An integrated dynamic failure assessment model for offshore components under microbiologically influenced corrosion
Introduction
Corrosion plays a critical role in the failure of infrastructures in the oil and gas industry. Several catastrophic steel failure events in marine and offshore environments have been attributed to undesirable corrosion phenomenon (Paik and Kim, 2012). The complexity of the corrosion mechanisms in the petroleum industry depends on numerous operational, environmental, and material related factors (Chandrasekaran, 2016; Chandrasekaran and Jain, 2016; Little and Lee, 2007). These factors may include biofouling, presence of carbon dioxide, pH, pollutants, temperature, pressure, water velocity, carbonate solubility, salinity, amount of suspended solids, presence of bacteria, material composition, and surface roughness. In petroleum production processes, oil, gas, and water exist under various flow regimes and process conditions; due to the ionic nature of oil and gas, the emulsion is formed at a low concentration of water. The emulsions may exist in the form of either water-in-oil or oil-in-water (Papavinasam et al., 2007, 2010). The oil-in-water emulsions, having contact with the pipe wall, provide a stimulating environment that may favor microbial growth and carbonic acid formation due to CO2 dissolution (Nesic et al., 2005; Papavinasam et al., 2007). The resulting carbonic acid may initiate different types of internal corrosion, and the bacteria enhance their growth, especially in carbon steel pipelines (Renpu, 2011).
Several corrosion-induced failures occur in vertical and/or deviated subsea oil pipelines as a result of water hang-up, which provides a stimulating environment that activates corrosion influencing factors such as bacteria, dissolved salts, and volatile fatty acids (Ibrahim et al., 2018). The water hang-up builds sufficient pressure that causes periodic spasms of slugging (Palmer and King, 2008). At high slug formation frequencies of 50–90 per minute, the corrosion rate may linearly increase with slug frequency (Jepson, 1996). The slug frequency depends on the inclination of the pipeline, and the corrosion rate increases by 50% when the slug frequency is doubled (Jepson, 1996). In addition, steel composition and microstructural configuration, as well as the micro-organisms in the area of corrosion initiation, play vital roles in the corrosion growth rate. This is shown in the high rate of material degradation in the dynamic and bacteria-infested environment (Marciales et al., 2019). According to the literature, the failure of the marine and offshore oil and gas infrastructures due to corrosion defects significantly depends on the depth of the defect and the corrosion growth rate. Therefore, a proper understanding of corrosive environmental dynamics and the material response is crucial in failure prediction and corrosion management in the oil and gas industry. Moreover, the associated dynamics in corrosion mechanisms under the microbial influence need to be adequately understood to predict and also inhibit the risks of failure of critical infrastructures in the marine and offshore industry.
Microbiologically influenced corrosion (MIC) describes the degradation process of various systems instigated by the presence and metabolic activities of micro-organisms such as bacteria and fungi (Beech and Gaylarde, 1999). The formation of metabolites (organic and inorganic acids) by the bacteria influences the electrochemical mechanisms and complicates the corrosion process. Previous research studies have shown that microbiologically influenced internal corrosion contributes to several onshore and offshore systems' failure with catastrophic consequences (Al-jaroudi et al., 2011; Eckert, 2003; Witt et al., 2016). The microbial growth is promoted by the availability of the supporting nutrients in the environment. These nutrients synergize with the metallic surface and abiotic corrosion product to provide a sustainable growing environment for the bacteria.
Heterogeneous material surfaces in the presence of water accelerate the formation of bacteria colonies called biofilms. The fused microbial cells and extracellular polymeric substances (EPS) form the biofilm, which provides a favorable mode of subsistence for the microorganisms, even in a hostile environment. The biofilm also promotes a more sustainable environment that enhances reproduction and growth of metabolisms, greatly influencing the corrosion mechanisms (Beech and Gaylarde, 1999; Machuca, 2016). The polymeric substances within the multispecies bacteria colony produce a mixed complex array of dynamic corrosive microenvironments that boost the steel material deterioration. This complexity poses difficulty in the understanding of the microbial process and addressing the subsequent challenges (Machuca, 2016).
