COLUMN | Naval architecture: Art, science or salesmanship? [Aft Lines]

Take computational fluid dynamics (CFD) as an example. This is now a tool used by many design houses, owing to its accessibility and affordability.

With each passing year, desktop computers are becoming more and more powerful and they can now utilise these highly complex, memory-hungry programs that once seemed near impossible only one or two decades ago. With hull forms that are now created totally within the digital environment, transferring the hull into these programs is child’s play, right?

So, looking back some 14 years ago to the International Towing Tank Conference (ITTC) symposium on CFD, interesting conclusions were made and highlighted:

"For unconventional ships such as multihulls, planing boats and new-concept hulls, it is a little harder to assess the state of matters due to the difficulty of finding systematic and well-documented studies in the open literature…the prediction error for unconventional ships…with all design details would be somewhat larger than with conventional ships."

This then sounded a warning that the accuracy of using CFD is still suspect and requires further verification, validation, and correlation.

Looking at last year’s (2024) ITTC CFD committee report, has things changed?

Yes, of course. CFD has matured significantly in the last decade, and many ITTC towing tank institutions use them regularly. These institutions have access to such verification and validation from existing models and their tank test data to tweak and modify their CFD models to become more reliable.

Several questions were asked where more than 40 institutions worldwide replied. The most striking consensus was, "We need more measurements to continue to understand how to use CFD reliably in our profession."