scholarly journals Uncertainty Quantification of the Effects of Small Manufacturing Deviations on Film Cooling: A Fan-Shaped Hole

Aerospace ◽  
2019 ◽  
Vol 6 (4) ◽  
pp. 46 ◽  
Author(s):  
Wei Shi ◽  
Pingting Chen ◽  
Xueying Li ◽  
Jing Ren ◽  
Hongde Jiang

The film cooling holes in the blade of modern gas turbines have commonly been manufactured by laser drilling, Electric Discharge Machining (EDM), and Additive Manufacturing (AM) in recent years. These manufacturing processes often result in small geometric deviations, such as conical angles, filleted edges, and diameter deviations of the hole, which can lead to deviations on the distribution of adiabatic cooling effectiveness (η) values, the value of the discharge coefficient (Cd), and the characteristic of the in-hole flow field. The current study employed flat plate fan-shaped film cooling holes with length-to-diameter values (L/D) equal to 3.5 and six to investigate the effects of these manufacturing deviations on the distribution of η values, the value of Cd, and the characteristic of in-hole flow field. An Uncertainty Quantification (UQ) analysis using the Polynomial Chaos Expansion (PCE) model was carried out to quantify the uncertainty in the values of η and Cd. The statistical characteristics (mean values, standard deviation (Std) values, and Probability Density Function (PDF) values) of η and Cd were also calculated. The results show that conical angle deviations exert no visible changes on the value of η. However, the Cd value decreases by 6.2% when the conical angle changes from 0–0.5°. The area averaged adiabatic cooling effectiveness ( η = ) decreases by 3.4%, while the Cd increases by 15.2% with the filleted edge deviation existing alone. However, the deviation value of η = is 7.6%, and that of Cd is 25.7% with the filleted edge deviation and the diameter deviation existing.

Author(s):  
Lucas Giller ◽  
Heinz-Peter Schiffer

The interaction between the strongly swirling combustor outflow and the high pressure turbine nozzle guide vanes were investigated at the cascade test rig at Technische Universität Darmstadt. The test section of the rig consists of six swirl generators and five cascade vanes. The three middle vanes are equipped with film cooling holes at the leading edges. The swirler nozzles are aligned with the center of the cascade passages. The operating settings are defined by the swirl number, the distance between the swirler nozzles and the vanes, the blowing ratio and the radial angle of the film cooling holes. Flow field measurements using PIV downstream of the swirlers and five hole probe measurements at the inlet and outlet plane of the cascade were accomplished. Measurements using the ammonia diazo technique to determine the adiabatic film cooling effectiveness on the surface of the center cascade vane were also carried out. It is shown that a swirling inflow leads to a strong alteration of the flow field and the losses in the passages in comparison to an axial inflow. Furthermore, the impact of the swirl on the formation of the cooling film and it’s adiabatic film cooling effectiveness is presented.


2021 ◽  
Author(s):  
Hai Wang ◽  
Chun-hua Wang ◽  
Xing-dan Zhu ◽  
Jian Pu ◽  
Hai-ying Lu ◽  
...  

Abstract Uncertainty due to operating conditions in gas turbines can have a significant impact on film cooling performance, or even the life of hot-section components. In this study, uncertainty quantification technique is applied to investigate the influences of inlet flow parameters on film cooling of fan-shaped holes on a stator vane under realistic engine conditions. The input parameters of uncertainty models include mainstream pressure, mainstream temperature, coolant pressure and coolant temperature, and it is assumed that these parameters conform to normal distributions. Surrogate model for film cooling is established by radial basis function neural network, and the statistical characteristics of outputs are determined by Monte Carlo simulation. The quantitative analysis results show that, on pressure surface, a maximum value of 61.6% uncertainty degree of laterally averaged adiabatic cooling effectiveness (ηad,lat) locates at about 4.0 diameters of hole downstream of the coolant exit; however, the maximum uncertainty degree of ηad,lat is only 4.5% on suction surface. Furthermore, the probability density function of area-averaged cooling effectiveness is of highly left-skewed distribution on pressure surface. By sensitivity analysis, the variation of mainstream pressure has the most pronounced effect on film cooling, while the effect of mainstream temperature is unobvious.


