Abstract Run-of-river hydroelectric power stations are a significant source of flexible hydroelectricity production. They are continuously upgraded to enhance efficiency and ensure ecological continuity in rivers. However, the implementation of inclined bar-racks with narrow bar spacing within downstream fish protection systems has been identified as an obstruction at the water intake, causing production losses. Therefore, optimizing this system to minimize its impact on power generation represents a crucial area of study. This paper presents a work on optimizing bar profiles to minimize losses while ensuring mechanical properties. Two parameterized profiles were developed to ensure feasible bar shapes. The optimization was achieved by implementing genetic algorithms, complemented by geometric and mechanical constraints, with the objective to minimize head loss and drag coefficients using two-dimensional (2D) computational fluid dynamics (CFD) simulations. The optimized profiles demonstrated favorable form coefficients and generate low head losses, with a reduction greater than 3.9 for head losses and 4.9 for drag coefficient compared with the classical rectangular profile, depending on the optimized profile considered. These profiles demonstrate improved adaptability to various spacer configurations and better resistance to clogging issues compared with conventional profiles. These profiles offer new shapes that are not typically used for these types of applications. One of the optimized shapes was validated using laboratory measurements in a free-surface hydrodynamic flume, and it was finally compared with other common bar profiles that have already been designed to present low head losses. The selected shape exhibited improved performance compared with the others, with a reduction in the bar-form coefficient exceeding 7% compared with the tadpole profile. The various optimized profiles, better adapted to the spacer technology and designed to prevent pebble clogging issues, will offer alternative possibilities for future installations with the objective of minimizing production losses.
Bon et al. (Sat,) studied this question.