Modeling localized corrosion of pipeline steels in oilfield produced water environments

Velázquez, J. C.; Cruz-Ramirez, J.C.; Valor, A.; Venegas, V.; Caleyo, F.; Hallen, J.M.

doi:10.1016/j.engfailanal.2017.04.027

Cited by 45 publications

(21 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In materials science and engineering, the most commonly implemented probability distributions are the normal distribution [24], Weibull distribution [25], Poison distribution [26], gamma distribution [27], log-normal distribution [28], exponential distribution [17], Gumbel distribution [29,30], and GEV distribution [31]. The frequency distribution of the layer thickness parameter is presented in a discrete mode.…”

Section: Mathematical Foundationmentioning

confidence: 99%

A Stochastic Model and Investigation into the Probability Distribution of the Thickness of Boride Layers Formed on Low-Carbon Steel

et al. 2019

View full text Add to dashboard Cite

The stochastic nature of the thickness of boride layers formed on carbon steel is described in this paper. Additionally, the probability distribution of the layer thickness is studied to determine the best-fit probability distribution. The study combines the use of an empirical model (power-law) and the Markov chain principles, with the purpose of demonstrating that it is feasible to develop a model that represents the non-uniformity of the thickness of boride layers that form on carbon steel. The results indicate that the mean and variance tend to increase when the time or temperature is increased. The findings of this paper demonstrate that an analytical solution to the Kolmogorov’s system differential equation can adequately represent the behavior of non-uniform boride layer formed on low-carbon steel, regardless of the temperature or duration of treatment.

show abstract

Section: Mathematical Foundationmentioning

confidence: 99%

A Stochastic Model and Investigation into the Probability Distribution of the Thickness of Boride Layers Formed on Low-Carbon Steel

et al. 2019

View full text Add to dashboard Cite

show abstract

“…In the Equation 1 [46], it was used to normalize the variables [46]: In the equation above, * X is the normalized value of the variable, X is the actual value, min X is the minimum value and max X is the maximum value. The boxplot graph is divided by minimum value, first quartile, median, third quartile and maximum.…”

Section: Depth [M]mentioning

confidence: 99%

Fracture Toughness and Charpy CVN Data for A36 Steel with Wet Welding

et al. 2017

View full text Add to dashboard Cite

This study presents K IC data obtained from K IC -CVN correlations from Charpy CVN values. For this study, T-welded connections were manufactured from ASTM A36 and E6013 electrodes in dry conditions. Then, a rectangular grinding at the weld toe was carried out and filled with wet welding. Charpy specimens were extracted to obtain CVN values. An exhaustive search through the literature of several authors was performed to collect experimental CVN data about wet welding being applied to A36 steel for comparison with CVN data obtained in this study. By using Charpy impact energy (CVN), K IC values could be predicted by K IC -CVN correlations. In addition, correlations were presented to obtain K IC values in the lower shelf, transition temperature zones and different zones for the energy-temperature curve of A36 steel. Of these correlations, Barsom's equation was adopted, because he applied the stress yield (σ YS ) of the material and it can be applied in all zones for the energy-temperature curve. The results revealed that CVN values are proportionate to K IC , this data decreases as water depth increases. This took place because several discontinuities, such as, porosity, slag inclusion, non-metallic inclusion, cracking and microstructures are present in the wet welding.

show abstract

“…They proposed a non-deterministic artificial intelligence method that estimated the corrosion at different sections of the pipeline. Many other investigations on the use of artificial intelligence in the estimation of the corrosion defect growth of carbon steel materials used for pipelines also abound in the literature [14][15][16][17][18][19][20][21][22][23]. Notable among these studied is the work on the fatigue crack growth where ANN was employed to investigate the corrosion-fatigue crack of a dual phase steel at different stress intensities inconsideration of the martensite content of 32-76% [23].…”

Section: Introductionmentioning

confidence: 99%

A Data-Driven Machine Learning Approach for Corrosion Risk Assessment—A Comparative Study

Ossai

2019

BDCC

View full text Add to dashboard Cite

Understanding the corrosion risk of a pipeline is vital for maintaining health, safety and the environment. This study implemented a data-driven machine learning approach that relied on Principal Component Analysis (PCA), Particle Swarm Optimization (PSO), Feed-Forward Artificial Neural Network (FFANN), Gradient Boosting Machine (GBM), Random Forest (RF) and Deep Neural Network (DNN) to estimate the corrosion defect depth growth of aged pipelines. By modifying the hyperparameters of the FFANN algorithm with PSO and using PCA to transform the operating variables of the pipelines, different Machine Learning (ML) models were developed and tested for the X52 grade of pipeline. A comparative analysis of the computational accuracy of the corrosion defect growth was estimated for the PCA transformed and non-transformed parametric values of the training data to know the influence of the PCA transformation on the accuracy of the models. The result of the analysis showed that the ML modelling with PCA transformed data has an accuracy that is 3.52 to 5.32 times better than those carried out without PCA transformation. Again, the PCA transformed GBM model was found to have the best modeling accuracy amongst the tested algorithms; hence, it was used for computing the future corrosion defect depth growth of the pipelines. This helped to compute the corrosion risks using the failure probabilities at different lifecycle phases of the asset. The excerpts from the results of this study indicate that my technique is vital for the prognostic health monitoring of pipelines because it will provide information for maintenance and inspection planning.

show abstract

Modeling localized corrosion of pipeline steels in oilfield produced water environments

Cited by 45 publications

References 40 publications

A Stochastic Model and Investigation into the Probability Distribution of the Thickness of Boride Layers Formed on Low-Carbon Steel

A Stochastic Model and Investigation into the Probability Distribution of the Thickness of Boride Layers Formed on Low-Carbon Steel

Fracture Toughness and Charpy CVN Data for A36 Steel with Wet Welding

A Data-Driven Machine Learning Approach for Corrosion Risk Assessment—A Comparative Study

Contact Info

Product

Resources

About