1、 Commentary on the Guide for Buckling and Ultimate Strength Assessment for Offshore Structures COMMENTARY ON THE GUIDE FOR BUCKLING AND ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES MARCH 2005 (Updated February 2014 see next page) American Bureau of Shipping Incorporated by Act of Legislature
2、 of the State of New York 1862 Copyright 2005 American Bureau of Shipping ABS Plaza 16855 Northchase Drive Houston, TX 77060 USA Updates February 2014 consolidation includes: March 2005 version plus Corrigenda/Editorials ABSCOMMENTARY ON THE GUIDE FOR BUCKLING therefore, uncertainties in loads and r
3、esistances are not specially addressed, but are inherently incorporated into the maximum strength allowable utilization factors. The formulations proposed are generally based on the premises that: They should not depart significantly from the formulations presented in ABS existing Rules and Guides a
4、nd be consistent throughout the whole Guide; Where departures from existing ABS formulations are recommended, they should tend towards a formulation presented in other widely used design standards, such as API RP 2A-WSD; Where appropriate, improvement in formulation accuracy, whether the starting po
5、int is ABS MODU Rules1, ABS Steel Vessel Rules2or API RP 2A-WSD3, should be included in the proposed formulations. In order to validate the two- or three-dimensional interaction equations of buckling and ultimate strength proposed in the Guide, a modeling uncertainty is introduced, which was suggest
6、ed by Hoadley and Yura(1985)4. The modeling uncertainty is the ratio of the distance from the origin to the test data point in question, L1, over the distance from the origin to the interaction curve, L2, and is written by: Modeling Uncertainty = L1/L2An example of the modeling uncertainty is shown
7、in Section C1, Figure 1. From this definition, the buckling and ultimate strength prediction is conservative if modeling uncertainty is greater than 1.0. The modeling uncertainty is especially useful because it can be used in one, two and three dimensions, and it is not a function of the exponent of
8、 each term in the interaction equation. In addition, it can be used to determine the amount of conservatism in a state limit when the experimental points are outside the range of the interaction equation when excluding factors of safety. This concept is also extended to determine the amount of conse
9、rvatism of a design when design loads are inside the range of the interaction equation including factors of safety. In spreadsheets developed by the ABS Offshore Technology Department, the so-called unity check method is used. In this method, the unity check is done by calculating the ratio of the d
10、istance, Q1, from the origin to the design load point A, over the distance, Q2from the origin to the point B on the interaction curve, as shown schematically in Section C1, Figure 2 and written by: Unity ratio = Q1/Q2The design is acceptable if the unity ratio is less than 1.0. Section C1 Introducti
11、on 2 ABSCOMMENTARY ON THE GUIDE FOR BUCKLING 2 on seamless pipe, Smith et al19; and 70 on ERW pipe, Steinmann and Vojta20and Yeomans21. This is considerably larger than that previously used to validate offshore tubular strength formulations. The increase is primarily due to the inclusion of relevant
12、 results from a large CIDECT test program (Yeomans21). The figure confirms that the ABS MODU Rules1and API RP 2A-WSD3formulations are identical. However, the statistics of the comparisons between the formulations and the test data indicate that differences do arise. For example, the means for the tw
13、o formulations are 1.0736 and 1.0743 respectively. An examination of the calculation details reveals that differences arise because of an API RP 2A-WSD local strength requirement. This applies for D/t 60; whereas the ABS MODU Rules local buckling limit is in excess of 60 (or using ABS MODU Rules1def
14、initions, D/t 59) for yield stresses up to 386 N/mm2. The mean and COV of modeling uncertainty of various codes are given in Section C2, Table 2. Section C2, Table 3 provides comparison between the ABS Buckling Guide and AISC Code16for two rolled-plate sections. The allowable buckling stress from th
15、e ABS Buckling Guide is remarkably close to that from the AISC Code when local buckling is ignored, as is the case for compact sections. Differences arise for non-compact sections, in which the allowable buckling stress from the ABS Buckling Guide is considerably smaller than that from the AISC Code
16、. This is reasonable because the ABS Buckling Guide includes the local buckling effect for non-compact sections. Section C2 Individual Structural Members ABSCOMMENTARY ON THE GUIDE FOR BUCKLING & ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES .2005 11 TABLE 2 Mean/COV of Modeling Uncertainty f
