Explorative gradient method for active drag reduction of the fluidic pinball and slanted Ahmed body.

Otherwise, the almost complimentary LHS optimum for actuators 2-5 and the one-dimensional optimum of § 5.3 should yield 10 % reduction with Graph $b 1 \approx 1$, Graph $b 2\approx 1$, Graph $b 3\approx 0.13$, Graph $b 4 \approx 1$ and Graph $b 5\approx 1$. The velocity components in the Graph $x$, Graph $y$ and Graph $z$ directions are denoted by Graph $u$, Graph $v$ and Graph $w$, respectively. The actuation velocities Graph $U 1, \ldots, U 5$ are independent parameters; Graph $U 1$ refers to the upper edge of the rear window, Graph $U 3$ to the middle edge and Graph $U 5$ to the lower edge of the vertical base; Graph $U 2$ and Graph $U 4$ correspond to the velocities at the right and left sides of the upper and lower windows, respectively. The optimal actuation reads Graph $b 1=0.7264$, Graph $b 2=0.5508$, Graph $b 3=0.1533$, Graph $b 4=0.6746$, Graph $b 5=0.7716$. In this case, the actuation energy of cases Graph $1D$, Graph $5D$ and Graph $10D$ would correspond to Graph $3.2\,\%$, Graph $3.0\,\%$ and Graph $7.9\,\%$ of the parasitic drag power, respectively. [Extracted from the article]