Pipeline Systems in Floating Offshore Production Systems: Hydrodynamics, Corrosion, Design and Maintenance

Abstract

Floating offshore production systems play a critical role in offshore resource development, where the structural integrity and operational safety of risers, umbilical cables, and mooring cables are of paramount importance. Focusing on the failure risks of these key components under harsh marine environments, this paper systematically reviews the coupled mechanisms of wave-induced loading, electrochemical corrosion, and material fatigue. Unlike traditional reviews on offshore pipelines and cables, this study not only examines the mechanical performance of deepwater pipelines and cables along with representative research cases but also discusses corrosion mechanisms in marine environments and corresponding repair and mitigation strategies. In addition, recent advances in machine learning-based digital twin frameworks and real-time monitoring technologies are reviewed, with an analysis of representative application cases. The findings indicate that interdisciplinary material innovations combined with data-driven predictive models are essential for addressing maintenance challenges under extreme ocean conditions. Furthermore, this review identifies existing research gaps in data fusion for monitoring technologies and outlines clear directions for the intelligent operation and maintenance of future deep-sea infrastructure.

1. Introduction

Fossil fuels are essential to human production and daily life [1,2]. While several renewable energy technologies have begun to emerge in recent years [3,4,5], they cannot fully replace fossil fuels in the short term [6]. The oceans and seas, as vast reservoirs of resources, are particularly rich in fossil fuel reserves [7,8,9]. Floating Offshore Production Systems play a critical role in the extraction of marine resources [10,11]. Marine pipelines and cables within these floating production systems are essential for ensuring their reliable and efficient operation [12,13,14]. There are three main types of marine pipelines and cables that play a critical role in floating offshore production systems: mooring cables, umbilical cables, and risers. Mooring cables are used to anchor the floating platform to the seabed, ensuring its stability and maintaining its position [15,16]. Umbilical cables connect the platform to subsea facilities, transmitting power, communications, and other essential services [17,18]. Risers are responsible for transporting oil, gas, or other resources from the seabed to the platform for processing or production operations [19,20]. The complex environment in which floating offshore production platforms operate presents several threats to the health of marine pipelines and cables within these systems. These threats include the impact of waves and currents [21,22], collisions with other offshore structures [23,24], environmental corrosion [25,26], variations in temperature and pressure [27,28], transporting pipeline and cables corrosion [29,30], and biological erosion [31,32]. If not adequately addressed, these factors can lead to environmental pollution [33,34], significant economic losses [35,36], and even pose risks to the lives of operators [37,38]. Figure 1 shows an offshore floating production system and common health and safety issues related to pipelines and cables in the system. Consequently, it is crucial to study the negative impacts of these threats on offshore pipelines and explore effective strategies for their placement and protection.

An initial analysis of health and safety issues related to marine pipelines highlights the need for these concerns to be addressed through interdisciplinary research, incorporating insights from multiple fields. The causes of damage to submarine pipelines can be divided into the following broad categories.

(1)

Deformation or breakage of pipelines due to sea currents during laying.

(2)

In the process of transportation due to the influence of third-party external forces leading to damage.

(3)

For submarine pipelines, the occurrence of pipe spans is inevitable during operation. Submarine pipelines can develop hanging spans due to current scouring. The hanging pipe sections are subjected to alternating loads from water flow, leading to vortex-induced vibrations and fatigue fractures, ultimately resulting in pipeline fatigue damage [39,40,41,42].

(4)

Submarine pipelines are susceptible to corrosion due to the acidic composition of the ocean, inherent pipeline defects, and the presence of sand in the transmission substances, leading to perforation and fracture.

Marine pipelines and cables serve as the critical lifelines for offshore energy transmission, yet they are increasingly exposed to multi-hazard environments, including extreme hydrodynamic loads, complex deep-sea corrosion, and mechanical failures. Despite extensive localized research, the field currently lacks a holistic, interdisciplinary framework that integrates fluid mechanics, material science, and structural integrity to guide the entire lifecycle of these assets. This paper mainly reviews three major types of marine pipelines and cables used in offshore floating production systems. This review aims to address this critical gap by developing a comprehensive research manual that fosters cross-disciplinary synergy. Moving beyond isolated studies, this work systematically maps the research trajectory starting from wave-induced effects and extending to advanced material design and anti-corrosion strategies. The scope encompasses an in-depth analysis of environmental impacts, corrosion mechanisms, maintenance protocols, and innovative design methodologies. By identifying current research bottlenecks and emerging trends, this review provides a strategic roadmap for the sustainable development and intelligent management of marine infrastructure.

2. Literature Selection Methods

This study selected the Web of Science Core Collection (WoSCC), indexed in the Science Citation Index Expanded (SCIE), and the China National Knowledge Infrastructure (CNKI) as data sources. The relevant literature was retrieved on 28 January 2025, through the online library of Guangdong Ocean University, China. The publication period was limited to studies published between 1950 and 2025.

To systematically explore multidisciplinary research on the safety of pipelines and cables in floating offshore production systems, a carefully designed search strategy was developed to integrate studies from engineering as well as environmental science and technology. Key structural components, including risers, mooring cables, and umbilical cables, were selected as primary keywords to ensure comprehensive coverage of the core elements of floating production systems. Safety-related terms such as corrosion, failure, and fatigue were incorporated to capture major threats to structural integrity. In addition, environmental factors including waves, currents, and vortex-induced vibration were included to highlight the influence of complex marine conditions on system safety.

The final search query was refined as follows:

(KP = (Riser OR Mooring Cable OR Umbilical Cable) OR TS = (Floating Offshore Production)) AND (TS = (Health OR Safety) OR KP = (Corrosion OR Failure OR Fatigue OR Biological Erosion OR Collisions OR Temperature OR Pressure OR Pipeline Contamination OR Wave OR Current OR Vortex-Induced Vibration)).

The literature selected for this review consists of reviews and research articles, with documents published in both Chinese and English. This is because the authors’ research background is primarily based in Chinese higher education institutions, making them more familiar with studies in both languages. Based on the above search strategy, a total of 657 papers were retrieved from WoSCC and 109 papers from CNKI. First, duplicate records were initially removed using the bibliometric analysis software CiteSpace (Version 6.3R1). Second, through manual screening, papers whose titles involved risers, umbilical cables, and mooring lines were selected. Subsequently, the literature was further filtered based on three criteria: research relevance, citation count, and journal reputation, with journal reputation assessed considering database indexing, impact factor, JCR quartile, and Chinese Academy of Sciences (CAS) ranking. Finally, a total of 260 papers were included. In addition, during the review process and based on recommendations from experts in the field, five additional DNV standards were incorporated. Meanwhile, due to the limited existing research on the performance and protection of pipelines and cables specifically within FPSO systems, this review incorporates high-quality case studies involving risers, umbilical cables, and mooring cables to provide a comprehensive analysis.

3. Wave and Current Impact

The impact of waves and currents are critical considerations in ocean engineering, particularly when studying marine pipeline and cable issues [43,44,45]. During the key operations of pipe-laying [46,47] and pipe-lifting [48,49], waves can compromise operational accuracy and even cause damage. These environmental factors not only directly impact the structural stability and mechanical properties of offshore pipelines and cables but also influence the reliability and safety of FPSO system long-term operations. Periodic loads from waves [50,51], vortex-induced vibrations [52,53], and the pressure and drag forces [54,55] from ocean currents create complex dynamic effects on pipeline and cables systems. These can lead to fatigue damage [56,57] and structural failure [58,59].

3.1. Morison Equation

The Morison Equation [60] is a crucial tool in marine engineering for calculating wave-induced forces on submerged structures, as expressed in Equation (1).

$$dF=C_m\rho dV{\dot{u}}_n+{\displaystyle \frac{1}{2}}{C}_d\rho dA\left|u_n\right|u_n$$

(1)

Here, dF represents the total wave force acting on a micro-section of the member with a projected area dA. The variables ${\dot{u}}_n$ and $u_n$ denote the instantaneous wave-induced fluid acceleration and velocity perpendicular to the object’s axis. The coefficients $C_d$ and $C_m$ correspond to the drag force coefficient and inertia force coefficient, respectively.

