Numerical Prediction of Cavitation and Related Nuisances in Marine Propulsion Systems

Book chapter, 2021

Cavitation is a major design constraint for marine propulsion systems, whether it concerns vibrations, noise, performance loss, or erosion. It thus needs to be considered as early as possible in the design process. For an open propeller arrangement, typically a baseline design is chosen without consideration of cavitation by optimisation based on empirical relations derived from systematic series. A more detailed blade design is then developed using potential flow simulation approaches considering cavitation, with a varying degree of reliability and refinement in the cavitation prediction. When used by an experienced blade designer, the behaviour of on-blade sheet cavitation is usually acceptably determined. Based on this, a first estimation of hull pressure pulses can be computed, and by visual inspection of the sheet cavity dynamics the designer can make a judgment on risk of cavitation erosion. For a standard contract, this is often sufficient to develop the final design that is then verified by experiments. A design turn-around time of a week or two is not uncommon.

For vessels with higher requirements, e.g. for cruise ships, research vessels or navy contracts, more effort is needed. Up to now, this has involved model experiments in a cavitation tunnel. With a time frame of a few months for model production and performing the tests combined with considerable cost, this is performed only in a late stage of the design process and if problems are discovered it may delay the delivery. Further, the data possible to extract from an experiment is quite limited and may strongly suffer from scale effects. Pressure pulse levels for the model scale system can be well recorded, while radiated noise is more troublesome in the confined space of a cavitation tunnel; cavitation erosion may be assessed by a paint test. Standard video recordings combined with stroboscopic light are used to determine cavity extent, and for more detailed studies of cavity dynamics, in particular in relation to determining risk of erosion, high speed video is employed. The presence of cavitation prohibits the use of more detailed optical measurement techniques. Scale effects to consider are the shape and dynamics of the wake inflow to the propeller and presence of laminar flow on the blade; both have a huge impact on cavity development.

Only recently, more advanced simulation approaches based on incompressible viscous RANS methods have become mature enough to be incorporated in commercial design procedures. For a marine propulsion system, the large scale separation between the geometrical scale of the ship and propeller and the small scales governing cavitation and tip vortex development limits the use of scale-resolving modelling and prohibits the use of compressible approaches. Still, however, reliable quantitative analysis is difficult to achieve in an industrial work flow where simulation time, and to some extent computational resources, is limited. For a standard propeller around design operating condition, performance predictions using RANS are not more reliable than potential flow methods, even when considering cavitation. However, during off-design conditions, separation or massive cavitation reduces the accuracy of potential flow methods. Also, for propellers with smaller blade area (to increase efficiency) or thicker blades (for ice-classed vessels) potential flow methods loose in reliability and RANS becomes necessary. For pressure pulse predictions, RANS may allow the prediction of the cavitating tip vortex, at least close to the blade, which is not possible in standard potential flow approaches. Although it is a challenge to propagate the tip vortex further downstream, the interaction between the cavitating tip vortex and the on-blade sheet cavity has a significant impact on the lower order pressure pulses and is a strong incentive to promote the usage of RANS. When it comes to cavitation erosion, a few approaches to indicate the risk of erosion have just emerged in research and their reliability and requirements on accuracy in flow predictions are still not proven. The published results are promising and it will be interesting to follow how industrial uptake develops.

This chapter aims to give an overview of numerical prediction of these cavitation issues in both research and in industry, with a discussion of the limitations and possibilities of different methods. A certain emphasis will however be given to state of the art as presented in recent scientific literature as an indication of where the topic is heading. This then includes to some extent performance predictions, i.e. thrust and torque over a range of operating conditions; a more detailed discussion on the determination of the cavitation inception point; current level of accuracy in hull pressure pulse predictions, extending into the related area of radiated noise levels; and finally an examination of proposed methods for assessing the risk cavitation erosion.