Essay: BigLift Shipping

BigLift Shipping, member of the Spliethoff group is one of the world’s leading heavy lift ship management companies, specialised in worldwide ocean transportation of heavy lift and project cargoes, with a history dating back to 1973. BigLift strives for innovation, excellence and operational reliability, adhering to high Health, Safety, Environment and Quality standards and operating to strict time schedules.’A great variety of heavy and over-sized cargoes for long-standing clients in the oil & gas, mining and power generating industries, is carried worldwide by a modern fleet of 15 specialised heavy lift vessels. All the vessels we manage are equipped with their own gear with lifting capacities up to 1,800 [mt] and some have a ro-ro capability for loads as high as 2,500 [mt].
A team of dedicated, highly skilled professionals, with years of experience and the mindset to think creatively, enables us to offer innovative and safe solutions for clients’ technically and logistically complex requirements. Careful planning, engineering, coordination and supervision to ensure safe transportation are all in a day’s work. (Biglift, 2015)
Rotterdam Mainport University of Applied Sciences
Rotterdam Mainport University (RMU) is the result of a joint venture between the STC Group and the Hogeschool Rotterdam. Using the hands-on experience from professional practice is essential in all programmes and it enables us to adequately respond to the needs arising in professional practice in the biggest port in Europe. RMU offers bachelor’s degrees in Mainport related topics such as:
Transport and Logistics,
Ship building and
Chemical Process Engineering.
Contractor
Daan van den Wildenberg. Graduate student Bachelor Shipbuilding / Maritime Technology.
Problem definition
Background
Prior to the actual transportation of heavy cargo, BigLift’s Engineering department carefully analyses the ship’s structural integrity with regard to dynamic loads for sea-going conditions. Usually the current practice is to make a simplification of the actual load cases that occur, for which BigLift has developed and uses an in-house Excel-based engineering toolbox. Specifically for tanktop loading BigLift’s current approach is limited and should be revised to a more sophisticated method to calculate stress levels in the tanktop structure.
Problem description
Currently for tanktop strength checks BigLift makes an Excel-based buckling analysis of plate fields that directly support cargo. But in this approach BigLift neglects any shear forces and bending moments transferred to the surrounding plate fields, stiffeners and main girders in the tanktop structure. This approach is not a correct representation of the actual situation and might be conservative as all energy is absorbed by the supporting plate field only.
This incorrect modelling and loss of accuracy in the actual approach could be partially solved by using FEM in structural assessments. This allows to analyse the structure as a 3D model, instead of the 2D model as plate field in the actual approach. Since no full-scale stress and deflection benchmark measurement dataset is available, FEM-modelling is expected to give the most realistic results for structural assessments.
The described problems of incorrect modelling and loss of accuracy mainly apply to tanktop load cases.
‘
Objective
BigLift is interested in gaining a better insight in the limitations in application of the actual Excel-based approach and the differences in outcome compared to FEM-modelling. A new methodology based on FEM should be developed for structural assessments. This should primarily be aimed at tanktop structural assessments.
This in-depth FEM-based methodology evaluates stress and buckling effects of an arbitrary load at a location and its surroundings on a tanktop structure, taking into account multiple criteria such as buckling, shear and bending stresses. For more general use this methodology could be integrate a FEMAP model and analysis in an Excel-based GUI.
The main objective can be achieved with the following sub-tasks:
Analysing the current buckling calculations of BigLift of several load cases
Modelling the tanktop construction of the Happy D in FEMAP
Modelling a previously existing load case in FEMAP
Comparing the results of FEMAP with the original calculations currently used by BigLift Shipping (e.g. comparing stress results in plate fields that have direct and non-direct contact with the load)
Setting the parameters in Excel for a load case as input
Programming the link between Excel and FEMAP for import and export of load cases and their results
Research questions
These research questions will be used as support for the sub-tasks.
Which method is currently used by BigLift Shipping?
How can FEM be used to calculate the stresses near the location of a load and its surroundings in the tanktop structure?
Which differences are there between the results from a previous existing load in FEMAP and the current Excel sheet?
Which parameters are necessary for setting up an Excel-based model with the integrated FEM calculation?
