The gap between theory and practice

01 Nov 2004

A paper presented at CIMAC in Kyoto this year was the result of a two-year project to study the interactions between the hull, propulsion shafting and two-stroke diesel engine, focusing on deflections and bearing loads in the propulsion machinery. Daewoo Shipbuilding & Marine Engineering Co. Ltd. (DSME), Republic of Korea, Det Norske Veritas (DNV), Norway, and MAN B&W Diesel A/S Denmark were joint partners in the project which used a DSME-designed VLCC for the project-study.
The 300,000 dwt VLCC is powered by a two-stroke MAN B&W 7S80MC diesel engine, rated at an MCR of 25,480kW at 79 rpm.
Geir Dahler, principal engineer at DNV?s Maritime Technology and Production Centre and one of the authors of the paper says: "The results should be relevant for all kinds of modern VLCCs because they are all similar in design. What you can?t do is apply the conclusions to container vessels or vessels with other hull shapes." However, he added: "This could be considered to be a good basis for ULCCs but I can?t quantify what the difference or impact of applying this data to ULCCs will be but they are more or less scaled up versions of a VLCC."
The R&D project combined analytical investigations of the hull, including calculation models for the propulsion machinery, with a comprehensive measurement programme carried out during sea-trials. Analytical investigations and measurements were used to increase the knowledge of, and to reduce the gap between, theory and practice with regard to shaft behaviour, structural deflections, bearing loads and required design offsets.
The limitations and possibilities involved in using finite element analyses and shaft-beam calculations were explored and uncertain variables, such as draught, tank filling, thermal loads, sterntube bearing offset, engine offset, engine tilt and propeller-induced forces were revealed, and their influences on the bearing loads, engine bearings as well as propeller and intermediate shaft bearings, were determined.
DNV says that while finite element (FE) hull analysis captures the general trend of hull deflections reasonably well, it fails to account for local variations in the curvature of the shaft, which can lead to inaccurate predictions of bearing loads. However, this may be a result of phenomena not accounted for in the analysis, such as thermal variations due to engine running and ambient heating, or in the assumptions made ? for example FE modelling methodology, the equivalent engine-stiffness matrix and equivalent crankshaft model applied ? or may be due to the model being too coarse (grid and stiffness matrix).
The analysis methodology, which combines strain gauge measurements and shaft-beam model calculations, gives calculated bearing loads that agree well with the jacking results for the aftermost bearings, such as the forward sterntube bearing and the intermediate shaft bearing.
The classification society concludes that the simplified crankshaft
model, described by an equivalent diameter and commonly used for shaft alignment calculations, is unable to describe the flexibility of the crankshaft very precisely.
DNV has no plans to carry out similar research work on other hull types for the time being. However, Dahler says that some of what was learned in this project can be applied to other types of vessels. "For example we have learnt a lot on the possibilities and limitations of FE analysis, especially with respect to its limitations such as boundary conditions to be included in the analysis model." He adds: "We have also had the same experience with the different calculations and methodologies for shaft alignment and bearing load calculations ? that is general knowledge that we can use for other ship types."
However, as DNV has already proved that the boundaries for the FE analysis for VLCCs was not the best for the purpose, unless other ships types go through the same process it will not be known how far other ship types differ. Dahler explains: "We discovered the limitations of the FE analysis when we focussed on the details such as figuring out the deflections and bearing loads. The FE analysis gives a fairly good picture of hull deflections in terms of mm, which is good for a huge VLCC. However a variation of the shaft bearing deflection, for example a fraction of a mm, may result in several tonnes of change in bearing loads. But we have learnt from this and we can now document that there are certain limitations in such analysis, especially regarding use of boundary conditions."
Although this project has been completed, DNV has follow up plans but Dahler was unwilling to discuss in what way at the time of going to press as it could be commercially sensitive. "This is not the end of the story ? it has generated a lot of knowledge for the designers and engine manufacturers and has created a lot of beneficial knowledge for DNV to describe the acceptance criteria. That is what DNV is about ? setting the acceptance criteria. It creates greater confidence in what we are doing but it has also raised a lot of new questions, some of which we were not able to deal with at all, so we are doing more but looking at different angles." n

Links to related companies and recent articles ...

DNV GL

view more