Blind disturbance separation and identification in a transitional boundary layer using minimal sensing

I Gluzman,J Cohen,Y Oshman

doi:10.1017/jfm.2021.733

Abstract

A novel approach is presented for identifying disturbance sources in wall-bounded shear flows. The underlying approach models the flow state, as measured by sensors embedded in the flow, as a mixture of disturbance sources. The degenerate unmixing estimation technique is adopted as a blind source separation technique to recover the separate sources and their unknown mixing process. The efficiency of this approach stems from its ability to isolate any, a priori unknown, number of sources, using two sensors only. Furthermore, by adding a single additional sensor, the method is expanded to also determine the propagation velocity vector of each of the isolated sources, based on sensor readings from three sensors appropriately located in the flow field. Theoretical guidelines for locating the sensors are provided. The power of the method is demonstrated via computer simulations and wind-tunnel experiments. The numerical study considers disturbances comprising discrete Tollmien–Schlichting waves and wave packets. Linear stability theory is used to model source mixtures acquired by sensors placed in a Blasius boundary layer. The experimental study investigates the flow over a flat plate, with hot wires as sensors, and a loudspeaker and plasma actuators as source generators. Based on numerical and experimental demonstrations, it is believed that the new approach should prove useful in various applications, including active control of boundary layer transition from laminar to turbulent flow.

Highlights

In flow control applications, model-based and rigorous design of controllers, using both experimental and simulation data, yields important insights into physical fluid systems for effective transition control
Following the new identification and isolation paradigm recently introduced by the authors (Gluzman, Oshman & Cohen 2020), in this paper we employ an alternative modelling approach that focuses on the physical disturbances that affect the flow field, such as those generated by sound or vorticity in the free stream, or by acoustic, mechanical or electrical actuators
This paper presents a novel approach for the isolation and identification of disturbances in a boundary layer

Summary

Introduction

Model-based and rigorous design of controllers, using both experimental and simulation data, yields important insights into physical fluid systems for effective transition control. As proposed in this work, is a process aiming at uniquely identifying the disturbance sources and the way they interact Based on this process, one may relate the individual sources with distinct physical disturbance generators, and, gain useful information about the disturbed flow field. One may relate the individual sources with distinct physical disturbance generators, and, gain useful information about the disturbed flow field This information can prove vital in, for example, the design of closed-loop strategies for the control of transitional boundary layers. This new approach may provide new insights and perspectives on the way we study transitional flows

Objectives

Methods

Conclusion