Effects of Damaged Fiber Ropes on the Performance of a Hybrid Taut-Wire Mooring System

Station-keeping of offshore large floating structures is a key technology for the exploitation of marine resources. For any type of offshore large floating structures, a reliable and economic positioning technology is of great necessity [1–3]. Note that, for a permanent positioning of floating structures in the field, catenary, and taut-wire mooring systems are used, as illustrated in Fig. 1 [4]. Catenary mooring systems often use steel chains or wire ropes as mooring lines and rely on the weight of the mooring lines to provide part of the restoring forces. However, the weight (and cost) of the mooring lines increases as their length increases. This relationship causes catenary systems to become less efficient and less economical in deeper water. Therefore, the use of synthetic fiber ropes as main sections of the mooring lines was proposed in the 1960s [5]. In 1997, the first polyester taut-wire mooring system was successfully installed in the Campos Basin, offshore Brazil [6]. Since then, the polyester ropes used in taut-wire mooring systems have become the preferred option for depths up to 1500 m. However, with numerous discoveries of natural resources in deeper waters, the question of whether the polyester ropes can provide sufficient stiffness to maintain acceptable platform offsets at various depths has been raised [7–9]. Based on the observation that the stiffness of high modulus polyethylene (HMPE) ropes is higher than that of the polyester ropes at the same minimum breaking strength, the HMPE ropes were considered as alternatives to the polyester ropes [7,9]. However, compared to polyester, HMPE shows a lower elongation at breakpoint as well as a higher creep rate and a higher Young's modulus, which causes a larger load range in a storm [10–12]. Considering these aforementioned factors, a preferred hybrid rope configuration was proposed [12,13].This configuration uses the stiffer HMPE rope in cooler water closer to the seabed and the polyester rope in warmer water closer to the vessel or platform. Consequently, taut-wire mooring systems utilizing various hybrid mooring lines have the capability to keep the position of the floating structures in deepwater locations [14,15].

Besides the function of positioning, a point worth emphasizing is that mooring systems also play a vital role in security assurance of floating structures in the harsh environment. According to the statistics of engineering practices, the occurrence of catastrophic failures, such as capsizing, fracture, leakage, uncontrolled drifting, and collision, is more frequent in the extreme marine environment than anticipated [2,3,16,17]. According to an investigation of 69 cases of failure of floating platforms since 1964, the number of failures of the mooring system is 47, reaching a high level of 68%, while the remaining 32% represent the other factors, as listed in Table 1 [3]. This shows that the performance of the mooring systems is of significance for the safe operation of the floating structures [2,3].

In taut-wire mooring systems, fiber ropes serve as load transmitters between the floating structures and their anchors. In engineering applications, fiber ropes may be damaged due to the following reasons: (1) rope handling during the installation, (2) wear experienced during the service, (3) ingress of sand and marine growth, (4) material and manufacturing defects, and (5) local sub-rope or element rupture during the service [17–20]. The damage to the mooring lines over time intensifies or even causes the failure of the mooring lines. Once mooring line damage or failure occurs, it will affect the dynamic response of the mooring system or even lead to loss of station-keeping capability. In addition, the cost of a single mooring line failure and replacement can be enormous. Thus, it is necessary to understand the mechanical behaviors of the damaged fiber ropes and the sensitivity of the mooring system’s performance to the damage levels of the fiber ropes.

In the aspect of the mechanical behavior of the damaged fiber ropes, the residual strength of damaged fiber ropes was experimentally investigated by Williams et al. [18], Ward et al. [19,20], and Flory [21]. All these studies adopted the method of cutting some sub-ropes or elements to create the initial damage state of specimens. In a study by Liu et al. [22], an empirical equation that accounts for the dynamic stiffness of damaged fiber ropes was proposed based on experimental investigations. The aforementioned research works are helpful in estimating the positioning performance of the mooring system with the damaged fiber ropes.