The equation incorporates both inertial and drag forces in hydrodynamics to predict fluid forces acting on small-diameter structures. It enables quick estimations of wave and fluid forces on elongated structures with a straightforward calculation process that avoids complex numerical solutions [61,62]. However, its applicability is typically limited to elongated structures whose diameters are much smaller than the wavelength. Recent intensive research has revealed potential extensions of this equation’s use to larger marine structures.

Zhou et al. [63] applied Morrison’s equations to calculate wave loads on large-scale round-ended sinkhole foundations for sea-crossing bridges subjected to regular waves. They compared the results obtained from Morrison’s equations with those from other methods. The findings indicated that Morrison’s equations might overestimate the contribution of wave drag forces, particularly for large-scale structures. The team also analyzed the sensitivity of Morrison’s equations to the relative scale D/L of the structure by varying the wave period and comparing the results with those of alternative methods.

Luo (2023) [64] used Morrison’s equations to calculate the wave forces in both horizontal and vertical directions for a suspension tunnel tube in a wave environment. The study revealed that Morrison’s equation might introduce significant errors when calculating wave loads on large-scale structures. To address this, the hydrodynamic coefficients, including additional mass and inertia coefficients, were optimized and corrected using the least-squares method. This adjustment made the wave forces predicted by Morrison’s equation more consistent with those calculated by the theory of wrap-around potential flow. The accuracy and validity of the modified Morrison’s equation in wave force prediction were verified by comparing its results with those obtained from the bypassing potential flow theory.

For calculating wave loads on large semi-submersible platforms, Liu and Xiao (2003) [65] combined Morrison’s formula with potential flow theory, particularly for the elongated portion of the platform. In this region, the forces are determined by the wave motion around the structure, influenced by the large-volume section of the platform.

Although the Morison equation is widely applied due to its simplicity, it exhibits significant limitations when used for large-scale marine structures. By neglecting wave diffraction and radiation effects, the conventional formulation fails to accurately capture wave–structure interactions when the relative diameter-to-wavelength ratio (D/L) becomes large, leading to distorted force estimations. At present, although coefficient calibration or hybrid approaches incorporating potential flow theory can improve prediction accuracy to some extent, a universally applicable unified criterion is still lacking. Therefore, in engineering design, it is necessary to balance computational efficiency and accuracy according to structural scale and wave conditions, and the establishment of standardized guidelines for model selection and parameter calibration remains a key focus for future research.

3.2. Current Scour

Current-induced scour can result in the removal of sediment around marine pipelines, exposing their surfaces and increasing abrasion and erosion. This can lead to pipeline overhangs, stress concentrations, buckling, or fractures. Additionally, scour can alter seabed morphology, cause uneven stresses and vibrations on the pipeline, reduce stability, and accelerate the corrosion process. Meanwhile, current scour poses a severe threat to the structural integrity and operational safety of FPSO systems by weakening the anchoring capacity of mooring systems, inducing fatigue failure in free-spanning pipelines and risers through vortex-induced vibrations (VIV), and causing instability in subsea infrastructure foundations. Figure 2 illustrates a typical free-spanning offshore pipeline sketch.

Zhang from Shanghai Jiao Tong University has summarized recent research findings on the local scour problem of fixed pipelines in marine environments [66].

Lucassen and Kjeldsen et al. [67,68] conducted experimental research on the current scour problem of submarine pipelines. They analyzed the effects of flow velocity, flume size, and pipeline burial depth on the equilibrium depth of scouring of submarine pipelines. They discussed the equilibrium depth of scouring as a function of flow velocity, flume size, and sand size and proposed a formula to predict the relationship between the equilibrium depth of scouring of pipelines and the diameter of the pipe and the flow velocity.

Chao and Hennesy [69] derived an analytical method for determining the equilibrium depth of local scour in pipelines under uniform flow based on two-dimensional potential flow theory. They proposed an analytical method specifically for clear water scour, which involves solving the following equations:

$${\left[1+{\displaystyle \frac{R}{S}}-{\displaystyle \frac{R^2}{2S^2+RS}}\right]}^2{u}_c^2{\rho }_w{f}_c={\tau }_{cr}$$

(2)

where ${\tau }_{cr}$ is critical shear stress, $u_c^2$ is critical velocity and $f_c$ is friction factor.

Li and others (2012) [70] improved upon the Chao-Hennesy resolution method by incorporating the factor of sediment inflow into the pit. Based on two-dimensional potential flow theory and the principle of energy balance, they combined this with the formula for flow-type and power-type sediment transport rates. This led to the development of a semi-empirical and semi-theoretical prognostic method applicable to determining the equilibrium depth of localized scour around pipelines under homogeneous flow conditions in a moving bed scour situation.

Subsequently, Mao (1987) [71] investigated the interaction mechanism between submarine pipelines and the seabed through indoor experiments. He proposed a mathematical model for the scour depth of pipelines based on the improved potential flow theory, discussed the effect of vortex shedding on scour downstream of the pipeline, and conducted a preliminary analysis of the impact of pipeline vibration on scour.

Sumer et al. (1988) [72] investigated the scour in the wake behind the pipeline using both experimental and numerical methods. They revealed the mechanism of wake scour and discussed the effect of bed shear on wake scour.

Van Beek (1990) [73] investigated the problem of localized scour around a submarine pipeline using numerical methods. The numerical model employed the Navier–Stokes (N-S) equation and the sediment movement equation to calculate fluid movement and sediment scour.

Chiew et al. [74] discussed the mechanism of localized scouring of pipelines in cohesionless seabed’s using experimental methods. They concluded that the pressure difference between the stagnation pressure upstream of the pipeline and the wake pressure downstream is the primary cause of scouring. They suggested that placing impermeable membranes in the seabed below the pipeline could prevent scouring. Additionally, Chiew [75] examined the effect of spoilers on pipeline scour, finding that spoilers installed at the top of the pipeline increase the blocking effect on the water flow, causing more water to be deflected downward and thus increasing the scour depth. The presence of spoilers also enhances wake scour, which is significant for the self-burial phenomenon of pipelines. Furthermore, the local scour of pipelines is strongly influenced by the parameters of both the fluid and the pipeline [76,77,78,79,80].

Hu et al. [81] investigated the effect of the incidence angle of water flow on the scouring of submarine pipelines, based on two-dimensional studies predicting scour depth. They proposed a machine learning approach to predict the equilibrium scour depth of submarine pipelines, considering the incidence angle of water flow. In their study, three models were proposed: GA-BP neural network, RBF neural network, and SVM. They found that the Froude number is the most influential factor affecting the scour depth of submarine pipelines.

Zhao et al. [82] discussed the effects of isolated waves and currents on localized scour. They proposed a coupled numerical model incorporating current and sediment scour modules. The hydrodynamics and local scour around various submarine pipelines, under the combined influence of isolated waves and surface back currents, were investigated using the immersed boundary method.

Dogan et al. [83] conducted an experimental study on the development of scour holes along the spreading direction of the pipe and the velocity of scour propagation. They found that the development along the spreading direction of the pipe was inversely proportional to the Keulegan–Carpenter number and directly proportional to the Shields parameter. Additionally, a new expression based on the Keulegan–Carpenter number was provided to evaluate the clear water scour condition data for this study.

$$V_h^{\ast }=4.5K{C}^{-1.67}$$

(3)

The effective range is 0.01 < $\theta$ < ${\theta }_{cr}$.

They also proposed that the propagation velocity of scour increases with the increase in pipe diameter and decreases with the increase in particle size.

Chen et al. [84] proposed a distributed model to evaluate the damage of submarine pipeline networks caused by regional mudflows. Figure 3 illustrates the framework of the proposed distributed model. The model comprises three components: a spatial discretization module, a regional submarine mudflow dynamics module, and a pipeline network impact force and damage assessment module. This model is designed to assess the response of pipeline networks to regional submarine mudflows, aiding in risk-informed decision-making, planning, and optimization of pipeline routes.