Author(s):  
Yifei Li ◽  
Yang Zhang ◽  
Xinrong Su ◽  
Xin Yuan

The influence of the cross flow in mainstream on film cooling performance and jet flow field is investigated experimentally and numerically. To show the effect of cross flow in mainstream without the influence of the other secondary flows, a curved test section is constructed to generate a cross flow, simulating the curved turbine passage. Both the straight and the curved passage are used to show the differences of cooling performance for shaped holes with and without the cross flow, with blowing ratio varying from M = 0.5 to M = 2.5. Pressure sensitive paint is used to measure the adiabatic cooling effectiveness, and the ink trace measurement is conducted to present the friction lines on the endwall platform. Numerical simulations are performed to show the flow field. The cross flow is accelerated in a curved passage and migrates the fluid near the endwall platform. Due to the cross flow in the mainstream, the deflection angle changes a lot along the normal direction to the endwall, and dominates the spatial distribution of coolant. Although the cooling trace follows the trend of wall surface streamlines, the migration of coolant is slower than the deviation of the friction line, and the difference increases with increasing blowing ratios. The cross flow enhances the lateral dispersion, decreasing the peak value of cooling effectiveness but increasing the laterally averaged cooling effectiveness. Higher blowing ratios lead to a higher intensity of a counter-rotating vortex pair that limits lateral dispersion near the outlet of cooling hole. But the effect of cross flow dominates the flow pattern downstream. The cooling performance has a significant difference with the influence of the cross flow. This study is essential to understand the interaction of the cross flow and the film cooling in gas turbines.


Author(s):  
Shantanu Mhetras ◽  
Je-Chin Han ◽  
Ron Rudolph

The effect of film cooling holes placed along the span of a fully-cooled high pressure turbine blade in a stationary, linear cascade on film cooling effectiveness is studied using the Pressure Sensitive Paint (PSP) technique. Effect of showerhead injection at the leading edge and the presence of compound angled, diffusing holes on the pressure and suction side are also examined. Six rows of compound angled shaped film cooling holes are provided on the pressure side while four such rows are provided on the suction side of the blade. The holes have a laidback and fan-shaped diffusing cross-section. Another three rows of cylindrical holes are drilled at a typical angle on the leading edge to capture the effect of showerhead film coolant injection. The film cooling hole arrangement simulates a typical film cooled blade design used in stage 1 rotor blades for gas turbines used for power generation. A typical blowing ratio is defined for each film hole row and tests are performed for 100%, 150% and 200% of this typical value. Tests are performed for inlet Mach numbers of 0.36 and 0.45 with corresponding exit Mach numbers of 0.51 and 0.68 respectively. The flow remains subsonic in the throat region for both Mach numbers. The corresponding free stream Reynolds number, based on the axial chord length and the exit velocity, are 1.3 million and 1.74 million respectively. Freestream turbulence intensity level at the cascade inlet is 6%. Results show that varying blowing ratios can have a significant impact on film-cooling effectiveness distribution. Large spanwise variations in effectiveness distributions are also observed. Similar distributions were observed for both Mach numbers.


Author(s):  
Zhi Tao ◽  
Zhendong Guo ◽  
Liming Song ◽  
Jun Li

Abstract With the continuous increase of aerodynamic and thermal load, the endwall of modern gas turbines has become the critical region affected by the uncertainties in the manufacturing and operation process and thus is very likely to suffer performance degradation and thermal failure. Therefore, it is critical to understand and quantify the impacts of uncertainty factors on endwall aero-thermal performance. Based on Kriging surrogate, the frameworks of uncertainty quantification and global sensitivity analysis are constructed for a gas turbine blade endwall. The impacts of slot geometry deviations (slot width, endwall misalignment) and mainstream condition fluctuations (turbulence intensity, inlet flow angle) on endwall aero-thermal performance are quantified and analyzed. Results show that the actual performance of the endwall has a high probability of deviating from its nominal value. With respect to the nominal values, the maximum deviations of aerodynamic losses, averaged film cooling effectiveness and averaged Nusselt number reach up to 0.33%, 45% and 5.0%, respectively. The critical regions which are most sensitive to the input uncertainty parameters are identified. Furthermore, the inlet flow angle is proved to be the most significant parameter affecting the endwall aero-thermal performance through sensitivity analysis. The influence mechanisms of the inlet flow angle on endwall aero-thermal performance are clarified by detailed flow and thermal field analysis. Results show that the inlet flow angle significantly alters the size and strength of the secondary flow structures, resulting in a large variation of endwall aero-thermal performance. Quantitatively, a positive incidence angle of 2 degrees can lead to a 0.1% reduction of total pressure coefficient, a 12% reduction of averaged film cooling effectiveness and a 2% enhancement of averaged Nusselt number.