In terms of dynamic response of the mooring systems with the damaged mooring lines, Huang et al. [23] performed a fatigue analysis of a polyester taut-wire mooring system of a floating production storage and offloading (FPSO) model and investigated the response of the mooring system with an one-line failure. Qiao et al. [24] investigated the corrosion effect on fatigue life of two mooring systems. Ahmed et al. [25] was apparently the first to evaluate the dynamic response of truss spar platforms for various catenary mooring configurations with the damage levels of the mooring lines and simplified these lines as springs. However, to investigate the effect of the damage levels of the fiber ropes on the mooring performance, it is important that the nonlinear behavior of the fiber ropes should be considered, for example, by introducing the empirical equation of dynamic stiffness [15,23,26].

However, little work has been devoted so far on examining the effects of the damage levels of the fiber ropes on the dynamic response of a taut-wire mooring system. The reason may be that the damage effect on the fiber mooring lines has not been understood well. Hence, the present study may be of great necessity in the design hybrid taut-wire mooring systems at the preliminary stage.

To investigate the effects of the damage levels of fiber ropes on the performance of a taut-wire mooring system, a numerical model of the FPSO, and the details of the mooring system are presented as follows.

In the present study, an FPSO numerical model installed in 3000 m of water depth is developed. A procedure for establishing the numerical model is presented as follows. First, based on the ship shape diagram (as shown in Fig. 2(a)) and the principal dimensions of the FPSO (as listed in Table 2), using a mesh generator in hydrostar (software) to generate the numerical model mesh, as shown in Fig. 2(b). After creating the model mesh, the diffraction and radiation coefficient, the motion of the ship, the second-order loads using near field, and the middle field formulations are derived [27,28]. Then, the response amplitude operators (RAOs) transfer functions are constructed using hydrostar [28]. The RAOs as functions of wave frequency are shown in Fig. 3 for five different incident wave angles $θ=0deg,22.5deg,45deg,67.5deg,90deg$⁠. The wave heading is defined as the angle between the propagation direction and the direction of the axis O_x, which is positive in the forward direction of the vessel. The detailed procedure for developing the numerical model and implementation can be obtained from the hydrostar manual [28].

The mooring system is composed of four groups, and each group contains three hybrid mooring lines, as illustrated in Fig. 4. The configuration of the hybrid lines comprises the bottom chain linked to an anchor, the HMPE rope in the cooler water closer to the seabed, the polyester rope in the warmer water closer to the vessel and the hang-off chain linked to the fairlead. The hybrid mooring line has the following benefits. First, the bottom and hang-off chains have good resistance to abrasion of the seabed and wear at trumpet welds, respectively. Second, the hybrid mooring line including the HMPE and the polyester ropes can provide adequate stiffness needed to handle maximum loads during the station-keeping in a storm, while ensuring sufficient elasticity to control the offset of floating platforms induced by environmental loads. Third, the lower environment temperature in cooler water closer to the seabed decreases the HMPE creep rate and assures the survivability of the HMPE ropes. Furthermore, the hybrid line allows the mooring designer to engineer the stiffness of the mooring system by controlling the length of the HMPE rope components.

In Table 3, detailed parameters of the hybrid mooring line, including the length, diameter, and minimum breaking strength (MBS) of rope segments, are listed. In addition, the dynamic stiffness of the fiber ropes is introduced to obtain the dynamic response of fiber mooring systems [1,7,23]. The “dynamic stiffness” is defined as that the ratio of the rope load to the strain between the trough and peak loads imposed during cyclic test, typically normalized by the minimum breaking strength. In the present study, to investigate the effects of the damage levels of the fiber ropes on the performance of the mooring system, the next paragraphs contain a more precise explanation of the dynamic stiffness of initially damaged fiber ropes.