Although significant progress has been made in the study of local scour around submarine pipelines, most existing research is based on simplified assumptions such as two-dimensional steady flow, rigid pipelines, and idealized seabed conditions. These assumptions limit the ability to fully represent the complex marine environment, including wave–current coupling, oblique flow incidence, and seabed heterogeneity, and introduce pronounced scale effects and restrictions on model applicability. Local scour–induced free spans and flow-induced vibrations pose serious threats to pipeline structural integrity; however, the accuracy and general applicability of existing prediction models and mitigation measures remain insufficient under multi-field coupled conditions. Future research should integrate large-scale experimental studies, high-fidelity numerical simulations, and field observations to develop a unified evaluation framework that incorporates structural dynamic response and reliability-based design, thereby establishing more robust engineering design guidelines and mitigation strategies.

3.3. The Vortex-Induced Vibration

Nonlinear hydrodynamic effects induced by wave action can lead to complex kinematic responses, among which vortex-induced vibration (VIV) is a representative phenomenon. VIV significantly compromises FPSO systems by driving high-frequency cyclic oscillations in risers, umbilical cables, and mooring lines, thereby accelerating fatigue damage and increasing the risk of structural failure or loss of containment.

An experimental study on the vortex-induced vibration (VIV) of submarine pipelines with long free spans was conducted. When submarine pipelines and cables are laid on cohesionless soil, uneven seabed’s or scouring can lead to the development of local scour around the pipeline. This local scour results in the periodic shedding of vortices across the pipeline, subjecting it to transverse and downstream periodic fluid forces. These periodic fluid forces cause the pipeline to vibrate periodically, a phenomenon known as “Vortex-Induced Vibration” (VIV) [85,86,87].

Numerous researchers conducted an experimental study on the vortex-excited vibration of low mass ratio pipelines. The experimental results revealed three branches in the variation in amplitude with the approach velocity in these pipelines. The vortex-excited cylindrical amplitude and vibration frequency were found to be related to the approach velocity. As the approach velocity changed, the performance of the three branches—initial, lower, and upper—also changed. Notably, there was a significant hysteresis effect observed before the transitions between these branches [88,89,90,91,92].

Zhang [66] summarized the equations and proposed several forms of vortex shedding for cylindrical vortex-excited vibrations, such as 2S, 2P, P + S, 2C, 2T, etc., as shown in Figure 4.

Fan D et al. [93] conducted an experimental study on the steady-state hydrodynamic vibrations of small flexible cylinders (aspect ratio L/d = 61, mass ratio m0 = 1.38) within a backed structure on the semi-buried seafloor of a main pipeline. The study investigated the effect of the gap-to-diameter ratio for five different cases, including the open-flow case (G/d = ∞) at various folding velocities. The following conclusions were drawn:

(1)

Compared to the narrow-band single-frequency response in the CF (first harmonic) and IL (second harmonic) directions for the open-flow case, the presence of a larger semi-buried pipe increases the mean drag coefficient of the smaller flexible cylinders in the backed structure. Additionally, it exhibits a wider-band response when the gap-to-diameter ratio is small.

(2)

For small gap-to-diameter ratios, this leads to larger average IL displacements at the same rate of reduction.

(3)

At G/d = 1.0, a stronger third harmonic motion is observed in the CF direction. In the IL direction, the frequency response is much richer, exhibiting stronger first and fourth harmonic motions, detectable third harmonic motions, and larger second harmonic motions.

Xu [94] and colleagues investigated the multi-span classification method and interaction mechanism of vortex-excited vibration in submarine pipelines. They studied the interaction mechanism of multi-span pipelines from both static and dynamic perspectives and proposed a multi-span pipeline classification method based on modal frequency and energy transfer. Figure 5 and Figure 6 illustrate the processes of the modal frequency-based classification method and the energy transfer-based classification method, respectively.

Zhang et al. [95] numerically investigated the inward flow effect (IFE) and outward flow effect (ECE) of vortex-excited vibrations (VIVs) on the axial inward flow of free-spanning submarine pipelines with mis-flow vortex-excited vibrations. They explored the competing mechanisms involved. The partial differential equations for the pipeline system were developed using Euler-Bernoulli beam theory and a wake oscillator model.

Although significant progress has been made in the study of vortex-induced vibration (VIV) of submarine pipelines, most research is limited to idealized conditions such as uniform flow, two-dimensional assumptions, and isolated free spans, which cannot fully capture wave–current coupling, seabed irregularities, and complex pipe–soil interactions. Existing predictive models are constrained by parameter sensitivity and scale effects, and often neglect the nonlinear coupling between scour evolution and vibration response, limiting their applicability to full-scale engineering scenarios. VIV-induced fatigue damage and resonance pose serious threats to pipeline integrity, while the long-term effectiveness of current mitigation measures under dynamic seabed conditions remains uncertain. Future research should employ large-scale experiments and high-fidelity fluid–structure–soil simulations to develop an integrated predictive framework encompassing hydrodynamics, structural dynamics, and seabed evolution, thereby establishing more robust engineering design guidelines.

4. Corrosion

The corrosion behavior of marine pipelines is primarily influenced by seawater [96,97] and seabed sediments [98,99]. Seabed sediments exert a significant influence on the integrity of FPSO systems by creating complex electrochemical environments that induce differential aeration cells and promote microbiologically influenced corrosion (MIC). These mechanisms adversely affect the long-term structural reliability of key subsea components, including risers, umbilical cables, and mooring lines, particularly in sediment-covered or partially buried conditions.

The complexity of the seawater environment complicates the corrosion process, which involves electrochemical reactions affected by factors such as the composition of the rust layer and the nature of the solution [100,101,102,103]. To better understand corrosion behavior, it is essential to use relevant corrosion models that can explain these patterns and interactions.

Compared to the seawater environment, seabed sediments present a non-uniform system consisting of solid and liquid phases [104,105,106]. In submarine sediments, steel primarily undergoes electrochemical and microbiological corrosion. The limited dissolved oxygen in seafloor sediments and the poor exchange between seafloor water and ocean water usually result in more oxygen-poor corrosive environments [107,108]. Different types of seafloor sediments affect the rate of oxygen diffusion, leading to uneven oxygen distribution [109,110,111], large potential differences, corrosion currents, and galvanic coupling corrosion. Additionally, variations in oxygen supply can initiate autocatalytic processes of localized corrosion, such as pitting [112,113] and crevice corrosion [114,115]. Figure 7 illustrates the analysis of corrosion in submarine pipelines.

4.1. Corrosion Behavior in Seawater

Figure 8 illustrates the environmental distribution of metals in the oceans. Depending on the environment in which metals are located in the ocean, they can be divided into the marine atmospheric zone, splash zone, tidal zone, submerged zone, and mud zone. The corrosion behavior of metals in the marine environment varies depending on the zone, and thus, marine pipelines are subject to different degrees of corrosion [116,117,118].

In the marine atmospheric and splash zones, frequent wet-dry cycles and high oxygen levels lead to faster corrosion rates, with uniform corrosion and pitting corrosion being common [119,120,121]. The tidal zone, influenced by tidal changes, presents a complex environment where differential aeration corrosion and crevice corrosion are more likely to occur [122,123]. The submerged zone, where metals are constantly submerged in seawater, experiences slower corrosion rates due to limited oxygen supply, but uniform corrosion and biofouling remain significant issues [124,125]. In the mud zone, metals buried in sediment face minimal oxygen exposure, making microbial corrosion, such as sulfide corrosion caused by sulfate-reducing bacteria, the primary concern [126,127].

Melchers R E et al. [128] established a corrosion model for mild steel in seawater by conducting long-term experiments on its early corrosion behavior. Figure 9 illustrates the time model of corrosion loss for multiphase phenomena. They concluded that the long-term immersion of carbon steel in seawater can be divided into five stages:

(1)

The activation control stage:

In the initial stage, during a relatively short period of time when carbon steel is immersed in seawater, metal activation is driven by the ample supply of oxygen.

