Optimization and sensitivity analysis of active drag reduction of a square-back Ahmed body using machine learning control

Abstract

A machine learning control (MLC) is proposed based on the explorative gradient method (EGM) for the optimization and sensitivity analysis of actuation parameters. This technique is applied to reduce the drag of a square-back Ahmed body at a Reynolds number Re = 1.7 × 105. The MLC system consists of pulsed blowing along the periphery of the base, 25 pressure taps distributed on the vertical base of the body, and an EGM controller for unsupervised searching for the best control law. The parameter search space contains the excitation frequency fe, duty cycle α, and flow rate blowing coefficient Cm. It is demonstrated that the MLC may cut short the searching process significantly, requiring only about 100 test runs and achieving 13% base pressure recovery with a drag reduction of 11%. Extensive flow measurements are performed with and without control to understand the underlying flow physics. The converged control law achieves fluidic boat tailing and, meanwhile, eliminates the wake bistability. Such simultaneous achievements have never been reported before. A machine-learned response model is proposed to link the control parameters with the cost function. A sensitivity analysis based on this model unveils that the control performance is sensitive to fe and α but less so to Cm. The result suggests that a small sacrifice on performance will give a huge return on actuation power saving, which may provide important guidance on future drag reduction studies as well as engineering applications.

Sensitivity analysis of large body of control parameters in machine learning control of a square-back Ahmed body

“…EDSM thus shows improved aerodynamic DR, e.g. 11% by Fan et al [4] and 17% by Li et al [28] for a square-back and a 35-degree Ahmed body, respectively. EDSM was also shown to outperform Monte Carlo sampling and GA [28].…”

“…Artificial intelligence (AI) has shown promising potential in active DR control (e.g. [4][5][6][7]).…”

“…The optimized parameters, including voltage amplitude, the burst frequency and the duty cycle, minimized the reattachment length of a backward-facing step and maximized the wall fluctuating pressure in a few iterations. Fan et al [4] considered three control parameters (i.e. excitation frequency, duty cycle and flow rate blowing coefficient) for all pulsed-jet actuators along the rear periphery of a square-back Ahmed body to maximize the DR based on EDSM.…”

“…A single-object optimization finds an optimal DR but the control efficiency may be low. Fan et al [4] used Taylor expansion around the EDSM-generated optimum. A response model, which can accurately estimate the cost values, was built for the sensitivity analysis.…”

Wing-attached Hatchback Vehicle in Adverse Weather Conditions

“…He observed the tendency to develop separated flow at a higher aspect ratio while a lower aspect ratio results in the reattached flow. Besides this, machine learning techniques is also implemented for square back Ahmed body by Fan et al [20] and the Explorative Gradient Method is used to reduce drag values by 11%. Again, the control law optimization is achieved within 100 test runs that assist to recover base pressure by 13%.…”