Prediction of Stall - Report Example

Prediction of stall Introduction Stall is a phenomenon, in aerodynamics, that results due to a sudden reduction in the lift forces generated by an airfoil. Stall occurs when the angle of attack (angle between the wing’s mean chord and the relative wind, parallel to the flight path, or the wind moving past the airfoil as the airfoil moves through the air) exceeds the wing’s ability to produce lift as a result of turbulent air flow over the wings. The low-pressure area above the wing reduces causing inability to lift. One can encounter a stall at any airspeed, any altitude and any amount of throttle. Stall speeds change with angle of bank, configuration and any changes in gross weight. Furthermore, an aircraft will stall at a higher speed with flaps retracted than with them extended. The phenomenon of stall is shown in the figure below:- Figure 1: Pictorial representation of the occurance of stall Figure 2: Relation between angle of attack and coefficient of lift showing stall angle In the subsequent paragraphs, we shall discuss the types, means to predict and control and available technologies to avert stall. Types of Stalls Stall occur due to a number of reasons and can thus be divided into a number of types1 like power off stalls (also called approach-to-landing stalls, practiced by pilots to simulate normal approach-to-landing conditions and configuration), Power-on stalls (also known as departure stalls, practiced by pilots to simulate takeoff and climb-out conditions) and Accelerated stalls (experienced at higher-than normal airspeeds due to applying abrupt or excessive controls to aircraft and may occur in steep turns, pullups or other abrupt changes in your flight path.). Recognizing an Impending Stall Understanding when aircraft is about to stall is very important. Some signs gives warning of stalls are: Pilot experience less control over his aircraft as its airspeed is reduced. This is partly because the airflow over the flight control surfaces is reduced. If he is flying a fixed-pitch propeller airplanes, he may detect a decrease in revolutions per minute (RPM) when approaching a stall while in power-on conditions. His aircraft experiences buffeting, uncontrollable pitching or vibrations that begin just before the stall occurs. Many aircraft in Flight Sim use a stall indicator to alert you when the airflow over the wing(s) approaches a point that lift cannot be sustained. The stall indicator is part of aircraft for a reason so no one should ignore it. Design Procedure The design calculation is an important procedure that solves two purposes as under:- Improve the take-off and landing performance of existing high-lift systems using an adjoint formulation. Setup a numerical optimization procedure that can be useful to the aerodynamicist in the rapid design and development of high-lift system configurations and that can also provide derivative information regarding the influence of various design parameters (gap, overlap, slat and flap deflection angles, etc.) on the performance of the system. Flow diagram of the above procedure is depicted as under:- Figure 3: Flow diagram of the design calculation procedure The variables that describe the relative element positioning can be used as design variables. These variables include flap and slat deflection angles, gaps, overlap, shapes of each of the elements and many more depending on method used2. Like in Continuous Adjoint Method, variables are flow-field variables and the physical location of the boundary. Then the governing equation which expresses the dependence of these variables can be written in flow field domain in form of partial differential equations. After introducing a Lagrange Multiplier and solving adjoint equations and applying boundary conditions we get gradient equation3. An active transparent stall control system utilizing sensors, actuators, and a closed-loop controller was designed and tested on a NACA 0020 wing model in a low-speed wind tunnel. The main objective of the control system was to enable active detection and control of local flow separation using a system of fast response pressure sensors DFEs, and a feedback controller, to delay wing stall. The method of predicting flow separation was based of the pressure fluctuations in the vicinity of flow separation the underlying principle of a smart-wing concept that utilizes the transparent stall control technique. Abrupt shifts in the pressure data, captured as spikes in standard deviation (STDEV) of surface pressures created a characteristic signature of imminent flow separation which served as the trigger for activation of DFEs. Pressure fluctuations captured from a fast-response (sampling rate 500-Hz) pressure sensor of pitching airfoil and STDEV are plotted. Based on these observations the controller architecture for this study was extended to include a memory unit. The controller detected characteristics shifts of the pressure data by monitoring STDEV values and deployed DFES in a preassigned configuration. In summary detection of incipient stall was achieved solely through mathematical analysis of pressure fluctuations recorded from pressure sensors that were embedded flush to the surface of the wing. Consequently, only a single fast-response pressure sensor located optimally on the surface of the wing can be used for stall detection. There are other techniques also to predict stall. They are discussed next. NACA 0012 The rapid growth of the wind-turbine industry has generated renewed interest in the dynamic stall phenomenon due to the fact that wind-turbine blades operate continuously under stalled flow conditions. According to the dynamic test results of a NACA-0012 airfoil model4, very unusual dynamic stall characteristics are generated by interactions