Approach
To achieve the main objective of this research, a well thought approach is necessary. The approach is divided in the following:
Problem investigation, current work method
Understanding the current Excel calculations sheets
Reviewing the Excel sheet with the theory of plate buckling
Analysing the load case
Analysing the cargo item on it stowage plan
Analysing the Excel calculations for this load case
Design a FEM methodology for careful analyses of the load case in three steps
Step 1: Modelling a similar buckling case as BigLift uses for the Excel calculation. This is only one plate simply supported on all edges. Comparing the results from the two different methods
Step 2: Modelling a number of stiffed panels with a load applied. Comparing the results with step 1
Step 3: Modelling the complete tanktop structure with the existing load case. Analyse the results with the previous steps
Evaluation of the difference between the current work method and the FEM method
Evaluation of the current Excel calculations and the results from FEMAP
Conclusion which method is more accurate and sophisticated
Preconditions
All the dimensions and load cases are based on the ship ‘Happy D-Type’. In order for careful analyses of the complete tanktop structure during a load, the tanktop model must be accurate according on the design of the ship. The model will be based on the drawings provided by BigLift Shipping of the hull structure. The shape of the load will be based on the drawings provided by the client for this particular voyage. The stowage plan is leading for the support of the cargo on the tanktop. The modelling and calculation will be done in FEMAP, a program from Siemens. After modelling and calculation the results need to be compared with the current buckling calculation of BigLift. This comprising is important for determine which method is more sophisticated.
‘
Load case
The load case for this research is cargo item (a LNG tank) shipped from Aviles ‘ Onslow. This tank is part of a LNG Plant in Onslow Australia. This plant is currently under construction and part of the Wheatstone project. The Wheatstone project will include an onshore facility located 12 kilometres west of Onslow. The foundation project includes two LNG trains with a combined capacity of 8.9 million tonnes per annum (Chevron, 2015).
The cargo item that’s calculated for buckling on the tanktop is a 170.6 [mt] LNG tank with a length of 39.89 [m], with of 8 [m] and 7 [m] high. The tank will be supported by two saddles which are located according to the stowage plan in Appendix I.
‘
Buckling calculation
Theory
The tanktop of a ship is basically a stiffened plate structure. The plating is supported by longitudinal stiffeners. But when the structure is subjected to compressive loads, the behaviour is quite different than under tensile loads. The structure under tensile stress may cause cracks, while compressive stress will result in buckling of the structure. Almost every part in the structure of a ship are sometimes loaded with compressive loads, therefore it is important to analyse the buckling stresses that occur during sea-going conditions, or in this case when adding heavy loads on a tanktop. In Figure 1: Example of plate buckling form is an example show of plate buckling behaviour (Okumoto, Takeda, Mano, & Okada, 2009).
Figure 1: Example of plate buckling form
For practical design purposes BigLift’s approach only checks a plate field without any stiffeners or other construction. Within this approach BigLift assumed that all the energy, forces and stresses only occur in the plate that’s calculated on buckling. BigLift’s method which is explained in 5.2 is based on the equation of Euler. The complete derivation of the Euler buckling formula is descripted in (Okumoto et al., 2009)
These calculations are according to the basic equations of plate buckling with various boundary conditions and different loading combinations.
‘
Buckling calculations are based on the Euler equation of buckling. (Okumoto et al., 2009)
??_E= (k ‘ ??^2 ‘E)/(12 (1- ??^2 ) ) ‘ (t/b)^2 ( 1 )
The formulas given below is equally applicable to either vertically or horizontally stiffened main supporting members, or members utilizing a combination of both systems. (Taggart, Architects, & Engineers, 1980)
(f_c/f_crc )^2+(f_s/f_crs )^2 ‘1.0 ( 2 )
Where
f_c = Calculated maximum compressive stress due to axial compression and bending.
f_s = Calculated average shear stress.
f_crc = Critical buckling stress corresponding to axial compression and bending loads.
f_crs = Critical buckling stress corresponding to pure shear loading.
