Mechanical Engineering - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemDrag reduction via manipulation of large-scale coherent structures in a high Reynolds-number turbulent boundary layer( 2018)The large-scale coherent structures in the outer region of a high-Reynolds-number turbulent boundary layer (TBL) have been shown to be highly energetic with an influence that extends down to the wall and thus directly affects the skin friction. This thesis is concerned with a drag reduction strategy that involves targeting these large-scale motions. The notion of high-Reynolds-number TBL refers to the case where an outer site in the spectrogram of the streamwise velocity fluctuations emerges, which is a manifestation of the energy of the large-scale structures. The largest length of the manipulated large-scale motions in this study measures 10δ in the streamwise direction (where δ is the boundary layer thickness). The characteristic Reynolds number of a TBL is considered to be high, as an outer site in the spectrogram of the streamwise velocity fluctuations emerges. The current experimental study investigates a feedforward control scheme in which the large-scale motions and very large-scale motions of a zero-pressure-gradient TBL at an approximate friction Reynolds number of 14400 were manipulated selectively. An array of nine wall-shear stress sensors—0.07δ apart in the spanwise direction—was utilized to measure the wall-shear stress fluctuations. The wall signature of the large-scale structures was resolved in real-time from the fluctuating signal of each individual wall-shear stress sensor. At 1.6δ downstream of the sensing point, an array of nine rectangular wall-normal jets was designated, each aligned in the streamwise direction with a corresponding wall-shear stress sensor, forming nine sensor-actuator pairs. On/off wall-normal jet airflows through the rectangular planes provided the actuation, and the penetration height reached the upper-bound of the log-region. Large-scale structures possess bilateral characteristics; their instantaneous streamwise velocity are either higher or lower than the mean streamwise velocity at each wall-normal height. These high- and low-speed regions are accompanied by respective down- and up-ward wall-normal velocity components. In a conditional sense, this results in a manifestation of counter-rotating roll modes in the spanwise–wall-normal plane. Therefore, the wall-normal jet actuators were programmed to be synchronized with either the high- or low-speed regions. As the wall-normal jet actuation was synchronized with the high-speed events, it was implicitly synchronized with the down-wash sections of the counter-rotating roll modes. This led to an opposition mechanism of the control scheme (opposing control scheme), and the intensity of the high- and low-speed events was reduced. A maximum reduction of 3.2% in the mean wall-shear stress was measured at 1.6δ downstream of the actuators. The opposite occurred when the wall-normal jet actuation was synchronized with the low-speed events. For this type of manipulation, the actuation was implicitly synchronized with the up-wash sections of the counter-rotating roll modes. This led to a reinforcing control mechanism (reinforcing control scheme) with regard to the intensity of the roll modes. The energetic high- and low-speed events were also enhanced. However, a maximum mean wall-shear stress reduction of 1.2% was measured at 1.6δ downstream of the actuators. It can be concluded that the detrimental top-down influence of the more energetic large-scale structures was overpowered by the beneficial influence of the streamwise momentum deficit downstream from the actuators. The streamwise momentum deficit is the inevitable by-product of the introduction of wall-normal jet into cross flow. A third control scheme investigated actuation that was synchronized with neither the high- nor the low-speed regions. Hence, there was no underlying control logic, and the results served as a baseline case. A 2.4% maximum reduction in the mean wall-shear stress was measured at 1.6δ downstream of the actuators. No behavioral change in the energy of the large-scale structures was observed. Thus, the entire amount of the associated skin friction reduction can be attributed to the generated streamwise momentum deficit downstream of the actuators. In summary, the experimental results support the conjecture of this thesis. That is, the energetic large-scale structures in the outer-region of a high-Reynolds-number TBL can, indeed, be decreased, and this results in their reduced top-down influence on the wall. Reduction in the large-scale component of the energy of the wall-shear stress fluctuations together with the mean wall-shear stress reduction is observed. If wall-normal jet airflows are used as the large-scale forcing, they ought to be synchronized with the high-speed events in order to obtain the above-mentioned results.