Periodic Report Summary 1 - ACTREAL (ACTive flow control for aeroengine turbine efficiency increase in REAL stage conditions)
Aviation is among the fastest growing economic sectors as shown by ICAO and IATA air passenger number trend, despite the 2008 industry crisis, together with fleet size increase. Many aspects are connected to this sector growth, amongst all fuel consumption increase -thus pollutant emissions growth- increase of size range of aircrafts, adoption of UAVs and micro UAVs. Fuel consumption rates, hence greenhouse gas emissions, are largely accountable to engine components efficiency (combustion chambers, fan, turbine...). The design of the low pressure turbine (LPT) is a relevant issue in modern high by-pass ratio engines, because a high work output is demanded by this component which directly drives the fan. In order to meet the requirements for lower weight and part count, design trends move toward ever higher blade loading, with increasing susceptibility to flow separation which considerably degrades performances. The use of flow control is a valuable tool to overcome the large separation that LPT profiles can experience.
In addition, the issue of flow separation -and the ability to control it- is also critical to external aerodynamics. The increasing adoption of UAVs and micro-UAVs brings about the necessity for understanding airfoil stall and solutions for its control.
This projects aims at contributing to the above-mentioned issue by increasing the knowledge of the fluid dynamics of separated flows and the mechanisms of flow control, in order to suggest optimal physics-based strategies for innovative technologies.
The main research objectives can be summarized as follows:
- Deep knowledge of fluid dynamics mechanisms of separation control in real environment.
- Evaluation of the effect of active control and optimization of control parameters for LPT efficiency increase.
- Understanding of the effects of real flow conditions on flow control behavior.
- Assessment of numerical and experimental methodologies suitable for active control study.
The first two years of the project have been carried out at the Aerospace Research Center of the Ohio State University, Columbus, Oh, USA and mainly focused on experimental activities.
In the first part experimental investigations have been accomplished on a low-speed linear turbine cascade through wake pressure measurements, Particle Image Velocimetry (PIV), hot-wire anemometry. Unsteady flow control techniques such as plasma actuators, pulsed jets, synthetic jets, acoustic excitation can rely on coupling of forcing with time scales present in the uncontrolled flow. Understanding the interplay between the forcing frequencies and the instabilities is therefore critical for efficient, effective control and ultimately closed-loop applications. The study aimed at optimizing control strategies taking advantage of the instabilities present in the flow, allowing for reduced power input into the system while maximizing the separation reduction effect.
Acoustic excitation at discrete frequencies was exploited to highlight the role of actuation frequency in controlling suction side separation of a highly loaded LPT blade. The study has revealed that the most effective frequencies are those in the proximity of the fundamental instability frequency of the shear layer. This mechanism is effective to reduce laminar separation on an LPT blade at low Reynolds number, with wake loss reduction of around 40%. The most effective frequencies scale with Re^3/2 which results from the scaling of the Kelvin–Helmholtz instability.
The effect of blade loading has been evaluated considering the different control behavior on a front-loaded and an aft-loaded LPT blade profile. The authority of flow control by excitation of linear instability mechanisms has been demonstrated on LPT profiles with front-loaded distributions. However, this mechanism fails to significantly reduce separation loss in the present aft-loaded profile. Nevertheless, another efficient mechanism can be exploited by single frequency excitation, which is shown to be nonlinear vortex merging promoted by forcing in the range of the subharmonic of the fundamental frequency of the shear layer.
The mechanisms of excitation of linear instabilities is therefore effective for smaller separation occurring on an aft-loaded profile but a threshold excitation level exists beyond which no further improvement is possible for exploitation of linear instabilities. When the vortex margining mechanisms is promoted through subharmonics excitation, no threshold level is found for effectiveness in stalled front-loaded LPT blades.
The effect of freestream turbulence has been investigated with the following results: whereas with linear instability exploitation high (engine representative) freestream turbulence levels are detrimental for the effectiveness of control, it is shown that the authority of subharmonic excitation is instead enhanced by increased turbulence levels such as those present in real stage flow.
A future application of this study is the exploitation of the perturbations already present in an engine to control LPT separation. Passive tuning of the frequency content of noise sources (such as blade passing frequency, trailing edge vortex shedding, leakage flows, mechanical vibrations...) can be accomplished to match the most effective frequencies for separation control, thus providing an inexpensive and efficient control technology.
Experimental investigations of pulsed jets control on a NACA643-68 laminar airfoil have highlighted that the mechanism of instability excitation can be efficiently exploited to reduce separation and increase lift on a laminar airfoil as well. Acoustic excitation has been used as a tool to study pulsed jets control mechanisms, with the support of wavelet analysis. It has been shown that the excitation of instabilities can be triggered by pulsed jets in the proximity of the leading edge on the suction surface. The frequency of the jets can be orders of magnitude lower than the instability frequency. Low duty cycles are effective inasmuch as they introduce perturbations with frequency content in the range of the most unstable frequency of the shear layer, originating from the harmonics of the pulsing frequency at the jet opening and closing. For high angle of attacks, effective frequencies one order of magnitude higher than the shear layer most unstable frequency are found to be more effective than the linearly most unstable frequency.