The critical buckling stresses, f_crc and f_crs may be approximated as follows (Taggart et al., 1980):
f_crc=f_t ( 3 )
f_crs= f_t”3 ( 4 )
Where
f_t= f_e “when ” f_e'(f_y ‘0.75) ( 5 )
f_t= f_y ‘ (1- (3f_y)/(16f_e )) “when ” f_e'(f_y >0.75) ( 6 )
f_e= 1.88 ‘ ’10’^6 ‘ (t/b)^2 ‘k=[kg/’cm’^2] ( 7 )
f_y = Specified minimum yield point of the material or 72 [%] of the specified minimum ultimate strength whichever is less.
t = Thickness of the web plate reduced by 10 [%] as an allowance for corrosion.
b = Depth of plate panel.
‘
K is a function of different types of loading, aspect ratio and boundary conditions. BigLift is using the following conditions for their buckling analyses.
Figure 2: Load condition Compression and Bending
K corresponding to axial Compression and Bending:
Where a’b ‘1.0=4 ( 8 )
Where a’b <1.0= (a/b+b/a )^2 ( 9 ) Figure 3: Load condition Shear stress K corresponding to Shear loading: Where a'b '1.0= '3 [5.34+4 'b/a'^2 ] ( 10 ) Where a'b<1.0= '3 [4+5.34 'b/a'^2 ] ( 11 ) ' Working Method Before the actual shipping of the load BigLift takes several steps. These steps insure the safe loading and transport of the cargo. These steps are descripted below. The Excel-sheets used for calculation are added in Appendix II. Stowage plan: The project manager will setup a stowage plan with notice of other cargo and the construction of the vessel. After completing the stowage plan, the engineering department will calculate the stresses and forces that may occur. Based on the information from the stowage plan and the drawings of the cargo, the calculations will be made. BigLift shipping has creating a direct link between the program that's used for their stowage's plans and Excel. With this link the cargo can be inserted in the Acceleration sheet. Acceleration Sheet This sheet is used by the project manager and the Engineering department of BigLift Shipping. After inserting the cargo, and the details of the weight and dimensions, in the Excel sheet the accelerations are calculated. These accelerations are based on the regulations drafted by DNV (DNV, 2009). BigLift is calculation their load cases on the Pitch + Heave motions and Roll +Heave motions. These motions are the most common and most powerful. But in this approach BigLift calculates the cargo as a box. But in almost every case the shape of the cargo isn't a box. So to reshape the calculations to a more realistic format the Summery Sheet is used. Deck Strength sheet: This sheet is used as summery page for the calculations of a specific load case. On this sheet there is an abstract of: the name of the cargo, the weight, Deck loads, a drawing of the stowage plan and which structural part is checked on the allowable buckling stresses. Summary Sheet: The Summary sheet is used for calculation of the accelerations of the cargo during sea-going conditions. On this sheet the Pitch + heave and Roll + heave forces and stresses are calculated. When the load case is statically indeterminate the Acceleration Sheet is used for calculation the Roll + heave forces and stresses. The maximum force from, Pitch + heave or Roll + heave, that occurs during sea-going conditions is used for the Buckling sheet. ' Distribution Load Sheet: The Distribution Load Sheet is used for calculation forces that occur in transversal direction during sea-going conditions. On this sheet is calculated which saddle is carrying the most load of the cargo. In this case the Saddle on the Aft of the ship is carrying 66.9 [%] of the weight. Therefore the force and stress that occur are distributed on the tanktop true the saddle. The load at the end of the saddle on the maximum roll + heave motion is used on the Buckling Calculation Sheet. Bucking Calculation Sheet: On this sheet there are two ways for calculating the load case on allowable buckling stress: The Compression and Bending and the Shear calculation. There are two ways to approach a load case. First the engineer can calculated the actual load case and make sure that no buckling occurs. Or the engineer can calculate the maximum allowable stress, and then checks if the stress of the load case is below this value. The Compression and Bending The K factor for Compression and Bending needs to be determinate as following: "If " a'b >1 “then ” K= 4 ( 12 )
