No abstract

For a certain operating point of a horizontal shaft bulb turbine (i.e. volume flow, net head, blade angle, guide vane angle) the efficiency for different pressure levels (i.e. different Thoma-coefficient σ) is calculated using a commercial Computational Fluid Dynamics (CFD-)-code including two-phase flow and a cavitation model. The results are compared with experimental results achieved at a closed loop test rig for model turbines.The comparison of the experimentally and numerically obtained efficiency and the visual impression of the cavitation show a good agreement. Especially the drop in efficiency is calculated with satisfying accuracy. This drop in efficiency in combination with the visual impression is of high practical importance since it contributes to determine the admissible cavitation in a bulb-turbine. It is seen that the incipient cavitation in Kaplan type turbines has no major importance in determing this admissible amount of cavitation.

No abstract

In this paper a method for detecting and furthermore estimating the intensity of cavitation occurrences in hydraulic turbines is presented. The method relies on analyzing high frequency signals with a convolutional neural network (CNN). The CNN is trained in an adversarial manner in order to get more robust results. After successful training the obtained network is modified in such a way, that it is possible to obtain estimations of the intensity. For evaluation purposes a separate dataset is investigated.