to the overall risk, with particular attention to the weighting of electrochemical testing parameters in risk assessment. Key risk influencing factors are identified, and their mechanisms of action and changing characteristics are analyzed. This analytical process can reveal the underlying causes of risk formation, especially the intrinsic link between electrochemical activity and corrosion risk, providing targeted guidance for risk prevention and control.

[0134] Fifth, establish a dynamic and updated early warning mechanism system. Regularly collect new monitoring data, especially periodic electrochemical test results, and update model evidence information in a timely manner to achieve dynamic tracking and early warning of risks. Simultaneously, establish a comprehensive feedback and improvement mechanism, continuously optimizing the assessment process and analytical methods, paying particular attention to adjusting the weight of electrochemical data in the conditional probability table, to continuously improve the accuracy and timeliness of risk assessment.

[0135] Through the above systematic implementation steps, this invention establishes a complete dynamic risk assessment technology system. This system ensures input quality through standardized data collection, establishes an assessment framework based on a scientific indicator system, and innovatively transforms electrochemical test data into weighting factors using a single-factor variable method, integrating them into the conditional probability table of a Bayesian network. It then utilizes advanced Bayesian networks to achieve dynamic reasoning, ultimately providing decision support through in-depth analysis. The entire implementation process is clearly structured and tightly integrated, forming a complete logical loop, ensuring the accuracy and practicality of the assessment results.

[0136] Reference Figure 5 As shown, the present invention provides a tank bottom plate corrosion risk prediction device based on electrochemical testing, comprising:

[0137] The sample and parameter acquisition module 11 is used to acquire soil samples outside the tank and sediment water samples inside the tank of the tank to be predicted, and to acquire the original environmental parameters of the samples.

[0138] The corrosion testing module 12 is used to construct a soil simulation solution and a sedimentation water simulation solution based on the soil sample outside the tank and the sedimentation water sample inside the tank, respectively, and to add a standard metal sample of the same material as the bottom plate of the tank to be predicted to the soil simulation solution and the sedimentation water simulation solution, respectively, so as to realize external corrosion testing and internal corrosion testing, and obtain the corresponding corrosion test results.

[0139] The risk level prediction module 13 is used to input the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted into a preset corrosion risk assessment model, so as to predict the corrosion risk level of the bottom plate of the tank to be predicted and output the corrosion leakage risk level of the bottom plate of the tank to be predicted.

[0140] The preset corrosion risk assessment model is a base plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system consisting of internal corrosion parameters, external corrosion parameters, and base plate quality.

[0141] Therefore, by constructing a highly realistic internal and external corrosion environment through the sample and parameter acquisition module and the corrosion testing module, and directly using samples of the same material as the bottom plate for testing, the limitations of traditional single-environment assessment or theoretical calculation are overcome. This makes the corrosion data more representative and reliable, laying a solid foundation for accurate prediction. Secondly, the risk level prediction module adopts a pre-built corrosion risk assessment model based on a Bayesian network, which can effectively integrate multi-source heterogeneous data (including environmental parameters, experimental test results, and bottom plate quality information) and quantify the complex relationships and uncertainties between various risk factors. This achieves a leap from isolated data analysis to multi-factor coupled reasoning, significantly improving the comprehensiveness and accuracy of risk prediction. Transforming complex corrosion problems into intuitive risk level outputs enables a scientific and quantitative assessment of the probability of leakage due to bottom plate corrosion, providing a direct and reliable basis for maintenance decisions and effectively preventing leakage accidents caused by bottom plate corrosion.

[0142] Furthermore, embodiments of this application also disclose an electronic device, Figure 6 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0143] Figure 6This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the electrochemical testing-based corrosion risk prediction method for storage tank bottom plates disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0144] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0145] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0146] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0147] The operating system 221 manages and controls the various hardware devices and computer programs 222 on the electronic device 20 to enable the processor 21 to perform calculations and processing on the massive amounts of data 223 in the memory 22. It can be Windows Server, Netware, Unix, Linux, etc. The computer program 222, in addition to including a computer program capable of performing the electrochemical testing-based corrosion risk prediction method for the tank bottom plate disclosed in any of the foregoing embodiments, may further include computer programs capable of performing other specific tasks. The data 223 may include data received by the electronic device from external devices, as well as data collected by its own input / output interface 25.

[0148] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for predicting the corrosion risk of tank bottom plates based on electrochemical testing. The specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0149] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0150] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application. The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly in hardware, software modules executed by a processor, or a combination of both. The software module may be located in random access memory (RAM), memory, read-only memory (ROM), electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, removable disks, CD-ROMs (Compact Disc-Read Only Memory), or any other form of storage medium known in the art.

[0151] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0152] The solution provided by the present invention has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for predicting corrosion risk of tank bottom plates based on electrochemical testing, characterized in that, include: Obtain soil samples from outside the tank and sediment water samples from inside the tank of the tank to be predicted, and obtain the original environmental parameters of the samples; Based on the soil sample outside the tank and the sediment water sample inside the tank, a soil simulation solution and a sediment water simulation solution are constructed respectively. A standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the soil simulation solution and the sediment water simulation solution respectively to achieve external corrosion test and internal corrosion test, and obtain the corresponding corrosion test results. The original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted are input into the preset corrosion risk assessment model to predict the corrosion risk level of the bottom plate of the tank to be predicted and output the corrosion leakage risk level of the bottom plate of the tank to be predicted. The preset corrosion risk assessment model is a base plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system consisting of internal corrosion parameters, external corrosion parameters, and base plate quality.