2011 ◽  
Vol 134 (1) ◽  
Author(s):  
Shantanu Mhetras ◽  
Je-Chin Han ◽  
Ron Rudolph

The effect of film-cooling holes placed along the span of a fully cooled high pressure turbine blade in a stationary, linear cascade on film-cooling effectiveness is studied using the pressure sensitive paint technique. The effect of showerhead injection at the leading edge and the presence of compound angled, diffusing holes on the pressure and suction sides are also examined. Six rows of compound angled shaped film-cooling holes are provided on the pressure side while four such rows are provided on the suction side of the blade. The holes have a laidback and fan-shaped diffusing cross-section. Another three rows of cylindrical holes are drilled at a typical angle on the leading edge to capture the effect of showerhead film coolant injection. The film-cooling hole arrangement simulates a typical film cooled blade design used in Stage 1 rotor blades for gas turbines used for power generation. An optimal target blowing ratio is defined for each film hole row, and tests are performed for 100%, 150%, and 200% of this target value. Tests are performed for inlet Mach numbers of 0.36 and 0.45 with corresponding exit Mach numbers of 0.51 and 0.68, respectively. The flow remains subsonic in the throat region for both Mach numbers. The corresponding freestream Reynolds numbers, based on the axial chord length and the exit velocity, are 1.3×106 and 1.74×106, respectively. Freestream turbulence intensity level at the cascade inlet is 6%. The results show that varying blowing ratios can have a significant impact on film-cooling effectiveness distribution. Large spanwise variations in effectiveness distributions are also observed. Similar distributions were observed for both Mach numbers.


2021 ◽  
Author(s):  
Zhiyu Li ◽  
Kaiyuan Zhang ◽  
Zhigang Li ◽  
Jun Li

Abstract Lean burn premix technology of gas turbines results in that the turbine combustor outlet has strong swirl flow characteristics which directly influences the inlet flow condition of the first turbine vane downstream from combustor and raise the thermal load of endwall. The aerothermal performance and film cooling effectiveness of first turbine vane endwall at different inlet swirl conditions is numerically investigated in this paper. The flow pattern, Nusselt number distribution and film cooling effectiveness of turbine vane endwall at the uniform and three kinds of swirl numbers (0.6, 0.8, 1.0) inflow conditions with clockwise and anticlockwise swirl orientations are analyzed using three-dimensional Reynolds-Averaged Navier-Stokes (RANS) and SST k-ω turbulence model solutions. The obtained results show that the flow field is apparently influenced by inlet swirl conditions. The separation in passage is clearly suppressed at clockwise swirl inflow conditions but anticlockwise make the flow pattern more complex. Inlet swirl can increase the overall Nu on the endwall. Especially in the area upstream of leading edge and area between the first and second row of film holes whose Nu are increased by 4 and 1.5 times compared with uniform inflow, respectively. Not only swirl number but also orientation can affect the film cooling effectiveness distribution of the vane endwall. The better film cooling effectiveness distribution and higher film cooling effectiveness downstream of each film hole row can be achieved at the clockwise swirl inflow conditions by weakening the accumulation of coolant near the suction side of the turbine vane endwall. Compared with uniform inflow, average film cooling effectiveness of endwall between the second row and third row is increased from 0.21 to 0.27 at clockwise SN = 1.0 for the maximum increase. The detailed flow field and aerothermal performance of the turbine vane endwall at different swirl inflow conditions is also discussed and illustrated.


