Optimizing MIMO control for fluidic pinball using machine learning
Abstract
We are looking for wake stabilization in a multi input multi output (MIMO) configuration.
The wake results from an obstacle made by three cylinders in an incoming flow. The means
of action are the cylinders rotation and the output is the velocity taken downstream. Previous
studies have shown that high and low frequency forcing stabilize the wake, revealing the nonlinear
interactions. Linear control being not applicable in our case we are looking for an optimal control
law regarding drag reduction using genetic programming, a model free MLC approach. Genetic
programming can explore a broad spectrum of laws, exploiting the nonlinearities, ranging from
open loop control to closed loop control.