(2)

The concentration control stage:

Dissolved oxygen diffuses and generates thin and loose corrosion products. Due to the thin rust layer, there is no obstruction to dissolved oxygen diffusion, and the corrosion process can be described by a linear function.

(3)

Oxygen diffusion control:

As the rust layer becomes thicker, the diffusion of dissolved oxygen through the rust layer becomes the controlling step of the corrosion reaction. This stage is characterized by oxygen diffusion control.

(4)

The growth control stage of SRB:

Due to the gradual thickening of the rust layer, it becomes difficult for oxygen molecules to reach the surface of the metal. Consequently, sulfate-reducing bacteria (SRB) start to grow in this stage.

(5)

The stabilization control stage of SRB:

As bacterial growth stabilizes, the corrosion rate of carbon steel also gradually stabilizes. This stage is characterized by a prolonged period of stable corrosion rate.

As early as 1925, Evans et al. [129,130,131] used vertical steel and zinc plates half-immersed in NaCl solution to study the effect of waterline action on metal corrosion. They found that the metal at the gas–liquid junction exhibited slight corrosion, while the corrosion at the waterline was particularly pronounced. This was attributed to different oxygenation states in the waterline region. The narrow, well-oxygenated waterline area became a cathodic zone, while the under-oxygenated area beneath the waterline became an anodic zone. This disparity created a macroscopic oxygen concentration cell, leading to galvanic corrosion of the metal.

Tomashov et al. [132] believe that the cathodic area of the macro-galvanic cell, which is the waterline region, contains multiple micro-galvanic cells. These two types of cells together influence the corrosion behavior of the metal. As the corrosion behavior of metals in the waterline region receives increasing attention, researchers worldwide have conducted extensive studies on the corrosion of metals in seawater and atmospheric environments [133,134,135,136,137,138].

Jeffrey et al. [139] conducted a study on waterline corrosion by immersing a 1-m-long vertical steel plate in seawater. Based on the different lengths of the steel plate immersed in seawater, they obtained a classic mass loss profile. The results showed that the mass loss near the waterline was minimal, indicating relatively mild corrosion. The most severe corrosion occurred in the splash zone and the fully immersed zone below the waterline. The splash zone experienced the greatest mass loss due to the influence of sea wind and waves, while the fully immersed zone below the waterline also showed significant mass loss. The study described the distribution areas of anodes and cathodes, and the electron transfer paths under different aeration conditions.

4.2. Corrosion Rate Prediction

Researchers have developed various corrosion prediction models to better understand the corrosion mechanisms in marine pipelines and enhance their protection [140]. Commonly used prediction methods include probabilistic statistical methods [141,142], reliability function analysis [143,144], regression analysis methods [145,146], time series prediction methods [147,148], gray model prediction methods [149,150], and artificial intelligence techniques [151,152,153].

Traditional prediction methods, such as probabilistic statistics and regression analysis, often require high-quality data and may not accurately capture the relationships between variables when fitting regression functions. Reliability function analysis, frequently employing the Monte Carlo method [154], involves substantial computational effort. Time series prediction methods, including autoregressive models [155], sliding average models [156], ARMA models [157], and ARIMA models [158], aim to predict future outcomes by constructing models based on the corrosion development law. However, time series-based methods face limitations in predicting multiple influencing factors that are not time-dependent. Additionally, artificial intelligence algorithms like the Least Squares Support Vector Machine (LSSVM) model [159] and gray models, such as the GM (1, N) model [160], offer alternative approaches to prediction.

Deng et al. [161] proposed a novel prediction method for the corrosion rate of submarine pipelines, combining gray correlation analysis and fuzzy neural networks (FNN). The FNN integrates fuzzy theory with artificial neural networks, addressing some of the limitations of traditional neural networks [162]. The theoretical foundation of FNN is rooted in fuzzy mathematics and the learning mechanisms of neural networks. In this context, the degree of affiliation refers to how closely an element u belongs to a fuzzy subset f. The fuzzy affiliation function is used to quantitatively calculate this degree of affiliation [163]. The T-S fuzzy system can continuously modify the affiliation function of the fuzzy subset. The experimental steps are shown in Figure 10. The experiment demonstrated that this hybrid prediction method maintains high prediction accuracy even when the number of environmental factors is reduced. Future work can focus on extensive testing and training with large datasets on the corrosion rate of submarine pipelines to enhance the generalization capability of this hybrid prediction method.

Wang et al. [164] conducted a research experiment utilizing a hybrid Bayesian network to predict the corrosion risk of submarine pipelines. The results indicated that the model effectively captures the dynamic nature of corrosion risk in submarine pipelines and offers actionable recommendations for mitigating this risk.

5. Maintenance and Repair of Marine Pipelines

5.1. Buckling Protection

Buckling protection constitutes a critical design consideration for marine pipelines and cables within FPSO systems. Risers and umbilical cables may experience thermally induced compressive loads under high-pressure, high-temperature (HPHT) operating conditions, while mooring lines are susceptible to instability under combined environmental and operational loading. Effective buckling protection and stability control measures are therefore essential to prevent excessive deformation, ensure structural integrity, and maintain long-term service safety. As marine development progresses into deeper waters, submarine pipelines must withstand the complex pressures, temperature fluctuations, and currents of the deep-sea environment. Effective buckling protection ensures the stability and safety of these pipelines through careful design and construction, which includes selecting suitable materials and structures, predicting and assessing buckling phenomena, and designing reinforcement measures such as additional supports and anti-buckling devices. Regular monitoring and maintenance are also critical. The significance of these efforts lies in extending the service life of pipelines, reducing maintenance costs, minimizing the risk of environmental pollution, and ensuring the efficient and safe exploitation of marine resources [165,166,167,168,169]. Figure 11 illustrates five typical buckling shapes for pipelines.

In 1977, Fabian [170] investigated the nonlinear pre-buckling state, bifurcation, and initial post-buckling behavior of an infinitely long cylindrical elastic tube subjected to bending, pressure, and axial loads. He analyzed the collapse behavior by determining the ultimate load and exploring the possibility and significance of axial wrinkling in the compression region of the shell before reaching the ultimate load. However, his study only considered linear buckling mode forms.

In 1984, Kyriakides et al. [171] discussed the activation of propagation buckles in submarine pipelines. If the external pressure is sufficiently high, a propagation buckle can be initiated through a localized depression in the pipe. This buckle will propagate at any pressure higher than the propagation pressure of the pipe. The pressure at which a localized geometric defect transforms into a propagating buckle is called the initiating pressure, which depends on the geometric characteristics of the damage. Kyriakides et al. [171] calculated the buckling initiation pressure and the buckling propagation pressure of a submarine pipeline with a localized depression under external pressure based on the “double bell” mode of a point connection.

In 1990, Huang et al. [172] investigated the post-buckling behavior of offshore pipelines with vortex-shaped geometric defects under the combined effects of axial force and external pressure. The authors employed a combination of dual-scale uptake techniques and Fourier expansion to derive the initial post-buckling coefficients of circular shells with general vortex-shaped defects. These defects exhibited an exponential decay in the circumferential direction and took the form of an arbitrary function in the longitudinal direction. Their study provided insights into the relationship between extreme loads and defect parameters under the combined action of axial force and external pressure. Additionally, the sensitive zone of shell response to vortex defects was identified based on extensive example calculations.

