PSI - Issue 77

Sunny O. Uguzo et al. / Procedia Structural Integrity 77 (2026) 521–528

524

Table 1: TCP section geometry definition.

TCP Characteristics Inner radius (r i ) Outer radius (r e ) Inner liner thickness Outer liner thickness Laminate thickness

Value (m)

0.076

0.1

0.008 0.008 0.008 0.001

Ply thickness

The methodology combined linear buckling analysis (LBA) and failure analysis. First, the LBA estimated the critical axial buckling load (P cr ) for various ply angles under bending, pressures, axial compression and thermal gradients. Then the P cr was reapplied as the axial load with other loads in a failure analysis to determine material state and governing failure at varying internal surface temperature (T i ) and thermal gradients. Cylindrical coordinate stresses obtained from FEA were transformed to principal coordinate stresses, while Hashin and Von Mises failure criteria quantified laminate fibre/matrix failures and liner yielding respectively (see also Li & Sitnikova, 2018). 3. Results and discussions 3.1. Effect of laminate architecture on critical axial buckling load Guided by literature, this study analysed selected TCP laminate configurations (see Table 2) using the TCP model described in section 2.2. Loads applied on the 3m TCP for the LBA included a unit axial compression buckling factor, uniform bending moment of 25kNm, internal pressure (P i ) and external pressure (P e ) of 40 MPa and 20MPa respectively and thermal gradient corresponding with varying T i from 4 – 120 o C. External pipe surface temperature (T e ) was fixed at 4 o C whilst a heat transfer coefficient (h) of 50Wm -2 o C -1 was implemented. Temperature dependent material properties were also applied. Obtained P cr vs. T i for each investigated TCP laminate architecture are detailed and illustrated in Table 3 and Fig. 3.

Table 2: Investigated TCP laminate architecture.

TCP Label Laminate configuration TCP A [±30] 4 TCP B [±55] 4

TCP C [±55 2 /±30 2 ] TCP D [±75 2 /±30 2 ] TCP E [±65 2 /±75 2 ]

Table 3: P cr variation vs T i for different TCP laminate architecture.

TCP Label

Laminate configuration

P cr at T i – 4 o C (kN)

P cr at T i – 30 o C (kN)

P cr at T i – 60 o C (kN)

P cr at T i – 90 o C (kN)

P cr at T i – 120 o C (kN)

TCP A [±30] 4 TCP B [±55] 4 TCP C [±55 2 /±30 2 ] TCP D [±75 2 /±30 2 ] TCP E [±65 2 /±75 2 ]

4975 1680 3745 3935 1445

4781 1593 3602 3807 1385

4557 1495 3439 3665 1322

4332 1397 3275 3521 1258

4097 1287 3102 3364 1177

Critical buckling load (P cr ) decreases linearly with rising T i and thermal gradient, showing thermal softening and a negative correlation with buckling resistance. The all-low angle laminate TCP ([30] 4 ) has the highest baseline P cr of 4975kN which reduces to 4097 when T i is 120 o C. In contrast, intermediate angle laminate TCP [55] 4 has lower baseline P cr (1680kN) and across T i variations. The latter also suffers steeper P cr thermal decay as T i rises (5.3 – 26.5% vs. 4 – 19.4%), confirming that low angle plies better preserve axial stiffness. Also, TCPs with hybrid layup laminates also exhibit a decline in P cr as T i rises. The [±75 2 /±30 2 ] laminate TCP starts highest at 3935kN, followed by [±55 2 /±30 2 ] with 3745kN, whilst all high angle laminate TCP [±65 2 /±75 2 ]was lowest at 1445kN. Thermally,