To investigate the effect of damaged lines on dynamic response of the hybrid taut-wire mooring systems, a representative extreme environmental load of combined wind–wave–current forces is used in the area of South China Sea S4, as shown in Table 7 [23,32]. The wave and wind condition is for a 100-year return period [32]. A constant profile for the current condition is for a 10-year return period [32]. In addition, to investigate the effect of one-line failure on the fatigue life of the adjacent lines, the long-term environmental conditions in the area of Nanhai S4 of China are selected and represented by 64 short-term environmental states. The corresponding environmental parameters are listed in Table 8 [23]. It should be noted that the wave condition is given by the JONSWAP spectrum with the peak enhancement factor γ = 2.5, the wind load is generated using the API wind spectrum, and the current is represented with the constant current spectrum [33]. In Tables 7 and 8, H_S, T_S, V_W, V_C, and P denote the significant wave height, the peak period, the mean wind velocity, the mean current velocity, and the occurrence probability of each sea state, respectively. In the present study, the analysis is performed only for the collinear direction of the wave, wind, and current case. The collinear direction is selected as $90deg$⁠, which is measured counterclockwise from the East, as shown in Fig. 4. The reason that all loads are applied in the same direction is based on the results of the following studies. In the fatigue analysis example of the API-recommended practice 2SK [34], the wind, wave, and current are assumed colinear. In addition, Ahmed et al. [25,35] investigated the response of a truss spar under the unidirectional waves, current, and wind loads based on the observation that waves are accompanied by the current and the wind in practical situations.

To estimate the fatigue life of the mooring systems, a linear damage accumulation rule (Palmgren-Miner), T-N curves, the rainflow cycle counting method, and time-domain dynamic theory are employed. This fatigue analysis method is used because the rainflow counting method, first proposed by Matsuishi and Endo [37], has been regarded as the classic method to estimate the fatigue life of structure elements [38]. The method is based on the counting of peaks and valleys, which is similar to the rain flow falling on a pagoda and running down the edges of the roof, to determine the equivalent constant amplitude cycles [39]. This fatigue analysis method has been widely applied in the designs of the marine structures [34,40–42].

The detailed calculation procedure for the fatigue analysis is delineated as follows. First, the long-term environmental events are represented as discrete cases. Second, a time-domain dynamic analysis of the mooring system is performed to obtain the tension time history of the mooring lines of each case. Third, the annual fatigue damage, D_j, is calculated using the rainflow method and T-N curve. Finally, the total annual fatigue damage, D, and the fatigue life, L, are obtained using the Palmgren-Miner rule.

In the present study, a case of using undamaged mooring lines is selected as the reference case in designing a representative analysis set. Based on the method described above, the response of the intact mooring system under the environmental loads from Table 7 is calculated. Here, the collinear direction of the wave, wind, and current is defined as $90deg$⁠, which is measured counterclockwise from the East, as shown in Fig. 4. Hence, Lines 4–6 are on the windward side of the FPSO and Lines 10–12 are on the leeward side of the FPSO. The axial tensions of the undamaged mooring lines at the fairlead are obtained, as shown in Table 9. As indicated in the table, the maximum load and the load range of the windward Line 5 is larger than those of the other mooring lines. These results are consistent with those specified in the guideline of API-RP 2SM [41], which shows that the windward lines withstand higher environmental loads than the leeward lines. According to the tension results of the intact mooring system, the windward Line 5 is selected as the reference line to design the analytical cases. Hence, nine cases are designed to investigate the effect of the damaged lines on dynamic response of the hybrid taut-wire mooring system, as shown in Table 10. Case 1 is the reference case of the intact mooring system. Cases 2–4 are designed to investigate the dynamic response of the mooring system with the initial damage level of Line 5 increasing until failure. Case 5 is used to investigate the dynamic response of the mooring system with the failure of Line 5, and 30% of the initial damage level of Line 4 occurs. Case 6 is designed to investigate the dynamic response of the mooring system considering the failures of both Lines 4 and 5. Case 7 examines the dynamic response of the mooring systems considering the effects of the damaged windward Lines 4–6. Case 8 examines the dynamic response of mooring systems under the effects of the damaged leeward Lines 10–12. Case 9 examines the dynamic response of the mooring system when all the lines suffer 5% damage. In addition, to investigate the effect of a failed windward line on the fatigue life of the remaining windward rope, which is adjacent to the failed one, fatigue analyses of Line 6 in Cases 1 and 4 are performed.

