Natural gas submarine pipeline internal corrosion simulation analysis method

Abstract

The invention discloses a natural gas submarine pipeline internal corrosion simulation analysis method. Aiming at the problems of difficulty in data acquisition, serious corrosion and the like in corrosion monitoring in the submarine pipeline, key parameters along the pipeline are accurately calculated by fusing field key node data acquisition and OLGA software numerical simulation; determining corrosion main control factors according to the key parameters, deeply analyzing the main control corrosion factors and corrosion rate data by utilizing an Apriori algorithm, and mining an association rule of the main control corrosion factors and the corrosion rate data; and finally, based on the association rule obtained by mining, constructing a semi-theoretical semi-empirical corrosion prediction model, and providing a scientific basis and an effective method for monitoring and protecting corrosion in the submarine pipeline.

Description

Technical Field

[0001] The invention relates to a natural gas submarine pipeline internal corrosion simulation analysis method and relates to the field of natural gas submarine pipelines. Background Art

[0002] Against the backdrop of continued global energy demand growth, submarine natural gas pipelines, with their advantages of high transmission efficiency and strong stability, have become a core channel for deep-sea natural gas resource development and terrestrial energy supply. However, due to the long-term exposure of submarine pipelines to the high-pressure, low-temperature environment of the deep sea, and the fact that the transmission medium often contains corrosive components such as hydrogen sulfide, carbon dioxide, and water, the inner walls of the pipelines are highly susceptible to chemical corrosion, electrochemical corrosion, and erosion corrosion. Internal corrosion can lead to thinning of pipeline walls and reduced strength, which can in turn cause serious accidents such as leaks and pipe bursts. This not only wastes natural gas resources and pollutes the environment, but can also endanger the lives of personnel on and around offshore platforms, resulting in significant economic losses and social risks.

[0003] Because the pipeline is laid in the deep sea, data collection faces dual challenges in terms of technology and cost. On the one hand, the extreme deep-sea environment places extremely high demands on monitoring equipment, making it difficult for conventional sensors to operate stably, resulting in discontinuous data collection. On the other hand, due to the long length of the pipeline and economic and technical limitations, data can only be collected at a limited number of key nodes, and it is impossible to cover the entire pipeline. Parameters such as temperature, pressure, CO2 partial pressure, and wall shear force significantly affect the corrosion rate within the pipeline. However, due to the limitations of deep-sea operations and monitoring technology, it is difficult to fully obtain these parameters through on-site measurements. The lack of these key parameters makes it difficult to accurately assess the corrosion rate within the pipeline, making it difficult to formulate anti-corrosion measures. Traditional methods such as coating protection and corrosion inhibitor addition are difficult to achieve effective prevention and control due to a lack of data support.

[0004] To accurately monitor the internal condition of natural gas submarine pipelines, this study combines data collection from key on-site nodes with numerical simulation using the OLGA software to systematically analyze operational parameters along the pipeline, such as pressure, temperature, and flow rate. By preprocessing and extracting features from the collected data, the primary factors influencing pipeline corrosion rates are identified. The Apriori association rule mining algorithm is then used to deeply analyze the quantitative relationship between these key factors and corrosion rates. Ultimately, based on the integration of theoretical derivation and empirical data, a semi-theoretical and semi-empirical corrosion prediction model formula for natural gas submarine pipelines is constructed, providing a scientific basis for pipeline corrosion status assessment and maintenance decisions. Summary of the Invention

[0005] The present invention aims to solve the deficiencies in the above technologies and designs a method for simulating and analyzing corrosion in natural gas submarine pipelines.

[0006] The natural gas submarine pipeline internal corrosion simulation analysis method specifically includes the following steps:

[0007] Step 1: Deploy sensors at key nodes of the natural gas submarine pipeline to collect field data such as temperature, pressure, and flow;

[0008] Step 2: Input the key node data on site into the OLGA software to establish a natural gas submarine pipeline model, and simulate and calculate key parameters along the pipeline, such as temperature, pressure, CO2 partial pressure, H2S partial pressure, wall shear force, and liquid holdup.

[0009] Step 3: Data preprocessing and determination of main controlling factors of corrosion;

[0010] Step 4: Use the Apriori algorithm to analyze the main controlling corrosion factors and corrosion rate data, and mine the association rules between the main controlling corrosion factors and corrosion rate;

[0011] Step 5: Based on the association rules of the main controlling corrosion factors, a semi-theoretical and semi-empirical model of the corrosion rate in the pipeline is established.

[0012] Furthermore, the key data collected in step 1 include multiple corrosion factors such as pipeline pressure, temperature, flow rate, CO2 partial pressure, H2S partial pressure, etc.

[0013] Furthermore, the step 3 specifically includes the following steps:

[0014] Step 3.1: Preprocess the data;

[0015] Step 3.2: Use grey relational analysis (GRG) to analyze the main controlling factors of corrosion.

