GPU-accelerated numerical analysis of jet interference in a six-jet Pelton turbine using Finite Volume Particle Method

Abstract

The Pelton turbine is an impulse turbine typically installed for high head hydroelectric power plants. For a site with a rated head H and discharge Q, a higher specific speed turbine results in a more compact generating unit with reduced manufacturing costs but requires a larger number of jets. However, by increasing the number of jets and specific speed, the water jets tend to interfere, creating a significant energy loss. In the present research, the interaction between two adjacent jets in a six-jet Pelton runner is simulated using a GPU-accelerated particle-based in-house solver based on the 3-D Finite Volume Particle Method (FVPM). The numerical simulations are performed at eight operating points ranging fromN/NBEP=0.89 to N/NBEP=1.31,where N is the runner rotational speed, and BEP is the Best Efficiency Point. The torque and efficiency trends, as well as the speed range in which the jets interfere, are well-captured, which provides confidence in the use of the numerical simulations for the design optimization of Pelton turbines. The simulations, in particular, evidence a significant torque and efficiency drop at high rotational speeds, due to jet interference. Furthermore, jet disturbance yields load fluctuations at rotational speeds both lower and higher than the NBEP, which is likely to amplify fatigue damage. Both phenomena are worth considering that in the design process of a Pelton machine.