Prédiction de l'érosion de cavitation: approche énergétique
Predicting the rate and location of cavitation erosion is a complex problem that motivated an important amount of basic and applied research in the fields of hydrodynamics, mechanical science and metallurgy. Using an original approach, the so-called energetical approach, we proposed a prediction model, through the erosive power term, in an attempt to conciliate these various aspects of the cavitation erosion phenomenon. Our attention was also devoted to defining a transposition procedure in order to correctly apply this model in practical fields such as model tests. The energetical approach is based on the knowledge of the energy spectrum associated with a leading edge cavitation, also known as sheet cavitation. The collapse of vapour cavities at the rear part of the attached cavity is known to be sufficiently violent to produce damage on most types of materials, even on very high strength ones such as stellite or stainless steel. However, the experimental determination of such a spectrum has been a matter of great difficulty. As a consequence, no prediction model has received enough consensus up to now to be successfully applied. We established this energy spectrum on the basis of large experimental work performed on a hydrofoil mounted in the IMHEF-LMH high speed cavitation tunnel. We aimed at the determination of the fundamental components of a cavity potential energy : its volume and the driving pressure forcing it to collapse. The production rate of the cavities completes the set of basic parameters needed to completely define the fluid energy spectrum. We first measured the wall pressure field over the profile. This allowed us to have a measure of the overpressure in the downstream closure part of the leading edge cavity. Measuring the travelling time of vapour vortices through a laser beam allowed us to determine their individual main size. Then, we used the continuous wavelet transform to determine their shedding frequency according to this size. On a statistical basis, we found a Strouhal relationship between their main dimension, their shedding rate and the mean flow velocity. This relationship applies to both the steady and unsteady states of the cavitation behaviour. However, in this latter situation, two specific sizes were also found that are essentially defined by the main cavity length. In this case, their production rate is rather ruled by the well-known Strouhal number based on the cavity length and on the flow velocity. We introduced an original visualisation technique, called stereo-tomography, in order to take a measure of the volume of vapour vortices shed by the main cavity. Our analysis of the volume distribution and of the vapour structures morphology led us to consider that all types of vapour vortices could be seen as equivalent to spherical volumes. The characteristic volumes related to the unsteady behaviour are defined by the knowledge of the main cavity length. According to these results, we were able to define the energy spectrum as a function of the main hydrodynamic parameters.We then derived the erosive power term, taking into account the minimal damaging energy, which is a parameter depending only on the mechanical properties of the material. Moreover, we introduced the erosive efficiency in order to quantify the erosiveness of a cavitation development considering the above energy threshold. Finally, we suggested a simple model predicting the location of erosion, as a function of the material energy threshold and of the flow parameters. Considering the transposition side of the erosion problem, we introduced the erosive power coefficient as a scaling factor for the erosive potential between geometrically similar flows. The comparison between the fluid and the deformation energy spectra, carried out from former erosion tests, showed a remarkable proportionality relationship between both. This relationship is defined by the collapsing efficiency, which we found to be constant over the considered range of energies. We then managed to establish a link between the fluid and the material being damaged. This result claims the merits of the energetical approach and ensures its validity.