2. The method for predicting corrosion risk of storage tank bottom plate based on electrochemical testing according to claim 1, characterized in that, The step involves adding a standard metal sample of the same material as the bottom plate of the tank to be tested to the soil simulation solution and the sediment water simulation solution, respectively, to achieve external corrosion testing and internal corrosion testing, and to obtain the corresponding corrosion test results, including: A standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the soil simulation liquid to perform potentiodynamic polarization curve testing and electrochemical impedance spectroscopy testing to obtain the first electrochemical parameter in order to realize external corrosion testing. A standard metal sample of the same material as the bottom plate of the tank to be predicted is added to the sedimentation water simulation solution to perform potentiodynamic polarization curve testing and electrochemical impedance spectroscopy testing to obtain a second electrochemical parameter for internal corrosion testing. The internal corrosion rate and external corrosion rate are calculated based on the first electrochemical parameter and the second electrochemical parameter, respectively, and are used as corrosion test results.

3. The method for predicting corrosion risk of tank bottom plate based on electrochemical testing according to claim 2, characterized in that, Before inputting the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted into the preset corrosion risk assessment model, the method further includes: A risk indicator system is constructed based on internal corrosion parameters, external corrosion parameters, and base plate quality. Create corresponding network nodes for each layer of risk indicators in the aforementioned risk indicator system; Define a discrete state for each network node that corresponds to a preset risk state in the risk indicator system; Based on the causal relationships in the risk indicator system, directed edges are used to connect network nodes representing lower-level risk indicators to network nodes representing higher-level risk indicators, thereby constructing a directed acyclic graph structure.

4. The method for predicting corrosion risk of tank bottom plate based on electrochemical testing according to claim 3, characterized in that, Also includes: A prior probability distribution is defined for network nodes in the directed acyclic graph structure corresponding to underlying risk indicators without parent nodes, and a conditional probability table is defined for network nodes in the directed acyclic graph structure corresponding to non-underlying risk indicators with parent nodes, so as to realize the network parameter definition process of the preset corrosion risk assessment model.

5. The method for predicting corrosion risk of tank bottom plate based on electrochemical testing according to claim 4, characterized in that, The conditional probability table for network nodes in the directed acyclic graph structure corresponding to non-bottom-level risk indicators with parent nodes includes: Based on the first electrochemical parameter and the second electrochemical parameter, the weight parameters of each corrosion risk factor on the corrosion rate are calculated by the single-factor variable electrochemical test method. Based on the weight parameters, a conditional probability table is defined for network nodes in the directed acyclic graph structure corresponding to non-bottom-level risk indicators with parent nodes.

6. The method for predicting corrosion risk of tank bottom plate based on electrochemical testing according to claim 1, characterized in that, The process of predicting the corrosion risk level of the bottom plate of the tank to be predicted and outputting the corrosion leakage risk level of the bottom plate of the tank to be predicted includes: The original environmental parameters, corrosion test results, and bottom plate information of the tank to be predicted are processed in a state-based manner according to the judgment criteria of the corresponding underlying risk indicators in the risk indicator system by the preset corrosion risk assessment model, so as to input the state-based processing results into the corresponding network node in the preset corrosion risk assessment model. Probabilistic inference is performed using the preset corrosion risk assessment model to update the posterior probability distribution of all network nodes in the Bayesian network. The risk level of bottom plate corrosion leakage of the tank to be predicted is determined based on the posterior probability distribution of the network nodes representing the top-level risk indicators.

7. The method for predicting corrosion risk of tank bottom plate based on electrochemical testing according to any one of claims 1 to 6, characterized in that, After outputting the bottom plate corrosion leakage risk level of the tank to be predicted, the method further includes: The corrosion risk assessment model is used to analyze the contribution of the bottom plate of the tank to be predicted to corrosion leakage risk. The contribution of each underlying risk indicator to the corrosion leakage risk level of the bottom plate is calculated and ranked to determine the target corrosion risk factor.

8. A device for predicting corrosion risk of storage tank bottom plate based on electrochemical testing, characterized in that, include: The sample and parameter acquisition module is used to acquire soil samples outside the tank and sediment water samples inside the tank of the tank to be predicted, and to acquire the original environmental parameters of the samples. The corrosion testing module is used to construct soil simulation solution and sedimentation water simulation solution based on the soil sample outside the tank and the sedimentation water sample inside the tank, respectively, and to add a standard metal sample of the same material as the bottom plate of the tank to be tested to the soil simulation solution and the sedimentation water simulation solution, respectively, so as to realize external corrosion test and internal corrosion test and obtain the corresponding corrosion test results. The risk level prediction module is used to input the original environmental parameters, the corrosion test results, and the bottom plate information of the tank to be predicted into a preset corrosion risk assessment model, so as to predict the corrosion risk level of the bottom plate of the tank to be predicted and output the corrosion leakage risk level of the bottom plate of the tank to be predicted. The preset corrosion risk assessment model is a base plate corrosion probability model constructed based on a Bayesian network, and the network structure of the Bayesian network is constructed based on a risk index system consisting of internal corrosion parameters, external corrosion parameters, and base plate quality.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the method for predicting corrosion risk of tank bottom plate based on electrochemical testing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the method for predicting corrosion risk of tank bottom plate based on electrochemical testing as described in any one of claims 1 to 7.

Images