Font Size: a A A

Suction Control And Its Optimization Of Boundary Layer Separation At Low Reynolds Numbers

Posted on:2015-11-03Degree:DoctorType:Dissertation
Country:ChinaCandidate:W L ZhangFull Text:PDF
GTID:1222330482967738Subject:Fluid Mechanics
Abstract/Summary:PDF Full Text Request
Flow separation appears usually in boundary layer at low Reynolds numbers, which will induce serious effects on aerodynamic performances and flight stability of aircraft. In order to suppress flow separation and improve the aerodynamic performances of aircraft, suction control is adopted. The detailed investigations on suction control are made by my software programmed by C language.The parameters of suction control mainly include suction coefficient, suction angle, suction hole diameter, suction hole spacing, et al. It is very important to analyze the effects of these parameters and their interactive effects on suction control. At present, most studies focus on the effect of aerodynamic performances, but don’t consider the energy consumed by suction control as an active flow control, except Wahidi’s experiment. So, the control effects of suction are comprehensively studied and assessed from the aerodynamic performances and energy consumed by suction in this dissertation.Suction control is one of the flow control methods with multiple parameters and objectives. More difficulties will occur to get the optimal set of suction parameters when more design parameters are took into account. The optimization is divided into single-objective and multi-objective. Especially, there are rare reports about Pareto multi-objective optimization of suction control at present. Therefore, it is another key technology in this dissertation to achieve the optimization with different control objectives by combining the optimization algorithm.In the three-dimensional flow around the finite-span wing, there is not only separation flow on the wing surface, but also wingtip vortex induced from wingtip. Wingtip vortex would bring aerodynamic noise and components flutter. When suction control acts, on the one hand, the effects of aerodynamic performances and figure of merit (FOM) are necessary to be analyzed, on the other hand, the variations of structure, trajectory and strength of wingtip vortex are also studied. Consequently, it’s the further investigation of suction control in the three-dimension that the influences of suction control on separation flow and wingtip vortex around the finite span wing are comprehensively analyzed.The main research contents are listed as followed:At first, the large eddy simulation (LES) software is developed and programmed in C language to solve the unsteady separation flow. AUSM+-up scheme coupling with high-order construction of WENO is used for spatial discretization. The implicit dual-time stepping with LU-SGS scheme is employed for time integral. Wall-Adapting Local Eddy-Viscosity (WALE) model is adopted as sub-grid model. The parallel computing technology is used to accelerate computation, reduce computed time. Then, the separation flow around NACA0012 airfoil at low Reynolds numbers is studied in detail using the self-programming software, the corresponding characteristics of the flow are analyzed.The flow around NACA0012 airfoil at low Reynolds number is selected as baseline case to study the suction control. The influences of suction parameters are comprehensively studied and assessed from the aerodynamic performances and energy consumed by suction. The results show that with the increase of suction coefficient, the ratios of lift to drag firstly begin to quickly increase, and then slowly decrease. The suction coefficient should be limited by a minimum value to obtain fast and effective control effect, also must be limited by a maximum value to ensure FOM is larger than one. The effect of suction angles on suction control is obvious, when the suction angle is larger, not only the lift-to-drag ratio increases, but also the cost of suction decreases. When the suction zone is located behind the separation point, the control effect is best. The hole spacings and diameters have smaller effects on airfoil aerodynamics performances, but larger influences on FOM distributions.The mechanism of suction control is analyzed in detail by the comparison of pressure coefficients, velocity profiles in boundary layer, characteristic parameters of boundary layer (such as displacement thickness, momentum thickness, shape factor). On the one hand, under the suction control, the momentum loss from separated boundary layer reduces, the streamwise momentum increases, the ability of resisting adverse pressure gradient is strengthened, so that the separation flow is suppressed. On the other hand, the low pressure in suction zones also makes contribution to the improvement of the aerodynamic performances.In order to gain the optimal set of multiple suction parameters, the optimization method is developed for solving single-objective and Pareto multi-objective problems, by coupling Genetic Algorithm with Radial Basic Function Neural Network. The results show that this optimization method has good convergence and accuracy. The maximum increase in lift-to-drag ratios about 2.4 times is achieved after performing single-objective optimization of lift-to-drag ratio. When single-objective of FOM is optimized, only drag coefficient decreases obviously, but the change of lift coefficient is little. The uniform distributed and satisfied Pareto front is gained by Pareto multi-objective optimization, and provides a selective database of effective solutions.At last, the application of suction control to the separation flow around the three-dimensional finite-span wing is discussed, and the variation of wingtip vortex under suction control is also analyzed. Suction coefficients have significant effects on the aerodynamic performances. Lift coefficients and lift-to-drag ratios gradually increase, drag coefficients and FOM gradually decrease. The influences of suction angles on aerodynamic performances are not obvious. Under the suction control, the trajectory of wingtip vortex varies significantly, and its change rate is reducing, but the structure and strength of wingtip vortex have small change.
Keywords/Search Tags:low Reynolds numbers, separation flow, large eddy simulation, suction control, RBF neural network, genetic algorithm, optimization, wing-tip vortex
PDF Full Text Request
Related items