Numerical Investigations on the Use of Roughness Patterns for Boundary-Layer Flow Control

点击翻译按钮获取中文摘要

In the present proposal investigations of new means for boundary-layer flow control are intended, based on small bristle-like hair structures attached to a wall in specific patterns. The main purpose of this proposal consists of obtaining knowledge about those wall patterns which are most usefully for technical applications.As already shown in preliminary work performed in DFG grant "Investigation of grid effects on surfaces covered with sensor hairs for selective detection of characteristic flow-field patterns (coincidence sensor)" thin hairs may significantly influence the local flow field. Despite this, we were able to show that the signal of an individual sensor is not inadmissibly falsified by this. Only if a second sensor is placed in the vicinity of another one, position-dependent variations of the second sensor's signal output appear which bear new and exciting application aspects. Results of our previous work on the flow around wall-mounted finite-length cylinders in single and tandem configuration will be used for validation of a new-to-implement modeling of bristle-like structures using body forces in the Navier-Stokes equations. After an additional validation for unsteady flow conditions, the numerical method will be used to perform direct numerical simulations based on the complete Navier-Stokes equations with modeled roughness elements. The distances between individual elements will be modified in such that their individual contributions intensify in the most appropriate way.An easily imaginable realization of this idea is an arrangement in rows parallel to the main flow direction with a certain distance between the rows and a certain length or height of the structures such that these hair structures exert a similar effect on the flow as already-known "riblets" in a turbulent boundary layer. Related questions will be studied in the present project using numerical investigations. However, the present approach is not restricted to riblets in turbulent flow, as other patterns for control of a boundary layer over a wall exist as well, like, e.g., so-called "zig-zag or pimple stripes" for provoking laminar-turbulent transition or delaying flow separation. For investigations of this problem, the impact of relevant roughness-element patterns on laminar-turbulent transition in a flat-plate boundary layer will be studied via numerical simulation. Results will be compared with data for critical Reynolds numbers found in literature. Additionally, the influence of roughness patterns on disturbance growth in a laminar boundary layer and hence on the transitions position will be quantified as well.