In 2022, Ning et al. [173] proposed an efficient pipe cell model to simulate the bulging and buckling behavior of pipelines based on Euler-Bernoulli beam theory. They accurately considered the continuously distributed pipe-soil interaction (PSI) using the Gauss-Legendre integration method. Nonlinear PSI, thermal expansion forces due to temperature increases, and axial loads due to internal pressure differences were integrated into the pipe element. Continuously distributed zero-length Winkler-type springs were employed to simulate the interaction between the pipe and the surrounding soil, including both vertical and axial PSI. The researchers derived the stiffness matrices for expansion forces caused by temperature rise and internal pressure differences. They designed three sets of validation cases to verify the correctness and effectiveness of the pipeline model, leading to the following conclusions:

(1)

Compared to traditional solid and shell elements, the pipeline unit requires fewer elements to simulate a long pipeline and eliminates the need for distributed springs or solid elements to represent the surrounding soil in traditional finite element analysis. The pipeline unit can describe and synthesize nonlinear pipe-soil interaction (PSI), thermal expansion, temperature effects, and internal pressure effects. Consequently, using pipe cells is more effective for simulating the bulging and flexure of pipelines.

(2)

For pipes with small out-of-straightness (OOS), bulge-flexure behavior may occur when the temperature reaches a critical value. Therefore, when using the Newton-Raphson iterative method, the load increment needs to be reduced to accurately simulate the pipe’s bulge-flexure behavior.

Figure 12 illustrates the turbulent buckling induced by cumulative expansion due to high temperature and high pressure (HTHP) and geometric defects.

5.2. Overhaul and Monitoring of Submarine Pipelines

Overhaul and monitoring strategies for marine pipelines and cables play a vital role in enhancing the operational resilience of FPSO systems. Key components, including risers, umbilical cables, and mooring lines, benefit from a transition from reactive to predictive maintenance enabled by condition monitoring and integrity management, thereby minimizing the risks of production downtime and environmental disasters associated with structural failures.

Wang [174] studied the current status of corrosion monitoring technology research in submarine pipelines. The monitoring methods for marine pipelines are mainly divided into full coverage monitoring and local corrosion monitoring.

The full coverage monitoring methods for submarine pipelines mainly include ultrasonic internal detection technology [175,176], vortex current inner inspection technology [177,178], magnetic leakage inner detection technology [179,180], and Magnetic Tomography Method (MTM) [181,182]. Although the implementation costs of these methods are high, they provide comprehensive, intuitive, and reliable data that cover the entire submarine pipeline, offering significant reference value.

Localized corrosion monitoring has become a crucial aspect of assessing the condition of sea pipelines. Key methods include field signature method (FSM) [183,184], fixed-point thickness monitoring [185,186], the inductive probe method [187,188], the hanging piece method [189,190], the linear polarization method [191,192], the electrochemical noise method [193,194], the electrochemical impedance method [195,196], and the resistance probe method [197,198]. These techniques enable more frequent monitoring while maintaining relatively low costs, with some allowing for continuous data collection.

Employing a combination of these localized monitoring methods enhances the comprehensiveness of the data gathered, providing insights such as the uniform corrosion rate, pitting corrosion rate, types of corrosion, and information on corrosion products. This multi-method approach ensures a richer and more detailed understanding of corrosion processes affecting sea pipelines.

Li et al. [199] conducted research on third-party damage monitoring technology for submarine pipelines based on ship dynamic tracking. By dynamically monitoring and identifying ship behaviors, this technology can effectively determine whether submarine pipelines are subjected to damage from ships, thereby enabling risk monitoring and early warning of third-party damage. However, due to various technical challenges, the monitoring and early warning of submarine pipeline damage caused by external forces from ships remain in the research and experimental stages.

At present, ship dynamic monitoring technology is primarily implemented using AIS (Automatic Identification System), satellite, radar, and other related technologies [200,201]. Developing early warning technology for third-party damage to submarine pipelines based on ship dynamic tracking will create an anti-damage barrier for these pipelines. This advancement will shift the safety management of submarine pipelines from experience-based methods to digital management, providing a reliable guarantee for their safe operation.

Zhang et al. [202] investigated the dynamic characteristics of pipeline inspection gauges (PIGs) under the excitation of submarine pipeline girth welds. Internal inspection technology for submarine pipelines forms the foundation for comprehensive inspection and evaluation of pipeline integrity. Figure 13 illustrates the PIG model. PIGs perform online detection of corrosion, deformation, and cracks within pipelines. The researchers proposed a dynamic model of the PIG sealing disk based on Kelvin Spring damping and a nonlinear dynamic model of the PIG system. Additionally, they used MSC/Adams software (version 2021.0.1) to simulate the dynamic response of the PIG when colliding with a circumferential weld at different velocities and center of mass positions.

Feng et al. [203] proposed an automatic cable tracking method (ACTM) based on side-scan sonar (SSS) to equip autonomous underwater vehicles (AUVs) with autonomous decision-making capabilities for the rapid localization and stable tracking of submarine cables under significant uncertainties. They designed a behavior-based planning method to enable AUVs to make online decisions such as cable search, tracking, and recapture after tracking loss, facilitating autonomous surveys of flexible cables in uncertain seafloor environments.

When a cable is detected, the tracking task is innovatively approached as a path tracking problem in the horizontal plane. The virtual mass extracted from the cable is modeled using a uniform motion model based on target tracking theory, and the state of the cable is estimated using a Kalman filter. The control objective is established based on the relative geometrical relationship between the underwater robot and the cable to ensure optimal imaging. Multiple cable states are incorporated into the guidance law to enhance tracking performance. The method’s effectiveness was validated through numerical simulations and field experiments.

Zhang et al. [204] proposed using forward-looking sonar to obtain the precise position of submarine pipelines and a tracking algorithm to maintain the tracking of these pipelines. This approach addresses the issue of positional deviation caused by the accumulation of errors in GPS signals within the vehicle’s combined navigation system during extended operations, which can result in significant inaccuracies when real-time measurements are made by an autonomous underwater vehicle (AUV) using a multi-beam sonar system.

Conventional tracking algorithms often fail to detect curved pipes, losing the tracking target at certain curved sections. To address this, they propose an algorithm to identify and fit curved pipes separately. The algorithm employs an optimized Rosenfeld algorithm to efficiently extract the center line of the pipeline in a binary image and uses straight-line fitting and power function fitting to obtain more accurate paths for straight and curved submarine pipelines, respectively.

A realistic tracking scenario of the submarine pipeline in Qingdao Bay was simulated. Numerous experiments concluded that the target point position of this method is easy to calculate and can be continuously tracked, demonstrating the effectiveness of the submarine pipeline tracking technology. Figure 14 shows the pipeline tracking system and Figure 15 shows the flow chart of the subsea pipeline tracking algorithm.

6. Marine Pipeline Design

6.1. Risk Analysis of Marine Pipelines and Cables

Submarine pipelines are essential systems in the process industry, designed for long-term use in marine environments. However, these harsh environments often pose significant threats to the condition of submarine pipelines [205]. For instance, these pipelines are vulnerable to scouring by currents and waves, third-party damage, earthquakes, and design flaws, all contributing to their high likelihood of failure [206]. Therefore, risk assessment of submarine pipelines is crucial due to the potentially devastating impacts of leakage failures on human safety, the environment, and the economy [207]. Within FPSO systems, marine pipelines and cables—comprising risers, umbilical cables, and mooring lines—play distinct yet interdependent roles in hydrocarbon production and station keeping. Risk analysis provides critical decision support by identifying and evaluating multi-dimensional threats such as environmental loads, structural failures, and accidental events across these components. While the subsequent analysis focuses primarily on leakage failure in submarine pipelines due to its high likelihood and severe consequences, the adopted risk assessment framework is applicable to other FPSO components with appropriate adaptation.

Yu et al. [208] aimed to obtain reliable brittleness values through expert assessment in the absence of historical failure data and to perform probabilistic risk assessment of submarine pipeline leakage failure. They proposed a probabilistic Petri net (PPN) method based on intuitive fuzzy evidential reasoning, which quantitatively evaluates the probability of submarine pipeline leakage failure by determining the failure probability of the basic event through the experience of domain experts. In the same year, Yu et al. [205] proposed an improved Failure Modes and Effects Analysis (FMEA) method based on cloud modeling and the extended risk assessment methodology (VIKOR) to enhance the effectiveness of submarine pipeline failure risk assessment. This FMEA method addresses the uncertainty and ambiguity of risk assessment information while leveraging the advantages of the extended VIKOR method in solving complex risk analysis problems. Additionally, an extended two-level submarine pipeline failure risk factor hierarchy was established to improve the comprehensiveness of risk assessment. The improved integrated dynamic weighting algorithm considers the weights of different FMEA team members, making the results more reliable and valid.