Based on the dynamic analysis method presented above, the maximum horizontal offsets of the FPSO for every case are calculated and presented in Table 11. It can be observed from Table 11 that the maximum offset of the FPSO changes with the damage levels of the mooring lines under the same environmental loads from Table 7. Cases 2–4 in Table 11 indicate that with the damage to Line 5 increasing, the maximum horizontal offset of the FPSO increases accordingly. Specifically, once the failure of Line 5 occurs, the maximum horizontal offset of the FPSO significantly increases from 23.37 m to 31.86 m. Case 5 in Table 11 indicates that the 30% damage level of Line 4 adjacent to the failed one leads to ((38.85–22.42)/22.42) 73.28% increase compared with Case 1. When Line 4 also fails as shown in Case 6, the increase in the maximum horizontal offset is quite significant, i.e., 185.28% of the Case 1 offset. When the windward Lines 4–6 are at 20.0% damage level (Case 7), the increase in the maximum offset is ((26.90–22.42)/22.42) 19.98% of the Case 1 offset. When the leeward Lines 10–12 are at 20.0% damage (Case 8), the increase in the maximum offset is 17.71% of the Case 1 offset. A comparison of the horizontal offset between Cases 7 and 8 shows that the maximum horizontal offset of Case 7 is slightly larger than that of Case 8, as shown in Fig. 5. Comparing Case 9 with Case 1 in Table 11, the uniform 5.0% damage to all lines causes a 1.63-m increase in the maximum horizontal offset.

These results show that the maximum horizontal offset of the FPSO increases with the increasing damage levels of the windward mooring lines because the damage causes the degradation of dynamic stiffness of the fiber ropes. Furthermore, with the stiffness degradation of the fiber ropes, the entire stiffness of the mooring system decreases. Particularly, when the failure of one windward line occurs, the increase in the maximum horizontal offset is about 42.11% of the Case 1 offset. If the failure of two windward lines occurs, the increase in the maximum horizontal offset is about four times the offset of the case of the one-line failure. This indicates that the offset of the FPSO is sensitive to the failure of the windward lines. Hence, capturing and monitoring the offset of the floating structures may help to assess the integrity of the mooring systems. In addition, Table 11 also shows that the offset heading of the FPSO is affected by the damaged mooring lines. From the results of Case 5 in Table 11, it can be observed that the damage level of Case 5 (30% damage level of Line 4 and the failure of Line 5) leads to an increase of 0.186 deg compared with that of Case 1. This indicates that when the damaged lines changes the symmetry of the mooring arrangement under the unidirectional environmental loads, the offset heading increases. Moreover, the abrupt change in heading due to the sudden mooring line failure may increase the roll motion of the FPSO and affect the operational safety [44]. Hence, the present study helps to understand the effect of the damaged mooring line on the offset and heading of the FPSO.

In this section, the line tensions at fairleads of the mooring ropes are reported. In Tables 9 and 12, the MBS is 19600 kN for the polyester and the HMPE mooring ropes. Table 9 shows the line tensions of the intact mooring system. In addition, Table 12 shows the maximum tension and tension range values of Lines 5 and 6 for each case. It is observed that when Line 5 is at 20% damage (Case 2), the decrease in the maximum tension of Line 5 is (35.99–33.59) 2.40% MBS, compared with Case 1. In terms of tension response of Line 5, the difference between Cases 1 and 2 is small. When Line 5 suffers 30% damage (Case 3), the decrease in the maximum tension of Line 5 is (35.99–32.53) 3.46% MBS. The tension response of Line 5 in Case 3 is similar to that in Case 2. When failure of Line 5 occurs (Case 4), the increase in the maximum tension of Line 6 is (38.68–35.74) 2.94% MBS, compared with Case 1. The tension response of Line 6 for Cases 1 and 4 are presented in Fig. 6. When the failure of Line 5 occurs and Line 4 incurs 30% damage (Case 5), the increase in the maximum tension of Line 6 is (42.67–38.68) 3.99% MBS, compared with Case 4. This shows that the increase in the damage level of Line 4 causes the maximum tension of Line 6 to increase. When both Lines 4 and 5 fail (Case 6), the increase in the maximum tension of Line 6 is (56.79–35.74) 21.05% MBS compared with Case 1, which is a noticeable increase, as shown in Fig. 7. Furthermore, to analyze the sensitivity of the tension response of Lines 5 and 6 to the damage levels of mooring ropes, the maximum tensions of both Lines 5 and 6 at the fairlead from Cases 1 to 6 are shown in Fig. 8. This figure shows that the maximum tension of Line 5 decreases with the increasing damage level of Line 5. At the same time, the maximum tension of Line 6 increases with the increasing damage levels of both Lines 4 and 5. In addition, the tension ranges of both Lines 5 and 6 have a similar trend as their maximum tension, as shown in Fig. 9, which also shows that the tension ranges of Line 6 increase with the increasing damage levels of both Lines 4 and 5.

