Trans RINA, Vol 154, Part C1, Intl J Marine Design, Jan - Jun 2012
the response surface is used for interpolation only, and not extrapolation. Once the RSM has been generated, interpolation is more or less instantaneous (compared with a CFD or sea-keeping calculation which might take from a few minutes to several hours to perform).
As has been previously discussed, the design-space often has many dimensions. For more than two dimensions, it is very difficult to visualise the effect of parameter variation on the performance measures. By fixing all but two parameters and using the response
surface to
interpolate the performance for regular variation of these two parameters, it is possible, however, to generate an iso-parametric surface which can be plotted. For instance the effect on resistance of (say) length and prismatic coefficient can be plotted by fixing all parameters other than length and prismatic coefficient and then allowing these two parameters to vary within the design-space range. Similarly, iso-parametric lines can be produced by fixing all but one of the parameters (for example the effect of beam on stability can be examined by fixing all parameters other than beam). Obviously there are an extremely large number of such surfaces and lines that could be produced. This is because of the infinite number of different combinations of values for the parameters that are fixed. For this reason, the fixed parameters are often chosen for some baseline design; then the effect of varying one or two parameters from their baseline value can be seen by use of the iso-lines and -surfaces.
2.6 The FFW provides many of direct
AUTOMATED OPTIMISATION tools for
evaluation of the performing
optimisation of the parametric model. By using the pre- computed response surface models, the computational time for the optimisation is greatly reduced since the overhead
performance
measures is removed. 2.7
EXAMPLE APPLICATIONS A few design scenarios that the design team may be
faced with can be identified. Each of these demonstrates an application of the proposed methodology at quite different levels of detail.
2.7 (a) Ferry Route Viability
When studying the viability of a potential ferry route, there may be the possibility of a completely free design brief: no restriction on vessel
type, size or number;
economic viability is the chief criterion for a successful design. Key performance measures to be addressed might include: Route operability as a function of passenger comfort and safety; passenger expectation – fast daytime crossing or slower night crossing; required ticket fee structure in comparison with alternative modes of transport; environmental and economic impacts. As an example, the trade-off decision between fuel consumption (estimated from resistance) and customer
acceptance (captured in terms of MSI) might be studied, see for instance [13], which essentially leads to Pareto- optimal sets.
2.7 (b) Naval Patrol Vessel Design
The design of a naval patrol vessel may be relatively free, even to the extent that the type of vessel is not fixed. However, it is likely that the performance will have certain minimum constraints in some areas as well as other areas that will be required to be optimised. For example: total lifecycle cost; downtime; performance criteria: minimum speed; endurance; equipment payload; crew requirements; etc.
2.7 (c) Luxury Motor-Yacht Design
This is the example being used in this paper. Here the vessel type has been determined by the owner as well as a relatively narrow range of primary dimensions. Within the range of vessel size considered, the building cost is largely determined by the outfit rather than small changes
in hull shape and dimension. The
key
performance measures are resistance (more specifically, the attained speed or operating range without bunkering) and passenger comfort in a seaway as well as consideration that certain legislative requirements concerning stability must be met.
3. RESULTS
A design-space exploration was performed for the luxury motor-yacht example. Typical raw results are shown in Figure 6. For further details and results, the interested reader is referred to [14]. These results and associated response surfaces form the basis of the examples shown in the following section.
Figure
6:
Example raw results
of
design-space
exploration, showing strong correlation between Stability performance measure and LPP.
In Figure 6, each dot represents one of the design variants tested. It is interesting to note the correlation of the stability performance measure with the vessel length, with stability decreasing as vessel length increases (this is thought to be due to the fact that the stability was examined over a
fixed range C-22 of displacements ©2012: The Royal Institution of Naval Architects
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64