The marine and offshore infrastructures under the microbial biofilm complexity continue to experience severe degradation, especially in micro-organism communities where the sulphate-reducing bacteria, iron-oxidizing bacteria, manganese-oxidizing, sulphate-oxidizing bacteria, acid-producing bacteria, and the exopolymers coexist in the same colony or biofilm (Beech and Gaylarde, 1999). This synergistic consortium alters the electrochemical processes, resulting in microbiologically influenced localized pitting corrosion that reduces the integrity of the structure. The byproduct of the bacteria metabolism cracks the corrosion protection layer, thereby exposes the steel material to severe degradation (Chandrasekaran and Jain, 2016). The loss of structural integrity occurs when the structure becomes susceptible owing to the breakdown of the thin passive oxide film that resists corrosion. The breakdown is due to the formation of organic and inorganic deposits on the structure surfaces that compromise the stability of the oxide film (Beech and Gaylarde, 1999). Further growth of the micro-organisms sustains the growth of pit formation and pit density across the length of the pipeline, resulting in MIC induced failures of offshore, shipping, and process systems, for example, pipeline leakages and ruptures, ballast tank, and cargo tank leakages (Beech and Gaylarde, 1999; Eckert, 2003; Paik and Kim, 2012). These MIC induced failures lead to direct and indirect consequences with associated economic losses/risks.
To model the MIC potential and its propagation, several researchers have proposed mechanistic models (Marciales et al., 2019). For instance, Gu et al. (2009) proposed a bioenergetic-based theory to describe the thermodynamic mechanism for Type I MIC formation by sulphate-reducing bacteria (SRB). The Type 1 MIC formation occurs due to the process of microbes' respiration on exogenous oxidants. This process involves an extracellular electron transfer by the sulphate or nitrate ions into the microbial cytoplasm. This is mostly involved in electrogenic biofilms' formation. The authors emphasized the possibility of alteration in the electrochemical corrosion mechanisms during this phenomenon. They argued that the process of thermodynamic equilibrium-based potential analysis only determines the potential of MIC formation but does not alter the rate of corrosion wastage. They also concluded that an integrated mechanistic model through involvement of microbial growth kinetics, mass transfer, and various chemical, biochemical, and electrochemical reactions would provide a more reliable tool for prediction of MIC potential.
Sørensen et al. (2012) proposed a risk-based MIC model for the worst-case pitting corrosion rate and risk factors based on sulphate reducing archaea (SRA), sulphate reducing bacteria (SRB), and methanogens (MET). The authors showed that in the combined colony of the bacteria, the rate of wastage increases; they suggested a proactive plan for the potentially high pit generation rate due to the exponential growth of microbial cells. Al-Darbi et al. (2005) developed a mathematically and/or numerically-based polarization model. They described the cathodic SRB mediated polarization as it affects the corrosion rate overtime at different pit depth increments. The developed anaerobic model in the SRB environment was based on cathodic depolarization, which describes the corrosion rate dependency on the consumption rate of sulphate by the SRB and the change in pit depth. More details on the mechanistic models for MIC potential and rate prediction and their limitations are presented in Marciales et al. (2019). Generally, these models are not dynamically structured to reflect the complexity and nonlinear interdependency among MIC influencing factors for real-time application and failure probability prediction.
In the recent years, advanced models/approaches that better predict the microbial potential based on the screening and operating parameters, and their influence on corrosion wastage of oil and gas infrastructures have been introduced. These include experimental, regression, quantitative, and probabilistic tools (Heyer, 2013; Huang et al., 1997; Liu and Cheng, 2018; Caleyo et al., 2009; Taleb-berrouane et al., 2018; Wolodko et al., 2018). Papavinasam et al. (2010) experimentally analyzed effects of physical pipeline parameters and fluid characteristics on internal pitting corrosion. The researchers identified the pipe diameter, thickness, inclination angle, production rate, partial pressure, and concentration of bicarbonates, sulphate, and chloride as pit corrosion contributing factors. Pots et al. (2002) proposed a quantitative methodology to assess critical parameters contributing to the corrosion rate in a microbial infested environment. Six factors, such as water presence, pH, salinity, water wetting, dissolved solids, and temperature exhibited the key roles. However, the complex nature of the interactions of the biotic and abiotic parameters poses a challenge in the application of this model. Further models that correlate the operating parameters, metallurgical properties, pit formation, and its propagation over time are found in the open sources (Caleyo et al., 2009; Mahmoodian and Li, 2018; Ossai et al., 2015).