In the second year of the project a new transonic linear LPT cascade facility has been developed in order to the effects of compressibility on flow control. Compressibility effects can become important with ever more increasing blade loading. Effects of roughness are considered as well at low and high Mach numbers and compared with the smooth surface results. The uniqueness of this study compared to previous works on highly loaded LPT separation control is the investigation of sensitivity of control authority at different blowing ratios to a turbulent open separation, with strong compressibility and roughness effects.
The opportunity of building a transonic cascade wind tunnel, using multiple techniques in low- and high-speed facing different issues and challenges has expanded the researcher’s knowledge and expertise in testing facilities, laboratory management, measurement techniques and fluid dynamics phenomena.
Steady vortex-generator jets (VGJs) have been tested on an in-house design LPT profile in high-speed, through wake losses, isentropic Mach number distributions, shadowgraph and Pressure Sensitive Paint techniques. When a shock-induced separation occurs due to the high curvature and uncovered portion of the blade suction surface, steady VGJ actuation provides maximum reduction of wake losses of about 25% for blowing ratios close to unity. The increased blowing favors the filling of the boundary layer upstream of separation, but this positive effect is counterbalanced by the generation of a stronger normal shock, thus generating higher wake losses for blowing ratios higher than 2 than in the uncontrolled case. Unlike low-speed flow, where a minimum effective blowing ratio (B) exists beyond which loss reduction remains constant, in high-speed the effect of shock strengthening cannot balance the positive effect of separation loss reduction, with the consequence of degradation of performance of higher blowing rates.
Another real-world effect that can affect flow control performance is surface roughness. Usually, specifications require values of surface finish that keep the roughness below the hydraulically smooth limit. However, actual degradation-induced gas turbine surface roughness can change performance significantly. A turbulent separation is present in the present blade due to the thickening of the boundary layer by roughness.
In incompressible flow regimes, steady VGJs are still capable of reducing the turbulent separation with increasing blowing ratio, but higher freestream high-momentum fluid entrainment is required: an optimum is reached at B = 3 (unlike B = 1 for laminar flows) when the jets first reach critical conditions, above which no further improvement is observed. At this rate of blowing ratio the smooth case performance are completely reestablished, with a loss reduction as high as 60%.
For higher exit Mach numbers when the transonic regime is reached, although the shock effects are weak, separation occurs due to the unhealthy state of the boundary layer encountering the uncovered portion of the blade. Flow control strengthens the shock and partially reduces separation. An optimum is found at B = 1 with a loss decrease of 15%.
In addition, the issue of flow separation -and the ability to control it- is also critical to external aerodynamics. The increasing adoption of UAVs and micro-UAVs brings about the necessity for understanding airfoil stall and solutions for its control.
This projects aims at contributing to the above-mentioned issue by increasing the knowledge of the fluid dynamics of separated flows and the mechanisms of flow control, in order to suggest optimal physics-based strategies for innovative technologies.
The main research objectives can be summarized as follows:
- Deep knowledge of fluid dynamics mechanisms of separation control in real environment.
- Evaluation of the effect of active control and optimization of control parameters for LPT efficiency increase.
- Understanding of the effects of real flow conditions on flow control behavior.
- Assessment of numerical and experimental methodologies suitable for active control study.
The first two years of the project have been carried out at the Aerospace Research Center of the Ohio State University, Columbus, Oh, USA and mainly focused on experimental activities.
In the first part experimental investigations have been accomplished on a low-speed linear turbine cascade through wake pressure measurements, Particle Image Velocimetry (PIV), hot-wire anemometry. Unsteady flow control techniques such as plasma actuators, pulsed jets, synthetic jets, acoustic excitation can rely on coupling of forcing with time scales present in the uncontrolled flow. Understanding the interplay between the forcing frequencies and the instabilities is therefore critical for efficient, effective control and ultimately closed-loop applications. The study aimed at optimizing control strategies taking advantage of the instabilities present in the flow, allowing for reduced power input into the system while maximizing the separation reduction effect.
Acoustic excitation at discrete frequencies was exploited to highlight the role of actuation frequency in controlling suction side separation of a highly loaded LPT blade. The study has revealed that the most effective frequencies are those in the proximity of the fundamental instability frequency of the shear layer. This mechanism is effective to reduce laminar separation on an LPT blade at low Reynolds number, with wake loss reduction of around 40%. The most effective frequencies scale with Re^3/2 which results from the scaling of the Kelvin–Helmholtz instability.
