Monitoring cavitation erosion: damage prediction based on ultrasonic measurements and digital twins

Context

Cavitation occurs in a flow that is initially liquid when the static pressure reaches the vapour pressure of the liquid, due to a drop in pressure or a local increase in flow velocity. This phenomenon then causes a phase change: part of the liquid water turns into vapour. The vapour structures thus formed are carried by the flow towards areas of higher pressure where they implode violently. These implosions generate pressure fluctuations, noise and vibrations. When they occur near surfaces and with sufficient intensity, they can damage materials, leading to a gradual loss of material, known as cavitation erosion.

To date, there is no fully validated method for predicting this phenomenon or for accurately measuring the rate of wear caused by cavitation. There are still many scientific challenges in fully understanding and characterising the associated physical mechanisms, with the purpose of ensuring long-term reliability of production tools.

Given its extensive fleet of hydraulic components (pumps, hydraulic turbines, valves, diaphragms), EDF Group remains heavily affected by this wear mechanism, both in terms of design and in terms of operation and maintenance.

The industrial objective is to develop a monitoring technique based on in-situ measurements and digital twins, enabling:

• to identify the presence of cavitating flow,
• to estimate the aggressiveness of the flow and machine wear,
• to define an appropriate operating strategy,
• to simulate cavitation damage scenarios (particularly in the context of hydroelectric power generation).

The methodology will be based on a non-intrusive ultrasonic method for detecting cavitation, using acoustic signal processing [1–3]. The experimental results obtained will be combined with a theoretical model enabling the prediction of equipment damage and service life, with the goal of optimising their operation and maintenance [4-5], in the form of a Digital Twin [6]. 

Objective and expected results

The primary objective of this thesis is to propose a method for detecting the presence of cavitation and to provide a characteristic measure of the ‘cavitation intensity’ (i.e., the aggressiveness) of the cavitating flow based on signals measured by ultrasound.

The experimental study will be conducted in two stages:

a) Initially, the focus will be on the scale of a vapour bubble imploding near a solid wall. Test campaigns to analyse the implosion of a laser-generated bubble in a pressurisable tank will be carried out in collaboration with various partners. This will enable to combine ultrasonic measurements with tracer analyses on samples of different materials, hydrophone measurements, pressure sensor readings and/or visualisations.

b) The method will also be applied to characterise cavitating flows within various test loops (in fixed geometry and rotating machinery) and during in-situ tests.

The methodology for testing, post-processing and modelling developed during the thesis will be combined with the development and use of a Digital Twin. This type of approach makes it possible to simulate a virtual replica of a given behaviour.

References

[1] Nati M., Digulescu A., Ioana C., Badina C., Fortes Patella R., Maruzwski P., 2025, Ultrasonic detection of erosive cavitation in hydraulic turbines, Proc. IAHR WG 2025 conference, Brno, October 1-3, 2025

[2] Badina C., Ernst O., Maruzewsk P., Ioana C., Fortes-Patella R., 2024, Non-intrusive monitoring of erosive cavitation, Proc. Conference Hydro 2024, 8-20 November 2024 Messe Congress Graz (MCG), Austria

[3] Digulescu A., Ioana C., Serbanescu A. 2019, Phase diagram-based sensing with adaptive waveform design and recurrent states quantification for the instantaneous frequency law tracking. Sensors (Basel). 2019 May 28;19(11):2434. doi: 10.3390/s19112434. PMID: 31141950; PMCID: PMC6603687

[4] Fortes Patella R., Challier G., Reboud J.L., Archer A., 2013, Energy balance in cavitation erosion: from bubble collapse to indentation of material surface, Journal of Fluids Engineering - Transactions of ASME, Vol. 135 / 011303-1 to 11

[5] Fortes Patella R., Choffat T., Reboud J.L., Archer A., 2013, Mass loss simulation in cavitation erosion: fatigue criterion approach, Wear, Reference: WEA100623, http://dx.doi.org/10.1016/j.wear.2013.01.118.

[6] A. Alonso, G. Robert et G. Besançon, A physics-based multi-regime approach for estimation of head losses in operating hydropower plants, Journal of Process Control, August 2025

The PHD will take place at Gipsa-lab, jointly with LEGI, within Univ Grenoble Alpes, in collaboration with EDF group.

Gipsa-lab is one of the major academic research groups in Systems and Signals in France, with about 400 people, including around 40% of PHD students (https://www.gipsa-lab.grenoble-inp.fr/en).

LEGI is a significant research group in hydraulics, carrying out a wide range of activities with a common ground: fluid mechanics and related transport phenomena (https://www.legi.grenoble-inp.fr).

EDF is one of the leading groups worldwide in electricty production, transmission and distribution, supplying energy and related services to nearly 40 million client sites around the world (https://www.edf.fr/en/the-edf-group).

With a Master degree, or equivalent, in Signal Processing, Control Engineering or Fluid Mechanics, candidates should have skills in physical modellng, signal and data processing, dynamical systems and learning, as well as simulation. Some taste in experimentation is also expected, in addition to good communication abilities.