The stochastic nature of MIC pit formation and growth requires dynamic models, such as the Markov, Poisson, Petri nets, and Bayesian network approaches for pit depth distribution prediction. For example, Hong (1999) used the combined inhomogeneous Poisson and Markov strategies to model pit generation and its depth growth. It was found that the point of surface wetting and coating breakdown play key roles in pit generation; the Kolmogorov forward equation through a time transformation condensation method was used. Similarly, other researchers (Ossai et al., 2016a; Valor and Caleyo, 2007; Valor et al., 2013) proposed a nonhomogeneous and continuous-time linear growth pure birth Markov procedure for pit depth distribution. They made efforts to predict pit growth characteristics and the corroded pipeline failure time under the influence of the pipeline's operating parameters.
Although the probabilistic models provide a better representation of the randomness in corrosion pit nucleation and growth, compared to deterministic tools (Kaduková et al., 2014; Shabarchin and Tesfamariam, 2016; Suarez and Polomka, 2018; Bazán and Beck, 2013), the reviewed probabilistic models do not consider the microbial influence on corrosion rate and failure probability. Some of them are also empirically formulated with multiple inspection data fitted for a comparative framework that have limitations due to sparse data availability and an associated high degree of uncertainties.
Generally, MIC significantly contributes to the failures of offshore systems, and the associated risks. Pipeline deterioration progressively increases the risk of failure over time. The safety of a pipeline is dependent on the management of the remaining useful life and the reliability estimation, which determine the intervention measures over time (Hasan et al., 2012; Mishra et al., 2019; Nizamani et al., 2016; Ossai et al., 2015). These can only be forecasted if the failure probability of the defective pipeline is known. The recent improvements in inspection techniques such as acoustic emission, guided waves ultrasonic testing (GWUT), visual imaging and photography, autonomous underwater vehicles (AUVs), remotely operating vehicles (ROVs), and marine inspection robotic assistant (MIRA) systems have enhanced the capacity for marine/offshore assets integrity assessment for failure-based prediction (Abbas and Shafiee, 2020; Ahmed et al., 2015; Carellan et al., 2014; Giurgiutiu et al., 2015).
Despite considerable attempts to understand, predict, and manage MIC in the oil and gas industry, one critical aspect of offshore systems integrity management under MIC that remains unsolved is how to dynamically predict the MIC rate, failure probability, and future MIC pit depth distribution of a corroding offshore system from single inspection data and by monitoring operating parameters. The existing models are not adequate for the precise prediction of the corrosion rate and failure probability in a dynamic and complex microbial infested environment. There are a limited number of dynamic quantitative models to evaluate the MIC rate and failure probability, considering dynamic nonlinear interdependency among contributory factors.
The main objective of this research is to develop an integrated BN-Markov process model for predicting the MIC rate, failure probability, and future pit depth distribution under microbiologically influenced internal corrosion and its effects on offshore system structural integrity. The MIC influential factors are represented using BN to capture their dynamics, nonlinear dependency, and interdependency. The system failure characteristics based on the critical pit depth state and the future MIC pit depth distribution are estimated using a Markovian process for an offshore system.
The remaining of the paper is structured as follows: Section 2 presents the failure assessment due to MIC. Section 3 briefly describes the proposed research methodology. Section 4 includes and illustrates the application of the methodology using a case study. Section 5 provides the research results and discussion, and Section 6 highlights the most important findings of this study.