Failure Modes and Effects Analysis (FMEA) is a proactive risk management tool designed to identify and evaluate the causes and effects of potential failure modes and implement necessary measures to prevent critical failures. Despite being a widely recognized method for analyzing and preventing failure modes in engineering, FMEA has not been extensively studied in the context of submarine pipeline failure risk assessment [209,210].

Among the existing studies, Li et al. [211] used the Decision-Making Trial and Evaluation Laboratory (DEMATEL) method to analyze the interdependence between failure factors of submarine pipelines and combined it with a fuzzy inference algorithm to calculate the risk levels of these factors. To improve the reliability of submarine pipelines.

Current FMEA methods applied to submarine pipeline risk assessment are essentially based on Multi-Criteria Decision Making (MCDM) techniques, which are also prevalent in other fields [212,213,214,215]. MCDM is a well-known branch of operations research and is considered a valuable tool for improving FMEA performance and addressing the limitations of the traditional Risk Priority Number (RPN) approach [216]. MCDM has been widely used to solve selection problems involving multiple conflicting evaluation criteria [217].

Yu et al. [218] extended the TODIM method to submarine pipeline risk analysis, highlighting the need for improvements to ensure its accuracy and applicability. Additionally, for multilevel index systems, the fuzzy comprehensive evaluation method has been utilized for analysis.

6.2. Simulation and Analysis Methods

Simulation and analysis methods provide essential support for FPSO systems by modeling the structural responses of marine pipelines and cables, including risers, umbilical cables, and mooring lines, under complex operational conditions such as scour, vortex-induced vibration (VIV), and thermal buckling. Although specific mechanisms may dominate different components, these methods enable comprehensive prediction of structural behavior, providing essential guidance for safe design and reliable operation of FPSO systems.

Hanging submarine pipelines are a common configuration in the transport of paint and other materials [219]. To predict the mechanical properties and ensure the structural integrity of these hanging pipelines, it is essential to model and analyze them accurately. Wang et al. [220] reviewed various methods for modeling and analyzing hanging submarine pipelines. Continuous modeling is typically employed for static morphology analysis of hanging pipelines. In contrast, discrete models can be used for both static and dynamic analysis of complex deep-water hanging pipelines, addressing geometric non-linearity, material non-linearity, boundary non-linearity, and other intricate issues. Figure 16 shows the classification of trailer line modeling methods.

For the static analysis of hanging submarine pipelines, the natural catenary theory method is often chosen due to its theoretical simplicity and ease of computation. However, when bending stiffness or higher computational accuracy is required, finite element methods such as the rigid catenary method and the nonlinear beam method can be utilized. These methods offer more precise analysis by accounting for factors like bending stiffness and other complex behaviors.

The natural catenary theory method has a straightforward theoretical structure, making it an excellent starting point for more refined solution methods. However, its calculation results in shallow water are not sufficiently accurate due to several factors: it neglects the bending stiffness of the pipeline, it assumes a discontinuity of the bending moment at the point of contact, and it fails to account for dynamic loads on the pipeline. Additionally, the model cannot solve for hydrodynamic forces and cable characteristic parameters (e.g., mass, elasticity) that vary along the cable. The natural catenary chain line morphology equation [204] for horizontal force H is:

$$z={\displaystyle \frac{H}{\omega }}\left[\cos \left({\displaystyle \frac{\omega x}{H}}\right)h-1\right]$$

(4)

In this catenary equation, $\omega$ denotes the unit weight (load per meter), while H signifies the constant horizontal tension throughout the span.

In 1967, Plunkett [221] enhanced the existing natural catenary theory by considering the bending stiffness of the pipeline and proposed the rigid catenary method. This approach addressed some of the shortcomings of the natural catenary theory method, yet it still could not analyze the dynamic response of the pipeline. The nonlinear beam theory, which discretizes the pipeline into beam unit micro-segments, neglects the influence of pipeline cross-section deformation and only considers pipeline bending deformation and axial tensile pressure. Compared to the catenary chain line concept, the nonlinear beam theory excels in handling complex boundary conditions and environmental loads. Furthermore, this modeling method is well-suited for numerical calculations using computer language, making pipeline correlation analysis more convenient.

The dynamic response analysis of hanging line pipelines often involves nonlinear elements. Currently, only various finite element methods combined with numerical solutions can effectively analyze the dynamic response of hanging submarine pipelines. Among the finite element methods, the nonlinear beam theory model and the Cosserat rod theory model exhibit higher accuracy. In contrast, the centralized mass method and the multi-rigid body pipeline model offer higher computational efficiency, with the advantage becoming more pronounced with increasing water depth. Different numerical algorithms are selected based on the specific simplification results of the model equations of motion.

de Lucena et al. [222] conducted a study on the optimized design of submarine pipeline routes using genetic algorithms, focusing on different constraint handling techniques. The study specialized in incorporating more relevant design criteria, categorized as “soft” and “hard” based on the consequences of their violation. It evaluated more efficient techniques to address constraints related to the “hard” criteria. The researchers found that classical static penalty methods were insufficient, especially in more complex cases. Therefore, they emphasized the importance of associating these criteria with more advanced constraint handling techniques, such as Adaptive Penalty Method (APM), e-constraints, and the Ho-Shimizu method. These advanced techniques demonstrated better performance in selecting the best feasible routes and required fewer evaluations.

7. Marine Pipeline Materials

The selection of marine pipeline materials fundamentally dictates the corrosion resistance, mechanical strength, and thermal stability of the system, serving as the primary defense against structural failure and environmental leakage while ensuring the durability and fatigue life of FPSO assets in extreme deepwater environments. Pipeline steel for submarine applications can be divided into three categories: cold [223,224], high sulfur areas [225,226], and submarine laying [227,228]. A comprehensive evaluation of pipeline laying conditions and the main forms and causes of failure reveals that pipeline steel must possess excellent mechanical properties. It should also have a large caliber, weldability, resistance to cold and low temperatures, corrosion resistance, and resistance to seawater and hydrogen-induced cracking (HIC), among other properties. The working environment for submarine pipelines is harsh, with long lines that are difficult to maintain, making stringent quality requirements for the steel essential.

The composition design of submarine pipeline steel aims to minimize the content of impurity elements such as phosphorus (P), sulfur (S), oxygen (O), nitrogen (N), and hydrogen (H). During the steelmaking process, precise control of the composition is required to meet design specifications. Additionally, efforts must be made to remove non-metallic inclusions from the steel, thereby improving its purity.

7.1. Materials Development

The growing energy demand is driving the development of marine oil and gas resources, with submarine pipelines playing an increasingly crucial role as the arteries for the collection and transportation of sea-based oil and gas. Submarine pipelines operate long-term in low-temperature, high-pressure, and highly corrosive marine environments. They must withstand internal and external pressure, axial force, bending moments, and other static loads, as well as the combined effects of temperature loads. Additionally, they endure alternating external pressures, waves, currents, and other dynamic loads, which subject the pipeline to various forms of stress and potential damage.

Given the extremely high safety requirements of submarine pipeline projects, the specifications for the size, shape, surface quality, and physical and chemical properties of thick-walled submarine pipeline steel are very stringent. These demanding conditions necessitate meticulous design and manufacturing processes to ensure the pipeline’s reliability and integrity in its challenging operating environment [229].

Therefore, the development of thick-walled submarine pipeline steel plates and steel pipes is particularly challenging, mainly reflected in the following points:

(1)

The substantial thickness of steel plates and pipes makes it challenging to achieve adequate low-temperature fracture toughness. Meeting the technical conditions [230] poses high demands, marking a significant milestone globally. Enhancing the low-temperature fracture toughness of thick-walled pipeline steel remains a key technical challenge in the development of pipeline steel worldwide.