[0016] Further, the step 3.1 is specifically as follows:

[0017] Taking the corrosion rate as the reference sequence, the corrosion factors and corrosion rates are dimensionlessly processed by formula (1);

[0018]

[0019] In formula (1), x j (h) is the jth influencing factor in the hth group of data, x j (l) is the average value of the jth influencing factor, x j '(h) is the dimensionless data, n is the number of influencing factors, and m is the number of data groups.

[0020] Further, the step 3.2 is specifically as follows:

[0021] The grey relational grade (GRG) is calculated by formula (2) and formula (3). The larger the GRG, the greater the correlation between the influencing factors and the corrosion rate. When GRG>0.6, the correlation is strong.

[0022]

[0023]

[0024] In formula (2), ξ j (h) is the correlation coefficient, y ’ (h) is the dimensionless reference order, is the reference sequence y(h) and x j (h) the minimum difference between the two levels, is the reference sequence y(h) and x j (h) The maximum difference between the two levels, ρ is the resolution coefficient, ρ∈(0,+∞), usually taken as 0.5;

[0025] In formula (3), r j is the degree of association, ranging from 0 to 1.

[0026] Furthermore, step 4 specifically includes the following steps:

[0027] Step 4.1: Input the total dataset D, support threshold σ, confidence threshold α, and lift threshold δ;

[0028] Step 4.2, calculate the support between the corrosion rate and the corrosion factor;

[0029] Step 4.3, calculate the confidence level between the corrosion rate and the corrosion factor;

[0030] Step 4.4, calculate the improvement between the corrosion rate and the corrosion factor;

[0031] Step 4.5: Derive all association rules that meet 4.2, 4.3, and 4.4.

[0032] Further, the step 4.2 is specifically as follows:

[0033] The support is calculated by formula (4). If the support threshold σ is met, the next step of calculation is entered;

[0034]

[0035] In formula (4), X and Y are the conditional dataset and result dataset respectively, Support(X→Y) is the support under the (X→Y) association rule, │X∪Y│ is the number of datasets containing both X and Y, and │D│ is the total number of datasets.

[0036] Further, the step 4.3 is specifically as follows:

[0037] The confidence is calculated by formula (5). If the confidence threshold α is met, the next step of calculation is entered;

[0038]

[0039] In formula (5), Confidence(X→Y) is the confidence under the (X→Y) association rule, and Support(X) is the support of the conditional dataset X.

[0040] Further, the step 4.4 is specifically as follows:

[0041] The lift is calculated by formula (6). If the lift threshold δ is satisfied, the association rule requirements are met;

[0042]

[0043] In formula (6), Lift(X→Y) is the lift under the (X→Y) association rule, and Support(Y) is the support of the result dataset Y. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the natural gas submarine pipeline internal corrosion simulation analysis method provided by the present invention. Specific implementation methods

[0045] See also Figure 1 As shown, the present invention provides a method for simulating and analyzing corrosion in a natural gas submarine pipeline, which specifically includes the following steps:

[0046] Step 1: Collect key data including pipeline pressure, temperature, flow, CO2 partial pressure, H2S partial pressure and other corrosion factors.

[0047] Step 2: Input the key node data on site into the OLGA software to establish a natural gas submarine pipeline model, and simulate and calculate key parameters along the pipeline, such as temperature, pressure, CO2 partial pressure, H2S partial pressure, wall shear force, and liquid holdup.

[0048] Step 3: Data preprocessing and determination of the main controlling factors of corrosion. The specific process is as follows:

[0049] Step 3.1: Preprocess the data;

[0050] Further, the step 3.1 is specifically as follows:

[0051] Taking the corrosion rate as the reference sequence, the corrosion factors and corrosion rates are dimensionlessly processed by formula (1);

[0052]

[0053] In formula (1), x j (h) is the jth influencing factor in the hth group of data, x j (l) is the average value of the jth influencing factor, x j'(h) is the dimensionless data, n is the number of influencing factors, and m is the number of data groups.

[0054] Step 3.2: Use grey relational analysis (GRG) to analyze the main controlling factors of corrosion.

[0055] Further, the step 3.2 is specifically as follows:

[0056] The grey relational grade (GRG) is calculated by formula (2) and formula (3). The larger the GRG, the greater the correlation between the influencing factors and the corrosion rate. When GRG>0.6, the correlation is strong.

[0057]

[0058]

[0059] In formula (2), ξ j (h) is the correlation coefficient, y ’ (h) is the dimensionless reference order, is the reference sequence y(h) and x j (h) the minimum difference between the two levels, is the reference sequence y(h) and x j (h) The maximum difference between the two levels, ρ is the resolution coefficient, ρ∈(0,+∞), usually taken as 0.5;

[0060] In formula (3), r j is the degree of association, ranging from 0 to 1.

[0061] Step 4: Use the Apriori algorithm to analyze the main controlling corrosion factors and corrosion rate data to mine the association rules between the main controlling corrosion factors and corrosion rates. The specific process is as follows:

[0062] Step 4.1: Input the total dataset D, support threshold σ, confidence threshold α, and lift threshold δ;

[0063] Step 4.2, calculate the support between the corrosion rate and the corrosion factor;

[0064] Further, the step 4.2 is specifically as follows:

[0065] The support is calculated by formula (4). If the support threshold σ is met, the next step of calculation is entered;

[0066]

[0067] In formula (4), X and Y are the conditional dataset and result dataset respectively, Support(X→Y) is the support under the (X→Y) association rule, │X∪Y│ is the number of datasets containing both X and Y, and │D│ is the total number of datasets.