17、or Column Buckling ABS MODU Rules API RP 2A WSD ABS Buckling Guide Mean 1.0736 1.0743 1.0547 COV 7.56% 7.51% 5.28% FIGURE 4 Column Buckling for Tubular Members Modeling uncertaintyNumberofspecimens051015202530354045500.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5A BS MODU RulesAPI RP 2A WSDABS Buckling GuideTABLE
18、3 Column Buckling for Rolled-plate Sections W-shape Square Hollow Section Geometry and Material Length 144.00 144.00 Section shape type 2 4 Specified minimum yield point o36.00 36.00 Modulus of elasticity E 2.90E+04 2.90E+04 Poissons ratio 0.30 0.30 Flange width b 3.75 7.96 Flange thickness tf0.25 0
19、.43 Web depth d 3.75 9.75 Web thickness tw0.25 0.29 Section classification Non-compact Compact The ABS Buckling Guide Allowable buckling stress considering local buckling 13.73 Allowable buckling stress ignoring local buckling 16.14 19.90 AISC Code Allowable buckling stress ignoring local buckling 1
20、5.99 19.46 Section C2 Individual Structural Members 12 ABSCOMMENTARY ON THE GUIDE FOR BUCKLING & ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES .2005 C3.5 Bending Moment The ABS Buckling Guide includes two failure modes that take proper account of plastic moment capacity and lateral torsional
21、buckling capacity for the members. The proposed buckling state limit is defined by the following equation: b/2CB 1 where b= bending stress due to bending moment CB= characteristic bending strength given as follows: i) For tubular members, the critical bending strength is obtained from the equation i
22、n Section 2/9.3 of the ABS Buckling Guide, in which the fully plastic capacity of the section could be developed. ii) For members with rolled or fabricated sections, the critical bending strength is determined by the critical lateral-torsional buckling stress. The critical lateral-torsional buckling
23、 stress is obtained by: C(LT)= ( )FrLTELTEFrrFFrLTELTEPPPP)()()()(if11ifwhere E(LT)= elastic lateral-torsional buckling stress, which is given below (Timenshenko and Gere15) = 22)(kLSMEICcComparisons are presented for tubular members in Section C2, Figure 5 between the existing ABS MODU Rules, API R
24、P 2A-WSD and the ABS Buckling Guide for bending and the test data. The bending database consists of 57 results published by Steinmann and Vojta20, Kiziltug et al22, Sherman23,24, Korol and Hudoba25and Korol26. In the ABS MODU Rules1, bending strength is limited to the range where D/t E/90or 0D/Et 0.
25、11 and local buckling effect is ignored. Over this valid range, the ABS MODU Rules1underestimate the bending strength. Section C2, Table 4 presents the Mean/COV of modeling uncertainty of bending strength for tubular members. TABLE 4 Mean/COV of Modeling Uncertainty of Bending Strength for Tubular M
26、embers ABS MODU Rules API RP 2A WSD ABS Buckling Guide Mean 1.3678 1.1741 1.1463 COV 10.83% 9.40% 9.79% Section C2 Individual Structural Members ABSCOMMENTARY ON THE GUIDE FOR BUCKLING & ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES .2005 13 FIGURE 5 Bending Strength for Tubular Members Model
27、ing uncertaintyNumberofspecimens0246810121416180 0.5 1 1.5 2 2.5 3 3.5A BS MODU RulesAPI RP 2A WSDABS Buckling GuideThe critical bending strength for the beams with rolled or fabricated compact sections obtained from SSRC28, ECCS29, AISC LRFD30, DnV CN30.17and the ABS Buckling Guide is shown in Sect
28、ion C2, Figure 6. The criterion proposed in the ABS Buckling Guide is conservative for the short beam. In this case, the critical bending strength is governed by the development of full plasticity. The criterion proposed in the ABS Buckling Guide is acceptable for the beams with rolled or fabricated
29、 compact sections in the practical range of slenderness ratio. FIGURE 6 Comparison of Lateral-Torsional Buckling Strength 0.00.20.40.60.81.01.20 50 100 150 200 250 300 350 400Le/ryMcr/MpSSRCECCSAISC LRFDDnV CN30.1ABS Buckling Guidew24x55Section C2 Individual Structural Members 14 ABSCOMMENTARY ON TH
30、E GUIDE FOR BUCKLING & ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES .2005 C5 Members Subjected to Combined Loads C5.1 Axial Tension and Bending Moment C5.3 Axial Compression and Bending Moment The criteria for combined column buckling and bending moment in the ABS Buckling Guide is based on
31、the individual formulation for column buckling and bending combined via the interaction equation involving Euler amplification of deflections by axial loading. This applies when the ratio of axial stress ato the column strength CAis greater than 0.15, i.e., the buckling failure is dominant. Otherwis
32、e, a relationship that does not involve the amplification is adopted, in which a yield failure governs. The equations in the ABS Buckling Guide are: For tubular members: 5.0212121)/(11)/(111+EzabzmzCBzEyabymyCByCAaCC 1 when a/CA 0.15 5.022211+CBzbzCBybyCAa 1 when a/CA 0.15 For rolled or fabricated p
33、late sections: )/(11)/(1112121 EzabzmzCBzEyabymyCByCAaCC+ 1 when a/CA 0.15 CBzbzCBybyCAa221+ 1 when a/CA 0.15 The comparisons between the ABS Buckling Guide for combined column buckling and bending and test data for tubular members are presented in Section C2, Figure 7. For overall buckling, 49 test