(2)

Achieving high strength and high toughness in steel plates and pipes, while also ensuring excellent deformation resistance, presents a significant challenge. The precise matching of these properties is particularly difficult. The stability of performance requirements is extremely high, permitting only a small range of fluctuations. Maintaining consistent product quality on a large scale further complicates the process.

(3)

The wall thickness of the steel pipe is large, and the diameter-to-thickness ratio (D/t) is small, so it is difficult to make the pipe.

(4)

For safety reasons, the size, shape and surface quality of steel pipe are strictly required, and it is difficult to control the mass production. In industrial production, it is very difficult to meet the strict requirements of thick-walled submarine pipeline steel pipe and realize stable, mass production and control [231,232,233,234,235].

In contrast to land-based pipelines, submarine pipelines must meet more stringent requirements due to the harsh marine environment. These include higher demands on the strength, toughness, compressive performance, and dimensional accuracy of both the steel and the steel pipes used. According to the DNV-ST-F101 standard for submarine pipelines, there are rigorous requirements not only for the transverse and longitudinal strength of the steel pipe but also for its resistance to low-temperature dynamic tearing. As offshore oil exploitation advances from shallow to deep waters, deep-sea pipelines impose even stricter demands on pipeline materials, welding, construction, and maintenance. Enhancing the strength, toughness, compressive properties, corrosion resistance, and dimensional accuracy of steel pipes is crucial to improving the safety of pipeline installation and operation [236,237].

Zhang et al. [238] have developed and studied X70 and X65 thick-walled submarine pipelines for the Li wan project in the South China Sea. These pipelines must meet the yield strength requirements specified by DNV offshore pipeline technical standards [239]. The standard allows for a yield strength fluctuation range of 120 MPa, which is more than 20% smaller than the permissible fluctuation range for conventional land pipelines (150 MPa). Additionally, the transverse and longitudinal strength fluctuations of the same steel pipe must not exceed 40 MPa.

The main challenges in developing pipeline steel for Li wan, South China Sea, included the use of “a small amount of ferrite + bainite” in the needle-like ferrite microstructure design. To ensure good welding performance, low carbon and low carbon equivalent alloys were employed. Clean steel smelting and continuous casting technologies were used to meet high toughness requirements, while two-stage controlled rolling technology was applied to ensure the desired properties of the pipeline steel.

Lang et al. [240] investigated the HE sensitivity of a new generation of X70 grade acid-resistant submarine pipeline steel treated with varying Mg contents, based on titanium and aluminum deoxidation rolling. The study used constant load and slow strain rate tensile (SSRT) methods to assess the hydrogen embrittlement (HE) sensitivity.

Table 1. Chemical compositions of tested steels %.

Steel	C	Si	P	S	Mn + Cr + Ni + Mo	Nb + V	Ti + Ai	Mg
Mg0	0.044	0.26	≤0.05	≤0.003	≤1.80	≤0.10	≤0.03	0
Mg3	0.041	0.25	≤0.05	≤0.003	≤1.80	≤0.10	≤0.03	0.003
Mg5	0.043	0.25	≤0.05	≤0.003	≤1.80	≤0.10	≤0.03	0.005

illustrates the chemical composition of the tested steels.

With a 0.003% Mg treatment, the morphology of Al₂O₃ and MnS inclusions improved, resulting in finer, more diffuse inclusions with a “core–shell” structure. However, a 0.005% Mg treatment increased the number of larger inclusions and led to the formation of chain-like inclusions. The “core–shell” structure from the Mg treatment exhibited higher dehydrogenation activation energy (greater than 100 kJ/mol), making it more difficult for hydrogen to desorb from the inclusion-substrate interface. This increased hydrogen trapping efficiency in the steel. Conversely, further increasing the Mg content from 0.003% to 0.005% reduced the hydrogen trapping efficiency in the tested steels.

7.2. Material Applications

The exploration and development of marine oil and gas resources have led to a growing demand for submarine pipeline construction. The industry for submarine pipeline materials has promising development prospects, particularly in the need for new materials and technologies. With advancements in marine pipeline laying technology, the depth of submarine pipeline installations has progressed from shallow waters of a few dozen meters to more than 3000 m in the deep sea [241,242,243,244].

The development of marine pipeline laying technology poses significant challenges for the industry. Notably, the Gulf of Mexico features a famous pipeline, 2412 m deep and 222 km long, using X65 grade steel, with an outer diameter of 610 mm and a wall thickness ranging from 24.1 to 34.3 mm, completed in 2014. Another record-setting pipeline in the Mediterranean Sea extends from Algeria to Sardinia, reaching a water depth of 2824 m. Currently, the highest steel grade used in foreign submarine pipeline projects in non-acidic environments is X70, while in acidic environments, the highest grade is X65 [244].

In recent years, the development of deep-sea pipelines has been rapid, with increasing depths. The design water depth of newly constructed submarine pipelines abroad has reached 3500 m, with most using X70 steel grade, large diameter, and thick-walled steel pipes. The maximum pipe diameter has reached 1219 mm, and the maximum wall thickness has reached 44.0 mm [245]. The application of typical foreign submarine pipelines is shown in

Table 2. Typical application of seabed pipeline abroad.

Line Name	Laying Depth/m	Grade of Steel	Pipe Diameter/mm	Wall Thickness/mm
The Gulf of Mexico ITP Pipeline	2412	X65	610	24.1~34.3
The Mediterranean Medgaz Pipeline	2160	X65	640	-
Ichthys Gas Pipeline	2160	X65	1067	29.6~40.1
The North Sea Lang Grad Pipeline	1000	X70	1016	34

.

The laying of submarine pipelines in China commenced relatively late due to the slow development of marine engineering and the relatively backward state of equipment and technology. Consequently, there remains a significant gap between the overall level of China’s offshore pipeline design and construction and the international advanced standards [246].

In recent years, China has rapidly advanced its offshore pipeline construction efforts. By the end of the “Twelfth Five-Year Plan,” China had laid hundreds of submarine pipelines, with a total length exceeding 6000 km. Submarine pipeline steel grades X56 and X60, suitable for H2S acidic service environments, have been applied in several Chinese submarine pipeline engineering projects, accompanied by the development of related welding process research.

By summarizing foreign experience and conducting extensive experimental research, several domestic pipeline steel and steel pipe manufacturers in China have mastered the production technology for high-strength, high-toughness pipeline steel. Consequently, the steel grade for long-distance submarine pipelines has been elevated to the X70 level, alleviating China’s long-term reliance on imported high-grade oil and gas pipeline steel [247,248].

Ma et al. [249] conducted research on the production and application of 10MnVNbNiMo steel for marine pipeline materials. This steel is produced using high-quality scrap and first-grade molten iron, with phosphorus content controlled within 0.010% during electric furnace tapping. The refining process is enhanced with deoxidation to create white slag, keeping the sulfur content of the finished product controlled within 0.002%. Refining involves precise composition adjustments to ensure the carbon equivalent is controlled between 0.17% and 0.19%. During the later stages of VD (vacuum degassing), calcium treatment and soft blowing are employed to promote the adequate flotation of inclusions, resulting in inclusion ratings of less than or equal to grade 1.0 in the pipe body samples. Reasonable process parameters are set according to the characteristics of the continuous casting of this steel type, ensuring that the surface and internal quality of the cast billets fully meet the rolling requirements.

8. Digital Twin and Real-Time Monitoring Technologies

As the wave of AI continues to surge, artificial intelligence is increasingly being applied across various scientific and engineering fields. Similarly, AI holds immense potential for the protection of offshore pipelines and subsea cables. The marine environment is complex and volatile; traditional periodic manual inspection models are not only prohibitively expensive but also suffer from significant response lags. Consequently, Digital Twin technology and real-time monitoring powered by AI are transforming protection strategies from reactive “after-the-fact” repairs to predictive maintenance. By significantly reducing costs, this shift is becoming the future trend and a key potential solution for the industry [250,251,252].