“If ” a’b <1 "then " K= '[(a'b)+ (b'a)] '^2 ( 13 ) The K factor can be used according the method descripted in (Taggart et al., 1980): f_e= 1.88 ' '10'^6 ' (t/b)^2 'k= [kg/'cm'^2] ( 14 ) This equation is based on the formula of Euler. But the Poisson ratio [??] from the Euler equation used by BigLift Shipping is ' 0.28, therefore the equation ( 14 ) is different from the equation of Timoshenko( 15 ) (Asmus, 2001). ??_cr= 1.9 ' '10'^5 ' (b/l)^2 'k=[N/'mm'^2 ( 15 ) Both formulas are based on the Euler equation but the difference is in the Poisson ratio. The Poisson ratio is the relation between material that is compressed in one direction, and the expansions in the other two directions perpendicular to the direction of compression (Lambe & Whitman, 1969). With equation ( 14 ) the critical buckling stresses corresponding to pure Compression and Bending loading can be calculated. According to the method descripted in (Lewis). "If " f_e'(??_y <0.75 "then " f_t= f_e ) ( 16 ) "If " f_e'(??_y >0.75 “then ” f_t= ??_y )'[1- ((3??_y)'(16f_e ))] ( 17 )
In this load case the is f_e'(??_y >0.75 )so f_t is calculated according to equation ( 17 ). These results are used for calculate f_crc:
f_crs= f_t ( 18 )
Then the Compression area is calculated:
A_s= (a+b) ‘t ( 19 )
The force that is actual causing the Compression and Bending is calculated as:
Q=q ‘b ( 20 )
This force is a distributed load on the structure that’s been analysed. Dividing equations ( 20 ) an ( 21 ) gives the stress that occurs.
f_s= Q’A_s ( 21 )
Shear Calculation
The K factor for shear is different than for Compression and Bending. The K factor for Shear needs to be determinate as following:
“If ” a’b >1 “then ” K= ‘3’ [5.34+4 ‘(b’a) ‘^2 ] ( 22 )
“If ” a’b <1 "then " K= '3' [4+5.34 '(b'a) '^2 ] ( 23 ) The K factor can be used according the method descripted in (Taggart et al., 1980): f_e= 1.88 ' '10'^6 ' (t/b)^2 'k= [kg/'cm'^2] ( 24 ) This equation is based on the formula of Euler. With equation ( 24 ) the critical buckling stresses corresponding to pure shear loading can be calculated. According to the method descripted in (Lewis). "If " f_e'(??_y <0.75 "then " f_t= f_e ) ( 25 ) "If " f_e'(??_y >0.75 “then ” f_t= ??_y )'[1- ((3??_y)'(16f_e ))] ( 26 )
In this load case the is f_e'(??_y >0.75 )so f_t is calculated according to equation ( 26 ). These results are used for calculate f_crc:
f_crs= f_t”3 ( 27 )
Then the shear area is calculated:
A_s=minimum of 2a ‘t or 2b ‘t ( 28 )
‘
The force that is actual causing the shear stress is calculated as:
Q=q ‘b ( 29 )
This force is a distributed load on the structure that’s been analysed. Dividing equations ( 28 ) an ( 29 ) gives the stress that occurs.
f_s= Q’A_s ( 30 )
Conclusion:
In this section the combined calculations of Compression and Bending and Shear are calculating if buckling occurs.
“No Buckling when: ” (f_c’f_crc )^2+'(f_s’f_crs ) ‘^2 <1.00 ( 31 ) "Buckling when: " (f_c'f_crc )^2+(f_s'f_crs )^( 2) >1.00 ( 32 )
Finite Element Method
The Finite Element Method (FEM) is a numerical solution of equations (matrixes). In FEM a construction is divided in a finite number of elements. Every element is connected with each other, these connections are called nodes. With the use of matrix equations, it’s possible to calculate the displacement, forces and stresses of these nodes in certain load cases. By using FEM software it’s possible to divide the model in small nodes. Then the program calculates the displacement of every node, which will result in a very careful analysis.
For the complete derivation of the Finite Element Method (Hofman, 1994).
Chapter 5 describes the current buckling calculation according the equation of Euler. But it neglects the surrounding structure and the method. Therefore a more accurate analysis of the complete tanktop structure is necessary. The choice for these analyses is based on the Finite Element Method.