[0068] Step 4.3, calculate the confidence level between the corrosion rate and the corrosion factor;

[0069] Further, the step 4.3 is specifically as follows:

[0070] The confidence is calculated by formula (5). If the confidence threshold α is met, the next step of calculation is entered;

[0071]

[0072] In formula (5), Confidence(X→Y) is the confidence under the (X→Y) association rule, and Support(X) is the support of the conditional dataset X.

[0073] Step 4.4, calculate the improvement between the corrosion rate and the corrosion factor;

[0074] Further, the step 4.4 is specifically as follows:

[0075] The lift is calculated by formula (6). If the lift threshold δ is satisfied, the association rule requirements are met;

[0076]

[0077] In formula (6), Lift(X→Y) is the lift under the (X→Y) association rule, and Support(Y) is the support of the result dataset Y.

[0078] Step 4.5: Derive all association rules that meet 4.2, 4.3, and 4.4.

[0079] Step 5: Based on the association rules of the main controlling corrosion factors, a semi-theoretical and semi-empirical model prediction formula (7) for the corrosion rate in the pipeline is established;

[0080]

[0081] In formula (7), CR is the corrosion rate (mm / a), R is the gas constant (8.314 J / (mol·K)), T is the absolute temperature (K), and E is the corrosion rate (mm / a). α is the activation energy (J / mol), P CO2 、P H2S is the gas partial pressure (bar), v is the flow rate (m / s), P is the system pressure (bar), and a, b, c, d, and α are all constants.

Claims

1. A method for simulating and analyzing internal corrosion of a natural gas submarine pipeline, comprising the following steps: Step 1: Deploy sensors at key nodes of the natural gas submarine pipeline to collect field data such as temperature, pressure, and flow; Step 2: Input the key node data on site into the OLGA software to establish a natural gas submarine pipeline model and simulate and calculate key parameters along the pipeline, such as temperature, pressure, CO2 partial pressure, H2S partial pressure, wall shear force, and liquid holdup; Step 3: Data preprocessing and determination of main controlling factors of corrosion; Step 4: Use the Apriori algorithm to analyze the main controlling corrosion factors and corrosion rate data, and mine the association rules between the main controlling corrosion factors and corrosion rate; Step 5: Based on the association rules of the main controlling corrosion factors, a semi-theoretical and semi-empirical model of the corrosion rate in the pipeline is established.

2. A natural gas submarine pipeline internal corrosion simulation analysis method according to claim 1, characterized in that: Step 4 specifically includes the following steps: Step 4.1: Input the total dataset D, support threshold σ, confidence threshold α, and lift threshold δ; Step 4.2, calculate the support between the corrosion rate and the corrosion factor; Step 4.3, calculate the confidence level between the corrosion rate and the corrosion factor; Step 4.4, calculate the improvement between the corrosion rate and the corrosion factor; Step 4.5: Derive all association rules that meet 4.2, 4.3, and 4.

4.

3. A natural gas submarine pipeline internal corrosion simulation analysis method according to claim 2, characterized in that: The step 4.2 is specifically as follows: The support is calculated by formula (4). If the support threshold σ is met, the next step of calculation is entered; In formula (4), X and Y are the conditional dataset and result dataset respectively, Support(X→Y) is the support under the (X→Y) association rule, │X∪Y│ is the number of datasets containing both X and Y, and │D│ is the total number of datasets.

4. A natural gas submarine pipeline internal corrosion simulation analysis method according to claim 2, characterized in that: The step 4.3 is specifically as follows: The confidence is calculated by formula (5). If the confidence threshold α is met, the next step of calculation is entered; In formula (5), Confidence(X→Y) is the confidence under the (X→Y) association rule, and Support(X) is the support of the conditional dataset X.

5. The method for simulating and analyzing internal corrosion of a natural gas submarine pipeline according to claim 2, characterized in that: The step 4.4 is specifically as follows: The lift is calculated by formula (6). If the lift threshold δ is satisfied, the association rule requirements are met; In formula (6), Lift(X→Y) is the lift under the (X→Y) association rule, and Support(Y) is the support of the result dataset Y.

6. The method for simulating and analyzing internal corrosion of a natural gas submarine pipeline according to claim 1, characterized in that: Step 5 specifically includes the following steps: Based on the association rules of the main controlling corrosion factors, a semi-theoretical and semi-empirical model prediction formula for the corrosion rate in pipelines is established (7); In formula (7), CR is the corrosion rate (mm / a), R is the gas constant (8.314 J / (mol·K)), T is the absolute temperature (K), and E is the corrosion rate (mm / a). α is the activation energy (J / mol), P CO2 、P H2S is the gas partial pressure (bar), v is the flow rate (m / s), P is the system pressure (bar), and a, b, c, d, and α are all constants.

Images