When all the windward Lines (4–6) are at 20% damage (Case 7), the decrease in the maximum tension of windward Line 5 is (35.99–35.49) 0.50% MBS, while the increase in the maximum tension of the leeward Line 11 is (32.04–31.05) 0.99% MBS compared with Case 1. Similarly, when all the leeward Lines (10–12) are at 20% damage (Case 8), the increase in the maximum tension of windward Line 5 is (38.15–35.99) 2.16% MBS, while the decrease in the maximum tension of the leeward Line 11 is (31.05–30.17) 0.88% MBS compared with the intact mooring (Case 1). When the uniform 5% damage to all lines (Case 9) occurs, the tension response of Line 5 for Cases 1 and 9 is similar, but Case 9 results in (35.99–35.33) 0.66% MBS decrease in the maximum tension of Line 5, compared with Case 1.

Observing that the line with the maximum load is usually the windward mooring line, therefore, the windward line should receive more attention. The damage to the windward line over time can lead to an increase in the broken danger. If the damage to a single mooring line is not detected, the tension of the remaining lines will be increased, which may cause the line to fail in the next instance of severe weather. Even though an FPSO is designed to cope with the failure of a single mooring line, in the event of subsequent line failures, the increased loads tend to cause additional lines to fail. The failure of multiple lines may result in an FPSO breaking away from the moorings and lead to catastrophic failures. The present study helps to understand the effects of the damage levels of the fiber ropes on the performance of the mooring system.

The above results of the tension response show that the maximum tension of the remaining line, adjacent to the failed one, significantly increases. Line 6, which is adjacent to the failed Line 5, is the line carrying the maximum load and is thus selected as the line on which fatigue analysis is performed. By comparing the fatigue life of Line 6 between Cases 1 and 4, the effect of failure of Line 5 on the fatigue life of Line 6 can be obtained. The tension time curves of Line 6 for Cases 1 and 6 under the 64 short-term environmental loads are utilized to obtain the damage levels of the mooring lines based on the rainflow method and T-N curves. A T-N curve of the polyester ropes is obtained from the guidance note of API in which G is 9 and K is 7.5 [42]. However, the T-N curve of the HMPE ropes has not yet been suggested in guidance notes of ocean engineering [34,41–43]. Only the DSM Dyneema Company has published some experimental fatigue data of the HMPE ropes [45]. Using these data of the HMPE ropes, the parameters of T-N curve were computed and found that G and K are equal to 1.26 and 28,704,210.73, respectively.

Based on the method described above, the fatigue damage and fatigue life of Line 6, which is adjacent to Line 5, are obtained for Cases 1 and 4, as listed in Table 13. This shows that when the failure of Line 5 (Case 4) occurs, the decrease in the fatigue life of the polyester rope and the HMPE rope in Line 6 is about 50.74% and 19.90% of the Case 1 fatigue life, respectively. It can be concluded that a mooring line failure leads to the change of the mooring configuration, which could increase the maximum load and load range in the remaining lines adjacent to the failed one. Furthermore, the increasing tension range of the line adjacent to the failed one induces more fatigue damage. The results presented in this study help to understand the effect of the failed windward mooring line on the fatigue life of the remaining lines, adjacent to the failed one, in the hybrid mooring system.