Section snippets
Failure assessment due to microbiologically influenced corrosion
Several probabilistic approaches have demonstrated high potential for assessing the failure of offshore systems with corrosion defects, especially pitting corrosion (Ossai et al., 2016a; Valor and Caleyo, 2007; Valor et al., 2013). The integrated BN-Markov process provides a better multi-dimensional dependency modeling capability for the prediction of the MIC rate and corroding offshore systems failure probability. The main elements of the integrated model are briefly illustrated in the
BN-Markov method for MIC rate and failure assessment
This section presents the developmental stages of an integrated BN-Markov methodology for prediction of MIC rate, failure probability, and future pit wastage propagation of a corroding offshore system under the microbial influence, as shown in Fig. 1. The modeling approach begins with assessing the contributory factors to microbial growth and their probabilities, microbial counts, and sets of inspection data, followed by the use of the BN model to forecast the MIC rate and then integrate it
Application of methodology: case study
The proposed methodology is demonstrated with a case study using inspection data and operating parameters (Eckert, 2003). Based on the established cause-effect interaction among monitoring operating parameters and the MIC defect rate, the limited available data is used as the mean value for the analysis. It is further used to simulate data set for a period of 5 years. Cases of offshore subsea pipeline failure due to corrosion defects have been reported in the literature (Liu et al., 2016; Mohd
Results and discussion
The primary objective of developing the integrated BN-Markov model is to precisely predict the MIC rate from a single set of inspection data and the operational parameters. The MIC rate is used as the transition intensity for determination of the probability of failure and the critical failure year of an internally corroded subsea pipeline. To model the MIC rate, the BN is built, connecting the intermediate and basic events by arcs, designating the dependency and interdependency among the
Conclusions
The present study demonstrates the application of an integrated BN-Markov methodology for the prediction of a time-dependent MIC rate, failure probability, and critical failure year of the corroding subsea pipeline. More emphasis is placed on the effects of the operating parameters and SRB on the MIC rate; this also covers the effect of MIC rate on the likelihood of failure of the pipeline after long-term exposure.
The developed model is tested using three corroded subsea pipeline segments; the
CRediT authorship contribution statement
Sidum Adumene: Conceptualization, Methodology, Software, Investigation, Writing - original draft, Writing - review & editing. Sunday Adedigba: Methodology, Validation, Formal analysis, Writing - review & editing, Supervision. Faisal Khan: Conceptualization, Methodology, Formal analysis, Writing - review & editing, Supervision, Project administration, Funding acquisition. Sohrab Zendehboudi: Methodology, Validation, Formal analysis, Writing - review & editing, Supervision.
Declaration of competing interest
We (authors of this manuscript) wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.
We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by
Acknowledgments
The authors acknowledge the financial support provided by Genome Canada and their supporting partners, the Canada Research Chair (CRC) Tier I Program in Offshore Safety and Risk Engineering, and the Natural Sciences and Engineering Research Council of Canada (NSERC).
References (77)
- et al.
An overview of maintenance management strategies for corroded steel structures in extreme marine environments
Mar. Struct.
(2020) - et al.
Dynamic safety analysis of process systems using nonlinear and non-sequential accident model
Chem. Eng. Res. Des.
(2016) - et al.
Process accident model considering dependency among contributory factors
Process Saf. Environ. Protect.
(2016) - et al.
Risk assessment of offshore crude oil pipeline failure
J. Loss Prev. Process. Ind.
(2015) - et al.
A novel modeling approach to optimize oxygen–steam ratios in coal gasification process
Fuel
(2015) - et al.
A Markovian approach to power generation capacity assessment of floating wave energy converters
Renew. Energy
(2020) - et al.
Stochastic process corrosion growth models for pipeline reliability
Corrosion Sci.
(2013) - et al.
Markov chain modelling of pitting corrosion in underground pipelines
Corrosion Sci.
(2009) - et al.
Rigorous models to optimise stripping gas rate in natural gas dehydration units
Fuel
(2015) - et al.
Probability assessment of burst limit state due to internal corrosion
Int. J. Pres. Ves. Pip.
(2012)
Human error probability assessment during maintenance activities of marine systems
Safety and Health at Work
New tools predict monoethylene glycol injection rate for natural gas hydrate inhibition
J. Loss Prev. Process. Ind.
Quantitative risk analysis of offshore drilling operations: a Bayesian approach
Saf. Sci.
Mechanistic aspects of microbiologically influenced corrosion of X52 pipeline steel in a thin layer of soil solution containing sulphate-reducing bacteria under various gassing conditions
Corrosion Sci.