Author(s):  
Shantanu Mhetras ◽  
Je-Chin Han ◽  
Ron Rudolph

The effect of film cooling holes placed along the span of a fully-cooled high pressure turbine blade in a stationary, linear cascade on film cooling effectiveness is studied using the Pressure Sensitive Paint (PSP) technique. Six rows of compound angled shaped film cooling holes are provided on the pressure side while four such rows are provided on the suction side of the blade. The holes have a laidback and fan-shaped diffusing cross-section. Another three rows of holes are drilled on the leading edge to capture the effect of showerhead film coolant injection. The film cooling hole arrangement simulates a typical film cooled blade design used in stage 1 rotor blades for gas turbines used for power generation. A minimum blowing ratio is defined for each film hole row and tests are performed for 1.0x, 1.33x, 1.67x, 2.0x and 2.67x of this minimum value. Tests are performed for an inlet Mach number of 0.36 with a corresponding exit Mach number of 0.51. The flow remains subsonic in the throat region. The corresponding free stream Reynolds number, based on the axial chord length and the exit velocity, is 1.3 million. Turbulence intensity level at the cascade inlet is 5% with an integral length scale of around 5cm. Results show that varying blowing ratios can have a significant impact on film-cooling effectiveness distribution. Large spanwise variations in effectiveness distributions are also observed. Results also show that the effectiveness magnitudes from superposition of effectiveness data from individual rows are comparable with that from full coverage film cooling.


Author(s):  
Daren Zheng ◽  
Xinjun Wang ◽  
Feng Zhang ◽  
Junfei Zhou ◽  
Qi Yuan

This paper presents a numerical investigation on a concept for enhancing the film cooling performance by modifying the shape of upstream ramps. The novel shape ramp, which is placed in front of the film cooling holes, is presented to alter the approaching boundary-layer flow and its interaction with coolant to increase the lateral spreading of the coolant. Five different shape ramps are investigated, including rectangular, wedge-shaped, convex, concave and wave-shaped ramps. The film cooling performance of different shape ramps are evaluated at the density ratio about 1, with blowing ratios ranging from 0.3 to 1.2. The numerical results for the upstream ramp show an agreement with experiment data when solving three dimensional average Navier-Stokes analysis with the k-ε model. Detailed adiabatic cooling effectiveness and total pressure loss coefficient are simulated. Results obtained indicate that film cooling characteristics in the region downstream of the film cooling holes are sensitive to the ramp shapes. The wave-shaped ramp shows the lowest total pressure loss coefficient among these five ramps. For M = 0.3 and 0.6, the highest centerline adiabatic cooling effectiveness occurs in the convex ramp. And this shape ramp also shows the highest spanwise averaged adiabatic cooling effectiveness at the blowing ratio of 0.3. Compared with the other shape ramps, the concave ramp can greatly increase both the centerline and the spanwise averaged adiabatic cooling effectiveness for M = 1.0 and 1.2.


Author(s):  
Shane Haydt ◽  
Stephen Lynch ◽  
Scott Lewis

Shaped film cooling holes are used extensively in gas turbines to reduce component temperatures. These holes generally consist of a metering section through the material and a diffuser to spread coolant over the surface. These two hole features are created separately using electrical discharge machining, and occasionally an offset can occur between the meter and diffuser due to misalignment. The current study examines the potential impact of this manufacturing defect to the film cooling effectiveness for a well-characterized shaped hole known as the 7-7-7 hole. Five meter-diffuser offset directions and two offset sizes were examined, both computationally and experimentally. Adiabatic effectiveness measurements were obtained at a density ratio of 1.2 and blowing ratios ranging from 0.5 to 3. The detriment in cooling relative to the baseline 7-7-7 hole was worst when the diffuser was shifted upstream (aft meter-diffuser offset), and least when the diffuser was shifted downstream (fore meter-diffuser offset). At some blowing ratios and offset sizes, the fore meter-diffuser offset resulted in slightly higher adiabatic effectiveness than the baseline hole, due to a reduction in the high-momentum region of the coolant jet caused by a separation region created inside the hole by the fore meter-diffuser offset. Steady RANS predictions did not accurately capture the levels of adiabatic effectiveness or the trend in the offsets, but it did predict the fore offset’s improved performance.


Sign in / Sign up

Export Citation Format

Share Document