Article: Wake Wash Prediction

From Mckwiki

Jump to: navigation, search

Wake Wash Prediction

Target: 1500 Words

Moving ships generate waves called “wake wash” which may be harmful to the shoreline or to other users of the waterway. Estimating the impact of these waves and controlling their formation requires the ability to predict their characteristics. Ship wake wash prediction is one component of minimizing a ship’s environmental impact.

Wake Wash Characteristics

A ship moving through the water generates a series of waves. These waves are the result of the ship displacing the water out of the way for it to pass. The waves represent a pressure field in the water, generated by pressures upon the body surface of the ship. The energy represented in this pressure field is a large part of the resistance of the ship.

The details of wave generation are complex, and have formed a field of study for over 300 years, from at least the time of Lord Kelvin.

Water waves are characterized by… Ships generate water waves having a characteristic feather pattern …. The ship’s waves are not single waves, but are actually a collection of many waves of different frequencies, and originating from different points along the ship’s hull. In the far field, when the ship is far enough away that it may be represented by a single point in space, the wake wash field may be adequately described by a spectrum.

Wake wash propagation

As the waves move away from the ship they undergo transformations and changes. In infinitely deep water this transformation is due to the fact that waves propagate at different speeds, depending upon their frequency. Further, since there is orbital motion of the water particles in an ocean wave, this means that there is a dissipative mechanism which will slowly decay the energy in the wave.

In water of finite depth there are also complex influences from the seabed upon the waves. The most common of these is an effect upon the speed of propagation that causes all the waves to arrive at the same time, which gives rise to the experience of breaking waves on a beach.

Wake wash prediction

The importance of this from the standpoint of wake wash prediction is the fact that the waves that arrive at the shoreline do not have the same characteristics as the waves close to the ship. Thus there are at least two types of prediction needed: Predicting the wave created by the ship, and predicting the waves experienced at the shore.

The most common technique is to predict the waves created at the ship, and then apply a second wave propagation model to predict the transformation of these waves as they reach the shore. This results in a general-purpose model of the wave associated with the ship, and a site-specific model of the wave transformation associated with any specific shoreline location.

Prediction by Physical Model

The ship generated waves may be predicted in several ways. The most accurate is of course not to predict the waves per se, but instead to measure them directly via full-scale tests of the ship itself. This method however requires the availability of the ship and is thus not useful during design or other early investigative phases.

The most accurate small-scale or predictive technique is to build a model of the ship and test it in a ship model testing tank. The test will be run at a range of speeds and loading conditions representative of all possible operations of the ship, and the ship-generated wake wash wave train is measured. The wave characteristics can be directly scaled to full scale by well understood principles.

The drawback to the model testing method is its expense, associated with the construction of the physical model and the employment of an expensive testing facility.

Prediction by Analog

During early investigative phases, when the ship design is not firmly decided, it is desirable to have a numerical predictive method, rather than the two physical predictive methods described above. For this case there are two that are predominant.

The first method is the use of Computational Fluid Dynamics (CFD). CFD constructs a numerical analog of the ship – a digital ship in a digital sea. CFD methods on modern computers may be very fast and of high resolution, and are excellent practical tools.

The drawback to CFD is that the actual mathematics of fluid flow (the Navier Stokes Equations) cannot be solved. The CFD tool therefore uses some simplification or subset of these equations, thus creating an analog of the fluid flow, but not a fully accurate solution.

The simplest CFD tools (which despite being called simple are still very powerful and complex tools) use the principle of Potential Flow. Potential flow is generally accurate for representing flows that are incompressible, irrotational, and inviscid. These restrictions are a good model of ship waves in the far field, in deep water. In the ship near field however there is a substantial role of viscosity which potential flow will not capture. This viscosity can lead to froth, breaking waves in the bow or stern, and other phenomena that will not be accurately represented in the CFD.

At the shoreline a potential flow CFD model will again suffer because it is not able to correctly represent the viscosity experienced along the seabed.

In response to these shortcomings of potential flow models, some CFD tools use Reynolds Average Navier Stokes (RANS) models. These models include the effect of viscosity in a broadly distributed manner. The RANS model will capture large scale viscous effects such as the ship boundary layer, but does not always accurately predict small-scale effects such as a breaking bow or stern wave.

Unfortunately, when using CFD there is no simple way of knowing whether the model is likely to be missing some part of the phenomenon, and thus knowing whether the wake wash prediction is likely to be wrong or not. The second analog method is to rely on measurements of similar ships. Many ship designers, builders, and owners will maintain libraries of measured wake wash characteristics. Based on these characteristics it may be possible predict the wake wash characteristics of a new untested vessel, by assuming various similarities with a parent ship. There is no accepted technique for this type of prediction by analogy, nor is there any accepted way of estimating the accuracy of the method a priori.

Personal tools