Failure analysis of oil tubes containing corrosion defects based on finite element method
International Journal of Electrochemical Science
Failure assessment and safe life prediction of corroded oil and gas pipelines
J. Petrol. Sci. Eng.
Mechanistic microbiologically influenced corrosion modeling — a review
Corrosion Sci.
The critical involvement of anaerobic bacterial activity in modelling the corrosion behaviour of mild steel in marine environments
Electrochim. Acta
Reliability-based lifecycle management for corroding pipelines
Struct. Saf.
Markov chain modelling for time evolution of internal pitting corrosion distribution of oil and gas pipelines
Eng. Fail. Anal.
Application of Markov modelling and Monte Carlo simulation technique in failure probability estimation — a consideration of corrosion defects of internally corroded pipelines
Eng. Fail. Anal.
Advanced method for the development of an empirical model to predict time-dependent corrosion wastage
Corrosion Sci.
Internal corrosion hazard assessment of oil & gas pipelines using Bayesian belief network model
J. Loss Prev. Process. Ind.
Stochastic modeling of pitting corrosion : a new model for initiation and growth of multiple corrosion pits
Corrosion Sci.
Applications of hybrid models in chemical, petroleum, and energy systems: a systematic review
Appl. Energy
Operational safety assessment of offshore pipeline with multiple MIC defects
Comput. Chem. Eng.
Design and control of MIRA: a lightweight climbing robot for ship inspection
Int. Lett. Chem. Phys. Astron.
Comprehensive modelling of the pitting biocorrosion of steel
Can. J. Chem. Eng.
Failure of crude oil pipeline due to microbiologically induced corrosion
Corrosion Eng. Sci. Technol.
Recent advances in the study of biocorrosion - an overview
Rev. Microbiol.
Pitting degradation modeling of ocean steel structures using bayesian network
J. Offshore Mech. Arctic Eng.
Characterization of Ultrasonic Wave Propagation in the Application of Prevention of Fouling on a Ship's Hull
Advanced Marine Structures. Advanced Marine Structures
Offshore Structural Engineering: Reliability and Risk Assessment
Ocean Structures: Construction, Materials and Operations. Ocean Structures
A Mechanistic and a Probabilistic Model for Predicting and Anlyzing Microbiologically Influenced Corrosion. Master Thesis
Predictive model for CO2 corrosion engineering in wet natural gas pipelines
Corrosion
Corroded Pipelines - Dnv-Rp-F101
Cited by (30)
A risk-based maintenance decision model for subsea pipeline considering pitting corrosion growth
2024, Process Safety and Environmental ProtectionQuantitative risk assessment of static equipment in petroleum and natural gas processing station based on corrosion-thinning failure degree
2023, Process Safety and Environmental ProtectionPseudomonas xiamenensis in the cutting fluids on corrosion behavior of aluminum alloy 2219
2023, BioelectrochemistryA dynamic model for microbiologically influenced corrosion (MIC) integrity risk management of subsea pipelines
2023, Ocean EngineeringCitation Excerpt :The mechanistic model of MIC is drawn with consideration of (i) environmental conditions, including salinity, CO2 partial pressure, pH, O2, temperature, water cut, and Sulphides, (ii) operational conditions including fluid velocity and pressure, (iii) material conditions including steel composition and Carbon content, (iv) biofilm, and (v) exposure duration. The data presented in Table 1 are based on literature (Adumene et al., 2021a, 2020a; Yazdi et al., 2021a), and mean values of continuous variables are based on operational and chemical analysis from an offshore facility off the East Coast of Canada. A chemical analysis of produced water shows the total number of microorganisms using the qPCR (quantitative Polymerase Chain Reaction) method (3.19 E+06 (16 S copies/mL sample)), different types of microorganisms percentage (Methermicoccus 4.79%, Desulfovibrio 8.72%, Dethiosulfatibacter 10.14%, Archaeoglobus 38.36%, Thermoanaerobacter 1.35%, Thermococcus 9.9%, Thermosipho 3.6%, Methermicoccus 4.79%), and sulfate reduction rate (SRR) (0.52 in situ).
Risk-based and predictive maintenance planning of engineering infrastructure: Existing quantitative techniques and future directions
2022, Process Safety and Environmental Protection