Analysis of different strategies for the prediction of hydro-abrasive wear in Pelton turbine buckets based on Computational Fluid Dynamics (CFD) simulations

The use of Computational Fluid Dynamics (CFD) techniques for predicting wear of materials in sediment-laden environments has been well established for long time. CFD simulations provide distributed information at the local scale, and, unlike laboratory testing, do not suffer from limitations in size and complexity of the components to be investigated. Meantime, the multi-scale and multi-physics nature of the wear processes entails approximations, simplifications and assumptions in the simulation models, which results in errors and uncertainties. The proper identification and management of the sources of uncertainty in CFD-based wear prediction is a key direction of our research, with the goal of providing guidelines to guarantee that a simulation set up is capable to fulfill the specific engineering requirements. Under high-head, Pelton turbines will suffer serious abrasive erosion after continuous operation, posing a threat to safety and stability of the unit. The water-air-sediments multiphase flow and the periodic rotation of the runner must be accounted for to numerically predict the erosion behavior of Pelton turbines, and the complexity of the system produces significant uncertainties. The proper handling of such complexity is necessary to obtain accurate predictions and, even earlier, make the simulations practically feasible. In this work, we will compare different strategies for the estimation of wear in Pelton turbine buckets. Specifically, we will assess the engineering effectiveness of three different set ups, consisting of a single fixed bucket, a single rotating bucket, and three subsequent rotating buckets. We will focus on both the quality of the information that each set up can provide, and also their computational burden, to provide readers with criteria to choose the most effective option, depending on their specific needs. We will focus our attention not only on the erosion output variables, but also on the flow and sediment dynamic related parameters which mostly affect wear.