‘
FEM-based tanktop model
In the FEM-based tanktop analyses the model will be exposed on several loads. The analysis that’s been used is the ‘NX Nastran Static and Buckling’. This analysis calculates the model on the static stresses that occur during the load case and on buckling as well. The program calculates a so called: ‘Eigenvalue’, for the buckling situation. This Eigenvalue is a factor for the number of times that the current load could be multiplied before the construction succumbs under buckling.
Step 1
For analysing a similar calculation as BigLift uses in the Excel-sheets, it’s necessary for modelling a plate in FEMAP with the same conditions as used in the Excel calculation. The FEM-calculation reviews the construction based on the mesh. A mesh is an assembly of element. If the elements are small the calculation is more accurate, but it will also result in longer calculation time. For these analyses the mesh size of 50 [mm] by 50 [mm] is used.
Conditions:
Material Steel – [-]
Load Force per Area on Surface 2 [N/mm2]
Constrains Fixed on the Shell – [-]
Analyses NX Nastran Static and Buckling – [-]
Mesh size – 50 [mm]
This will result in several stresses as result from compressing and bending, shear and of course buckling.
The results of this analyses is shown in figure 212
‘
Step 2
Since the FEM analysing of buckling situations is relative new for BigLift Shipping, verifying the results is difficult. Therefore splitting the modelling and analysing was necessary to insure the accuracies of the model. So after verifying that the previous step resulted in a good analyse, the model could be expanded. So stiffeners and other girders where added and the nodes where connected to each other. This resulted in a very basic and local construction of the tanktop. The footprint of the saddle was added as load case. Different values of the load where added to insure the accuracies of the model in different situations.
Conditions:
Material Steel – [-]
Load Force per Area on Surface 2 [N/mm2]
Constrains Fixed on the Shell – [-]
Analyses NX Nastran Static and Buckling – [-]
Mesh size – 50 [mm]
The results of these analyses are shown in figure 221
Step 3
After step 2, the construction of the tanktop is expended. The construction is modelled from frame 52 till frame 58. This is the construction, which according the Distribution calculation sheet of BigLift Shipping, is exposed at the most forces and stresses during sea-going conditions. The footprint of the load case is modelled as a surface with a very high Young’s modules. BigLift Shipping assumes in there calculation that the load is infinitely rigid. With this assumption the relation between the cargo and the construction of the vessel can be ignored. In this final step the analyses calculates the complete construction that’s been modelled. With this step it’s also possible to review the construction that acutely is exposed on the stresses.
Conditions:
Material Steel – [-]
Load Force per Area on Surface 2 [N/mm2]
Constrains Fixed on the Shell – [-]
Analyses NX Nastran Static and Buckling – [-]
Mesh size – 50 [mm]
‘
Analysing calculation results
‘
Implementation
Conclusion
‘
Recommendations
‘
Bibliography
Books
Asmus, K. (2001). Bijzondere Aspecten van de sterkte van Scheepsconstructies.
Hofman, G. E. (1994). Eindige elementen methode: HB Uitgevers.
Lambe, T. W., & Whitman, R. V. (1969). Soil Mechanics: Wiley.
Lewis, E. V. Principles of Naval Architecture (Second Revision), Volume I – Stability and Strength: Society of Naval Architects and Marine Engineers (SNAME).
Okumoto, Y., Takeda, Y., Mano, M., & Okada, T. (2009). Strength Evaluation. In Y. Okumoto, Y. Takeda, M. Mano & T. Okada (Eds.), Design of Ship Hull Structures (pp. 33-80): Springer Berlin Heidelberg.
Taggart, R., Architects, S. o. N., & Engineers, M. (1980). Ship design and construction: Society of Naval Architects and Marine Engineers.
Internet
Biglift. (2015). from http://www.bigliftshipping.com
Chevron. (2015). Wheatstone project. Retrieved 04-03-2015, from http://www.chevronaustralia.com/our-businesses/wheatstone
Rules and Regulations
Rules for Ships, Det Norske Veritas ?? Ch. 1 Sec. 1 (Det Norske Veritas 2009).