The effect of blade loading has been evaluated considering the different control behavior on a front-loaded and an aft-loaded LPT blade profile. The authority of flow control by excitation of linear instability mechanisms has been demonstrated on LPT profiles with front-loaded distributions. However, this mechanism fails to significantly reduce separation loss in the present aft-loaded profile. Nevertheless, another efficient mechanism can be exploited by single frequency excitation, which is shown to be nonlinear vortex merging promoted by forcing in the range of the subharmonic of the fundamental frequency of the shear layer.
The mechanisms of excitation of linear instabilities is therefore effective for smaller separation occurring on an aft-loaded profile but a threshold excitation level exists beyond which no further improvement is possible for exploitation of linear instabilities. When the vortex margining mechanisms is promoted through subharmonics excitation, no threshold level is found for effectiveness in stalled front-loaded LPT blades.
The effect of freestream turbulence has been investigated with the following results: whereas with linear instability exploitation high (engine representative) freestream turbulence levels are detrimental for the effectiveness of control, it is shown that the authority of subharmonic excitation is instead enhanced by increased turbulence levels such as those present in real stage flow.
A future application of this study is the exploitation of the perturbations already present in an engine to control LPT separation. Passive tuning of the frequency content of noise sources (such as blade passing frequency, trailing edge vortex shedding, leakage flows, mechanical vibrations...) can be accomplished to match the most effective frequencies for separation control, thus providing an inexpensive and efficient control technology.
Experimental investigations of pulsed jets control on a NACA643-68 laminar airfoil have highlighted that the mechanism of instability excitation can be efficiently exploited to reduce separation and increase lift on a laminar airfoil as well. Acoustic excitation has been used as a tool to study pulsed jets control mechanisms, with the support of wavelet analysis. It has been shown that the excitation of instabilities can be triggered by pulsed jets in the proximity of the leading edge on the suction surface. The frequency of the jets can be orders of magnitude lower than the instability frequency. Low duty cycles are effective inasmuch as they introduce perturbations with frequency content in the range of the most unstable frequency of the shear layer, originating from the harmonics of the pulsing frequency at the jet opening and closing. For high angle of attacks, effective frequencies one order of magnitude higher than the shear layer most unstable frequency are found to be more effective than the linearly most unstable frequency.
In the second year of the project a new transonic linear LPT cascade facility has been developed in order to the effects of compressibility on flow control. Compressibility effects can become important with ever more increasing blade loading. Effects of roughness are considered as well at low and high Mach numbers and compared with the smooth surface results. The uniqueness of this study compared to previous works on highly loaded LPT separation control is the investigation of sensitivity of control authority at different blowing ratios to a turbulent open separation, with strong compressibility and roughness effects.
The opportunity of building a transonic cascade wind tunnel, using multiple techniques in low- and high-speed facing different issues and challenges has expanded the researcher’s knowledge and expertise in testing facilities, laboratory management, measurement techniques and fluid dynamics phenomena.
Steady vortex-generator jets (VGJs) have been tested on an in-house design LPT profile in high-speed, through wake losses, isentropic Mach number distributions, shadowgraph and Pressure Sensitive Paint techniques. When a shock-induced separation occurs due to the high curvature and uncovered portion of the blade suction surface, steady VGJ actuation provides maximum reduction of wake losses of about 25% for blowing ratios close to unity. The increased blowing favors the filling of the boundary layer upstream of separation, but this positive effect is counterbalanced by the generation of a stronger normal shock, thus generating higher wake losses for blowing ratios higher than 2 than in the uncontrolled case. Unlike low-speed flow, where a minimum effective blowing ratio (B) exists beyond which loss reduction remains constant, in high-speed the effect of shock strengthening cannot balance the positive effect of separation loss reduction, with the consequence of degradation of performance of higher blowing rates.
Another real-world effect that can affect flow control performance is surface roughness. Usually, specifications require values of surface finish that keep the roughness below the hydraulically smooth limit. However, actual degradation-induced gas turbine surface roughness can change performance significantly. A turbulent separation is present in the present blade due to the thickening of the boundary layer by roughness.
In incompressible flow regimes, steady VGJs are still capable of reducing the turbulent separation with increasing blowing ratio, but higher freestream high-momentum fluid entrainment is required: an optimum is reached at B = 3 (unlike B = 1 for laminar flows) when the jets first reach critical conditions, above which no further improvement is observed. At this rate of blowing ratio the smooth case performance are completely reestablished, with a loss reduction as high as 60%.
For higher exit Mach numbers when the transonic regime is reached, although the shock effects are weak, separation occurs due to the unhealthy state of the boundary layer encountering the uncovered portion of the blade. Flow control strengthens the shock and partially reduces separation. An optimum is found at B = 1 with a loss decrease of 15%.