Over the service life of the taut-wire mooring systems, rope damage and mooring incidents have been occurring at a high rate. The dynamic response of the mooring systems with the damaged or failed lines has been of high concern to the offshore engineers and platform operators. Hence, it is important to investigate the effects of damaged ropes on the dynamic response of the taut-wire mooring systems.

In this study, based on the dynamic stiffness equation of damaged fiber ropes, the time-domain dynamic theory, the rainflow cycle counting method, and the linear damage accumulation rule of Palmgren-Miner, an analysis is presented to investigate the effects of the damage levels of fiber ropes on the safety performance of a hybrid taut-wire mooring system. The dynamic analyses of the FPSO with the damaged fiber mooring system under the typical extreme environmental loads were performed. The numerical results indicate that the offset of FPSO and the tensions of the remaining windward lines increase with the increasing damage levels of the windward ropes. For example, when the failure of one windward line occurred (Case 4), the increase in the maximum horizontal offset was 42.11% of the offset of the intact system. In addition, the increase in the tension of the remaining windward line was 8.2% of the line tension of the undamaged mooring system. Furthermore, once a windward mooring line fails, the tension range of other windward mooring lines increases. In addition, fatigue analyses of the intact and damaged mooring systems demonstrated that the failure of a windward line causes a change of the mooring configuration, which could remarkably increase the maximum tension, tension range, and the annual fatigue damage levels of the remaining windward lines, adjacent to the failed one. Hence, once a mooring line has serious damage phenomena, the displacement of the floating platform, the maximum tension, and the fatigue life of the damaged line need to be evaluated to ensure safe operation of the mooring system. The results also show that when the damage of all the windward lines occurs, the maximum tension of the windward lines decreases and the maximum tension of the leeward ones increases, and vice versa. These conclusions are valid only for a typical extreme environmental condition. However, the present work provides a method for investigating the effects of damaged fiber ropes on the performance of the hybrid taut-wire mooring system. This study, therefore, helps to ensure safe operation of the mooring system.

The first and second authors greatly appreciate the support of the Oregon State University School of Civil and Construction Engineering for the first author's two-year study.

• National Natural Science Foundation of China (Grant Nos. 51609079 and 51539008; Funder ID: 10.13039/501100001809).

• Fundamental Research Funds for the Central Universities (Grant No. 2017B11314; Funder ID: 10.13039/501100002942).

• International Postdoctoral Exchange Fellowship Program 2017 (Grant No. 20170013; Funder ID: 10.13039/100011040).

D =

the total annual accumulated fatigue damage

L =

the calculated fatigue life

M =

the structure mass matrix

P =

the occurrence probability of each sea state

R =

the ratio of tension range to MBS (minimum breaking strength)

d_j =

the dynamic simulation duration under the environmental state

n_jk =

the number of cycles within the tension range R under the environmental state j

p_j =

the occurrence probability of the environmental state j

D_j =

the annual fatigue damage under the environmental state j

H_S =

the significant wave height

K_r =

the dimensionless dynamic stiffness

$Ma∞$ =

the added mass matrix at the infinite frequency

N_k =

the number of cycles to failure at normalized tension range R

T_m =

the mean tension of fiber ropes

T_S =

the peak wave period

V_C =

the mean current velocity

V_W =

the mean wind velocity

G, K =

the coefficients of T-N curves of fiber ropes

MBS =

minimum breaking strength of mooring lines (kN)

R(t − τ) =

the retardation function

{x} =

the motion displacement of floating platform

${x¨}$ =

the motion acceleration of floating platform

${x˙}$ =

the motion velocity of floating platform

[B] =

the damping matrix

[F] =

the loads, including the wave, wind, current, damping, and mooring forces

[K] =

the stiffness matrix including hydrostatic stiffness and stiffness from mooring lines

α =

the fatigue safety factor

δ, ω, β, ψ =

coefficients of dynamic stiffness of damaged fiber ropes

ϕ =

the initial damage value of fiber ropes