Figure 3: Rough preliminary qualitative comparison

Linking numerical and experimental environmental  models

Model tests are rather expensive and time consum- ing。 Thus, especially when testing in irregular seas, the test conditions should be well chosen。 Within SinSee numerical investigations were used to iden- tify interesting conditions with respect to resonance phenomena for the latest series of tests。 Numerous simulations in following and head seas were run and evaluated。 Some of the most interesting or typical ex- amples with respect to parametric excitation as well as pure loss of stability were chosen as test cases。 Corresponding model seaways were generated and the test conditions (especially with respect to tim- ing between waves and vessel) designed。

This set-up is regarded as rather promising in or- der to investigate dangerous resonance problems in rough but not necessarily extreme conditions。 Fur- ther tests were run in extreme conditions to extent the test conditions and identify possible gaps in the numerical pre-investigation。

Vice versa - when focussing on validation - regular waves are frequently used for validation purposes, mainly because the comparison is relatively straight forward, especially if the ship’s response reaches a steady state (as far as amplitude and phase shift are concerned) also。 In irregular waves a comparison is more complicated。 Traditionally statistical quanti- ties are compared, but this of course does not de- liver a detailed analysis of possible weaknesses in the numerical tool。

Based on the new test set-up where the wave ele- vation is known at all positions of the tank and at all times during the test it is now possible to extent the validation of tools to more details。 At the same time the already mentioned pre-calculation of tests is possible, and might also prove to be a very trust- enhancing set-up, as here the motions are predicted before the actual test run。

Fig。 3 shows a first preliminary qualitative compari- son of a capsize in a wave group。

First steps towards decision support for operation

The given examples show that a lot of interesting in- formation regarding the vessel’s behaviour in rough conditions is in principle available or could be gen- erated。 At the same time there is an increasing de- mand for operational guidance to support crews in their decision making while sailing in rough condi- tions。 Especially as international guidelines such as the MSC Circ。 707 are found to be insufficient for modern designs。

Figure 4: Polar plots illustrating the danger of sliding of unlashed cars for two different load cases in different significant wave length

Yet again, no standards exist with respect to func- tionality, scope and reliability of such systems。 Based on the results from the numerical design evaluation carried out at FSG a so called ”operational per- formance” booklet was developed to summarize the findings and communicate these to the ship’s crew。 Fig。 4 shows an example extracted from such a man- ual。 In these polar plots the danger of the occurrence of sliding of unlashed cars is presented for two differ- ent load cases and different wave length。

While this information enables the crew to better un- derstand their vessels behaviour in rough conditions and more appropriately plan their voyages and the corresponding necessity for cargo securing, it is not suitable for real time decision support in dangerous situations for three reasons: 1。 only some standard loading conditions are covered in the booklet in order to avoid an information overflow。 2。 It is a booklet, this means the user would have to be extremely  fa-

miliar with the content to find the correct answers fast and reliable while being in the potentially criti- cal situation。 3。 It is well known that it is extremely difficult to impossible for the crew to visually identify the basic characteristics of the seaway with sufficient reliability (especially in confused seas and at night)。