between the some flow phenomena. The combined effect of The accelerated-flow and moving-wall effects on dynamic airfoil stall can be represented by the dominating moving-wall effect, illustrated by classic Maguus lift results in both laminar and turbulent flow separation. The reversal of these effects when they influence flow separation via their effects on boundary layer transition, and the interaction between the oscillating flow separation and Karman vortex shedding explains how each of these flow phenomena can have strong effects by itself of the dynamic stall characteristics of wind-turbine and helicopter blades. However, the biggest problem is that they can severely distort the dynamic results obtained in subscale tests. As it is readily possible to avoid this problem if appropriate steps are taken in the planning of the test, it is important that the test engineer is aware of these intricate flow interactions. The effects of local surface buzzing vertical to flow direction near the leading edge on an airfoil were investigated at low Reynolds numbers5. The results showed that the aerodynamic characteristics including poststall lift and drag have been improved with milder stalling feature. The locally introduced unsteady disturbances by the present method caused the separated boundary layer to reattach to the surface by intensifying turbulence activity to enhance mixing and entertainment, which can be deduced from the result of oil surface visualization and pressure distributions. Further study for applying this to real situation will be needed by considering higher Reynolds numbers as well as higher frequencies including the power spectrum analysis and velocity measurement in the flow field. The dynamic stall characteristics of conventional airfoils used in helicopter blades, and airfoils whose shapes change dynamically with time are numerically studied. Two-dimensional Navier-Stokes equations in integral form are solved on a body-fitted grid that deforms as the airfoil changes its shape, and rotates with the airfoil in pitch6. The computed surface pressure distributions and the integrated loads show that the dynamically deforming leading edge airfoil has a superior performance compared to the NACA 0012 airfoil. These two shapes are quite similar except for subtle variations in the surface curvature near the leading edge. It is demonstrated that the deforming leading edge shape, which has a slightly larger leading edge radius, exhibits a much milder dynamic stall hysteresis. It does not exhibit the high nose down pitching moments, or the high drag values that are expected of an airfoil undergoing dynamic stall. The difference between the two flow fields is striking, given the fact that the airfoil deformations are rather small7. Dynamic Stall Control Using Dynamic Shape Adaptation A mathematical model is developed to predict the suppression of rotating stall in a centrifugal compressor with a vaned diffuser8. This model is based on the employment of a control vortical waveform generated upstream of the impeller inlet to damp weak potential disturbances that are the early stages of rotating stall. The control system is analyzed by matching the perturbation pressure the compressor inlet and exit flow fields with a model for the unsteady behavior of the compressor. The model was effective at predicting the stalling behavior of the Purdue Low-Speed Centrifugal Compressor for two distinctly different stall patterns. Predictions made for the effect of a controlled inlet vorticity wave of the stability of the compressor show that, for minimum control wave magnitudes, on the order of the total inlet disturbance magnitude, significant damping of the instability can be achieved. For control wave amplitude of sufficient amplitude, the control phase angle appears to be the most important factor in maintaining a stable condition in the compressor. The effect of introducing a lag to allow for the control wave to be convented through the compressor floe path in all cases was beneficial to control effectiveness. The convected wave assumption also produced a shift in phase angle for optimum control effectiveness. Studies of Compressible Dynamic Stall The complex surface-flow physics of compressible dynamic stall has been experimentally documented using flush-mounted heat-flux gauges9. The study has revealed the various surface-flow features and their changes under different flow conditions. It has also provided a description of the surface-flow events that lead to compressible dynamic-stall onset. The imprint of a leading-edge shock has been recorded for the first time in the surface-flow signature. The dominant features of the flow are 1) the rapid movement over the airfoil of the transition point in the unsteady flow on the upstroke and flow relaminarization on the down stroke; 2) the formation of the dynamic-stall vortex at the leading edge through the bursting of the laminar separation bubble, or, at higher Mach numbers, through shock-induced flow separation from somewhere in the middle of the bubble; and 3) an extremely rapid rise of the surface shear stress at the onset of stall for different onset mechanisms. Even though compressible dynamic stall can originate from several causes depending upon flow conditions, the study led to the conclusion that a deterministic precursor of dynamic-stall onset is a sharp rise in the surface shear stress in the leading-edge adverse pressure gradient region, which is a common and singular flow feature at all conditions tested Hence, it is expected that the same indicator will also serve as a precursor to full-scale rotor dynamic stall; even though a laminar bubble is unlikely to be present in the higher-Reynolds-number rotor case and some of the flow details will be different. A definitive result