Chen et al. (2025) [253] developed a digital twin–based predictive diagnosis method for subsea suspended pipelines to overcome the limitations of traditional inspection approaches, which typically lack long-term and multi-directional diagnostic capabilities. For these pipelines, a unit-level digital twin framework was proposed, enabling real-time monitoring, fault prediction, and operational optimization through digital twin functionalities. In addition, a PFD-TCN (Polynomial Fitting Denoising–Temporal Convolutional Network) model was introduced. By integrating polynomial fitting functions with a conventional TCN, the proposed model effectively reduces noise and improves pipeline strain prediction accuracy by more than 50% compared with the traditional TCN model. However, acquiring more comprehensive data generally requires the installation of additional subsea sensors, which can be costly. Moreover, the actual stress states of subsea suspended pipelines may deviate from idealized models, indicating the need for further investigation into pipeline force characteristics. Furthermore, subsequent validation of modeling approaches under real offshore conditions can also be economically expensive.

Similarly, Chen et al. (2024) [254] integrated physical monitoring with virtual simulation to develop an integrated platform management system. This system periodically collects monitoring data on environmental conditions, offshore platforms, mooring lines, and risers from field sensors, and employs long short-term memory (LSTM) neural networks to reconstruct virtual simulation models of the dynamic responses of offshore platforms under environmental loads.

Repalle et al. (2020) [255] proposed a novel approach that combines refined finite element analysis (FEA) with artificial neural networks (ANNs) to construct a digital twin model for risers. This model serves as a decision-support tool for integrity management and life-extension operations. By training advanced neural networks using partially available metocean and vessel motion data, the digital twin can be used to predict fatigue damage under various metocean conditions and internal pressure scenarios. Compared with conventional finite element analysis, this approach significantly reduces the time required for fatigue damage assessment while improving prediction accuracy.

Bhowmik (2021) [256] proposed a novel digital twin concept based on computer vision, in which convolutional neural network (CNN) algorithms are employed to perform real-time corrosion detection and classification using remotely operated vehicle (ROV) images and online inspection data. This approach significantly improves the performance of existing corrosion identification methods, reduces inspection time, and supports predictive and prescriptive maintenance strategies, achieving an accuracy of approximately 81% in corrosion detection and classification. However, certain limitations remain, as CNN-based corrosion detection methods do not inherently capture the uncertainty and risk associated with the detected corrosion.

Imran et al. (2023) [257] investigated recent artificial intelligence (AI) techniques for marine-related corrosion prediction and detection, and highlighted that both real-time and historical data are crucial for predicting corrosion failures within data-driven predictive maintenance (PdM) models. In some approaches, real-time data are used to frequently update prediction models, while historical data are leveraged to support the characterization of corrosion defect distributions.

Meniconi et al. (2024) [258,259] conducted a two-part companion study on fault detection in the Trieste subsea pipeline in Italy. The first paper focuses on the off-site preparatory work, including the selection of a small Side Discharge Valve (SDV) through hydraulic characterization to generate small but sharp pressure waves, and the formulation of field test procedures to ensure the system reaches a stable steady state. In the second paper, the authors analyze actual field test data using 1D numerical modeling and Inverse Transient Analysis (ITA) to identify pipe wall deterioration in certain sections, a finding subsequently confirmed by professional divers’ field investigations.

Current research on Digital Twin and real-time monitoring technologies for FPSO systems focuses primarily on the hull, topside structures, and production planning. In contrast, research concerning the real-time monitoring of subsea pipelines and cables within FPSOs remains in its infancy, representing a significant research gap [260,261,262].

9. Conclusions and Prospects

A wide survey of the literature for risers, umbilical cables, and mooring cables for a FPS in offshore floating production would indicate that, although current results provide a robust engineering platform, there are still considerable knowledge gaps in theories, models, and their applicability. Based on the review presented in this paper, the hazards, underlying mechanisms, typical models, and monitoring/mitigation measures related to marine pipelines and cables—particularly those in FPSO systems—are summarized in

Table 3. Synthesis of key failure hazards, mechanisms, typical models, and normative standards for pipelines and cables in marine and FPSO systems.

Hazard	Key Mechanism	Typical Model	Normative Standard	Monitoring and Mitigation Strategy
Vortex-Induced Vibration (VIV)	Periodic vortex shedding causing resonance and fatigue damage in free spans or risers.	Wake oscillator models; Morison Equation	DNV-RP-F105 [263]	Accelerometers, strain gauges, Digital Twin prediction, Helical strakes, fairings, span rectification
Local Scour and Free Spans	Sediment transport due to wave interaction removing seabed support.	Potential flow theory; Empirical scour depth formulas	DNV-RP-F105 [263], DNV-RP-F109 [264]	Side-scan sonar, multi-beam echo sounder, Rock dumping, concrete mattresses, sediment stabilization.
Corrosion	Electrochemical oxidation; Microbiologically Influenced Corrosion in anaerobic mud.	Melchers’ Multi-phase Model; Bayesian Network	DNV-RP-B401 [265]	Linear Polarization Resistance (LPR), Electrical Resistance (ER), Cathodic protection (sacrificial anodes), anti-corrosion coatings
Buckling	Thermal expansion under High-Pressure High-Temperature (HPHT) conditions constrained by pipe-soil friction.	Hobbs’ analytical method; Euler-Bernoulli beam theory.	DNV-RP-F110 [266]	Distributed Temperature/Strain Sensing, Snake-lay installation, intermittent rock dumping
Fatigue Failure	Cyclic loading from waves/currents or accidental loads	S-N Curve analysis; Palmgren-Miner rule.	DNV-RP-C203 [267]	Intelligent PIGs for crack detection, Stress relief joints, fatigue-resistant welding procedures.

.

Morison’s equation is very popular for basic hydrodynamic calculation work but fails to account for either wave diffraction or radiation effects. This is particularly a concern when dealing with very large structures or with situations where the diameter is a large fraction of the wavelength, leading to inaccurate predictions of hydrodynamic force. Semi-empirical models combining potential flow theory with adjustments provide a means to improve accuracy, though models and a method to establish calibration constants appropriate to specific structures and environment conditions are not established.

Secondly, for seabed scour and free spans of seabed pipes, the current models involve simplistic theories such as Two-Dimensional Steady Flow, rigid pipes, and empty seabed. Additionally, they fail to involve oceanic phenomena such as Interaction between Currents and Waves, Non-Uniform Current Flow, and different seabed materials. Essentially, this implies a less accurate method of predicting the scour and span changes with respect to safety. Additionally, research on Vortex-Induced Vibration (VIV) assumes a uniform flow condition and a span setup without putting adequate emphasis on the combination of scour, soil interactions, and the phenomenon of vibration.

Third, the factors in buried or partially buried pipelines involving the influence of sediments on electrochemical and microbiologically influenced corrosion processes are not quantified. The influence of different sediments on the diffusion of oxygen, bioactivity, and the rate of corrosion is not modeled.

Finally, while AI and Digital Twin technologies offer new ways to sense condition, warn early about risks, and manage operations, using them safely in offshore engineering has challenges. These include a lack of high-quality training data, trouble collecting real-time data, unclear model validation paths, and the “black-box” nature of many neural networks.

Looking ahead, cross-disciplinary teamwork is essential. Future work should combine large-scale physical tests, high-fidelity multi-physics simulations, and long-term field data. It is important to promote Explainable AI and Physics-Informed Neural Networks (PINN) within reliability-based design. These advances will strengthen safety, durability, and sustainable growth of floating offshore energy systems. At the same time, this study has certain limitations. Due to the relative scarcity of research specifically focusing on subsea pipelines and cables compared to FPSO hulls or topside systems, this paper incorporates a broad range of studies related to the three primary types of lines and cables to ensure comprehensiveness and credibility. Future efforts should focus on further expanding the database of research cases in this specific field.