34、 results exist extracted from Prion and Birkemoe31, Kiziltug et al22, Ellis32, Wagner et al33, Kato and Akiyama27and Smith et al18: 34 results were rejected on the grounds of being too thin and inadequately documented. For local buckling, 19 data have been extracted from Kiziltug et al22and Prion an
35、d Birkemoe31: no data were rejected. The mean and COV of modeling uncertainty of various codes are presented in Section C2, Table 5. TABLE 5 Mean/COV of Modeling Uncertainty for Beam-Column Buckling ABS MODU Rules API RP 2A WSD ABS Buckling Guide Mean 1.0811 1.0439 1.0180 COV 10.03% 9.52% 10.84% Sec
36、tion C2 Individual Structural Members ABSCOMMENTARY ON THE GUIDE FOR BUCKLING & ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES .2005 15 FIGURE 7 Beam-Column Buckling for Tubular Members Modeling uncertaintyNumberofspecimens0123456789100.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5ABS MODU RulesAPI RP 2A WSDA
37、BS Buckling GuideC7 Tubular Members Subjected to Combined Loads with Hydrostatic Pressure Unstiffened tubular members under external hydrostatic pressure are subjected to elastic or inelastic local buckling of the shell wall between restraints. Once initiated, the collapse will tend to flatten the m
38、ember from one end to the other. Similarly, ring-stiffened members are subject to local buckling of the shell wall between rings. The shell buckles between the rings, while the rings remain essentially circular. Consequently, it is desirable to provide rings with sufficient reserve strength to preve
39、nt general instability. Strength design interaction equations for the cases in which a tubular member is subjected to axial tension or compression, and/or bending combined with external hydrostatic pressure have been proposed in API RP 2A WSD. The hoop compression is not explicitly included in the a
40、nalysis, but its effect on member design is considered within the design interaction equations. The hoop collapse design check must be satisfied first. The method described is based on the explicit application of the capped-end axial compression, which allows for a more precise redistribution of the
41、 capped-end load based on the relative stiffness of the braces at a node. A collection of the test data and additional comparisons of the design equations to test data can be found in Miller and Salikis34. C7.1 Axial Tension, Bending Moment and Hydrostatic Pressure The member net axial stress is the
42、 calculated value, tc, since the effect of the capped-end axial compression is explicitly included in the design analysis. Therefore, the calculated axial tensile stress, tc, can be used directly in the cross-sectional strength check. Test data for tubular members subjected to combined axial tension
43、, bending, and hydrostatic pressure can be found in Miller et al35. C7.3 Axial Compression, Bending Moment and Hydrostatic Pressure In this method, the calculated axial stress, ac, is the net axial compressive stress of the member since the capped-end axial compression is included in the design anal
44、ysis. For the stability check, the axial compression to be used with the equation is the component that is in addition to the pure hydrostatic pressure condition (see Section C2, Figure 8). Therefore, the capped-end axial compression is subtracted from the net axial compressive stress. For the stren
45、gth check, the net axial compressive stress is used. In addition, the cross-section elastic buckling criterion needs to be satisfied. The comparison between the ABS Buckling Guide and test data for tubular members subjected to combined axial compression, bending, and hydrostatic pressure can be foun
46、d in Loh36,37. Section C2 Individual Structural Members 16 ABSCOMMENTARY ON THE GUIDE FOR BUCKLING & ULTIMATE STRENGTH ASSESSMENT FOR OFFSHORE STRUCTURES .2005 FIGURE 8 Capped-end Action Arising from Hydrostatic Pressure wave, current and wind loadsDead and live loadsCapped end acac= 0.50.5ac -0.5+
47、ac -0.5Tubular members subjected to combined compression, bending moment and hydrostatic pressure are to satisfy the following proposed equations at all cross-sections along their length. When ac/CA 0.15 0.15 and ac 0.5: )/()5.0(115.0121 EacbmCBCAacC+ 1 When ac/CA 0.15: CBbzbyCAac2221+ 1 When x 0.51
48、C: and Cx 0.5C, the following local buckling state limit should also be satisfied: 2111)5.0(5.0+CCCxCx 1 The comparison of the characteristic buckling strength from the above interaction equation to test data is shown in Section C2, Figure 9. This is based on test data for D/t 120 from Das (2000)38and includes both stiffened and unstiffened members under hydrostatic pressure or combined loadings. The mean and COV of the modeling uncertainty are 1.0299 and 13.28% respectively. Section C2 Individual Structural Members ABSCOMMENTARY ON THE GUIDE