can be obtained by repeating the present studies on a tripped airfoil. The information generated here can be used to develop suitable transition and relaminarization models to be included in the computational studies of the flow for more realistic results. Stall Control Using Dynamic Shape Adaptation Compressible dynamic stall has been successfully controlled using dynamic shape adaptation. This required a very small (0.6 mm) change in the chord length of dynamically adaptive airfoil that produced a nearly 150% change in leading edge radius of curvature10. The flow was found to be dynamic stall vortex free for specific values of variables. The favorable effects of dynamic shape adaptation realized through changes in the instantaneous potential flow resulted in broader pressure distribution with lower peak suction values and led to a redistribution of the unsteady flow velocity. The vorticity levels decreased to values where the dynamic stall vortex did not form. The peak suction variation loop over the oscillation cycle was found to be smallest for adopting airfoil. The deformation rate, the initiation angle of attack, and the amount of nose curvature change affect the success of approach significantly. The most benefit is produces while remaining within the attached flow envelope for a given mach number during dynamic shape adaptation. Bubble Bursting Slotted multi element configurations are widely used because they are very effective in increasing the maximum lift of airfoils during take off and landing. However, when stall occurs in the leading edge region of one of element, the outcome is a sudden and dangerous drop in performance of whole configuration. The net results of the experimental verification of a computer designed single slotted flap are presented. Many realistic configurations, including the computed optimum, were tested and compared to numerical predictions. For a number of these configurations, sudden leading edge stall due to the bursting of a laminar bubble was detected on the flap, followed by severe stall hysteresis. Present numerical design codes do not help the designer much in predicting this important phenomena, and wind tunnel testing remains necessary. Therefore, there is a need for better bubble bursting prediction methods. A possible direction for improvement is discussed, focusing on unsteadiness of flow11. The experimental and numerical results obtained on a single slotted flap configuration designed for the wing of a general aviation aircraft were presented. The results clearly show that the current numerical design codes are incapable of predicting bursting occurrence. This is indeed a major cause for concern: once bursting occurs, it can give rise to a large hysteresis loop. The angle of attack has to be lowered appreciably to regain prestall conditions. But the bubble bursting to phenomenon is still lacking of many issues, most of them related to the structure of the transitional region. A number of crucial points, especially related to the unsteadiness of the laminar separation have been identified as currently missing for an effective bursting prediction. So the further research should focus on importance of the laminar unsteadiness for the structure of the bubble, especially close to and after bursting condition. Unsteady experiments and unsteady navier-stokes calculations need to be performed to improve the existing bursting prediction methods. Conclusion There has been dramatic progress in dynamic stall research during the past 10 years. The level of understanding has advances from qualitative conjectures concerning the possibility of dynamic delay of stall, to quantitative measurement of the instantaneous character of the viscous flow during stall process itself, and the development of quantitative empirical models that reflect even small variations of stall loads during the cycle. There is much to be learned- little is known of compressibility effects or of the influence of Reynolds number or three dimensionality on dynamic stall loads, and there is need for an extended effort diverted toward measurement of details of viscous flow that results in dynamic stall. References Bangalore, A and Sankar, L. N., “Forward Flight Analysis of Slaft’ed rotors using Navier-Stokes methods”, Journal of Aircraft, 34.1 (January-February 1997): 80-86. Lars E. Ericsson, "Revisiting Unresolved Dynamic Stall Phenomena." Journal of Aircraft, 37.6 M. Baragona, L. M. M. Boermans, M. J. L. van Tooren, H. Bijl and A. Beukers, "Bubble bursting and stall hysteresis on single-slotted flap high lift configuration". AIAA Journal, 41. (July 2003). Mehmet Sahin and Lakshmi N. Sankar, "Stall Alleviation using a Deformable Leading Edge Concept". IEEE (2000). Mehul P. Patel, "Active Transparent Stall Control System for Air Vehicles". Journal of Aircraft, 40.5 (Sept 2003). M. S. Chandrasekhara, "Compressible dynamic stall control using dynamic shape adaptation". AIAA Journal, 39.10 (2001). M. S. Chandrasekhara, "Heat-Flux Gauge Studies of Compressible Dynamic Stall". AIAA Journal, 41.5 (May 2003). P. B . Lawless and S Fleetert, "Prediction of Active Control of Subsonic Centrifugal Compressor Rotating Stall". AIAA Journal, 35.12 (Dec 1997). Sangho Kim, Juan J. Alonso and Antony Jameson, "Two-dimensional High-Lift Aerodynamic Optimization Using the Continuous Adjoint Method" AIAA 2000–4741. S. Kim, J. J. Alonso, and A. Jameson. “A Gradient Accuracy Study For The Adjoint-Based Navier-Stokes Design Method”. AIAA paper 99-0299, AIAA 37th Aerospace Sciences Meeting & Exhibit, Reno, NV, (January 1999). Slaughter, Scott. "Recovering from Stalls and Spins". Abacus (2006). Y. W. Park, Soo-gab lee, Dong-Ho Lee, and S. Hong, "Stall control with local surface buzzing on a NACA 0012 airfoil". AIAA Journal, 39.7 (July 2001). Read More