東華大學圖書館 |

Design of Flush Air Data Systems Insensitive to Manufacturing Variance.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Design of Flush Air Data Systems Insensitive to Manufacturing Variance./
作者:	Dunbar, Grant .
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
面頁冊數:	124 p.
附註:	Source: Masters Abstracts International, Volume: 81-03.
Contained By:	Masters Abstracts International81-03.
標題:	Aerospace engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13864485
ISBN:	9781085623698

Design of Flush Air Data Systems Insensitive to Manufacturing Variance.
Dunbar, Grant .

Design of Flush Air Data Systems Insensitive to Manufacturing Variance. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 124 p.

Source: Masters Abstracts International, Volume: 81-03.

Thesis (M.S.)--University of Colorado at Boulder, 2019.

This item must not be sold to any third party vendors.

Flush air data systems are systems designed to measure flow angle using several channels of pressure measurements. A pressure coefficient can be calculated based on these pressures, which is then correlated to angle of attack through a regression model. Calibration of these systems is usually performed experimentally, which is cost- and time-intensive. These commitments can be reduced if calibration can instead be performed via simulation. However, geometric differences between the design and the physical flush air data system often prevent this from happening and even require copies of the same system (such as a production run of multi-hole probes) to be individually experimentally calibrated. If a flush air data system insensitive to manufacturing variations can be constructed, then calibration requirements (and thus cost) can be reduced.The case of a hemispherical multi-hole probe is first considered as it can be modeled using potential flow around a sphere after an initial CFD study that selected the hemispherical shape as generally the most robust and location of the pressure ports as one of the parameters that may be made more robust through design. A parameter of "relative sensitivity'' is developed to determine how large of an effect a theoretical manufacturing variation may have on the resulting system data. This parameter is calculated by taking the derivative of the pressure distribution with respect to displacement along the surface geometry, and then dividing that quantity by the derivative of the pressure distribution with respect to changes in angle of attack. This gives a measure that can be thought of as the difference in angle of attack measured by a pressure port due to a given displacement from the design location. The absolute value of the minimum of this parameter will occur at the location that is most robust to displacements in pressure port location (this parameter could be modified in the future to measure robustness to other quantities as well). In the hemispherical multi-hole probe case investigated here, it was found that the most robust pressure port locations were 45 degrees along the sphere from the front of the probe, which agrees with current probe designs.With the method proven on the simple hemispherical multi-hole probe case it was then extended to the case of a flush air data system on a NACA 0015 airfoil, utilizing the XFOIL panel code at first and then moving to an experimental study. Three pressure ports were required as a measure of angle of attack was desired independent of the effects of dynamic and static pressures. Two of these pressure ports were "measurement'' ports, with one "reference'' port. Using the relative sensitivity parameter, it was found with the XFOIL data that the measurement ports should be between 10% and 60% chord on the pressure and suction surfaces of the airfoil. The pressure at the trailing edge of the airfoil was almost completely invariant with angle of attack, and so was selected as the location of the reference pressure port. A pressure coefficient that could be correlated to angle of attack was defined using these three pressure measurements. The validity of a 3-port flush air data system based on the chord of an airfoil or wing was verified using the XFOIL data. While testing flush air data system configurations it was found that the pressure coefficient function isn't necessarily monotonic with respect to angle of attack, highlighting another item to be considered when designing a flush air data system.A NACA 0015 wing model was then tested in the University of Colorado Boulder low speed wind tunnel. The model had 31 pressure ports distributed chordwise on the pressure and suction surfaces to both build a pressure profile of the airfoil and to allow for choosing any 3 pressure ports as a flush air data system. The wind tunnel tests were run at angles of attack from 0 degrees to 10 degrees at speeds of 15 m/s, 20 m/s, and 25 m/s to replicate Reynolds numbers often seen on small UAVs (150,000, 200,000, 250,000). The wind tunnel tests confirmed the trends found in the XFOIL cases. An example FADS configuration was then investigated in more detail, with the pressure coefficient function being correlated to angle of attack using a fourth order polynomial fit. The resulting system was able to estimate angle of attack to an accuracy of approximately 0.2 degrees. The same system was also calibrated using a polynomial fit to the XFOIL inviscid model predictions, and it was found that calibrating using the XFOIL model resulted in angle of attack estimations accurate to within 1 degree for small angles.

ISBN: 9781085623698Subjects--Topical Terms:

1002622
Aerospace engineering.
Subjects--Index Terms:

Flush air data system

Design of Flush Air Data Systems Insensitive to Manufacturing Variance.
LDR:06049nmm a2200409 4500 001 2272498
005 20201105110108.5
008 220629s2019 ||||||||||||||||| ||eng d
020 $a 9781085623698
035 $a (MiAaPQ)AAI13864485
035 $a AAI13864485
040 $a MiAaPQ $c MiAaPQ
100 1 $a Dunbar, Grant . $3 3549934
245 1 0 $a Design of Flush Air Data Systems Insensitive to Manufacturing Variance.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 124 p.
500 $a Source: Masters Abstracts International, Volume: 81-03.
500 $a Advisor: Farnsworth, John.
502 $a Thesis (M.S.)--University of Colorado at Boulder, 2019.
506 $a This item must not be sold to any third party vendors.
506 $a This item is not available from ProQuest Dissertations & Theses.
506 $a This item is not available from ProQuest Dissertations & Theses.
506 $a This item must not be sold to any third party vendors.
520 $a Flush air data systems are systems designed to measure flow angle using several channels of pressure measurements. A pressure coefficient can be calculated based on these pressures, which is then correlated to angle of attack through a regression model. Calibration of these systems is usually performed experimentally, which is cost- and time-intensive. These commitments can be reduced if calibration can instead be performed via simulation. However, geometric differences between the design and the physical flush air data system often prevent this from happening and even require copies of the same system (such as a production run of multi-hole probes) to be individually experimentally calibrated. If a flush air data system insensitive to manufacturing variations can be constructed, then calibration requirements (and thus cost) can be reduced.The case of a hemispherical multi-hole probe is first considered as it can be modeled using potential flow around a sphere after an initial CFD study that selected the hemispherical shape as generally the most robust and location of the pressure ports as one of the parameters that may be made more robust through design. A parameter of "relative sensitivity'' is developed to determine how large of an effect a theoretical manufacturing variation may have on the resulting system data. This parameter is calculated by taking the derivative of the pressure distribution with respect to displacement along the surface geometry, and then dividing that quantity by the derivative of the pressure distribution with respect to changes in angle of attack. This gives a measure that can be thought of as the difference in angle of attack measured by a pressure port due to a given displacement from the design location. The absolute value of the minimum of this parameter will occur at the location that is most robust to displacements in pressure port location (this parameter could be modified in the future to measure robustness to other quantities as well). In the hemispherical multi-hole probe case investigated here, it was found that the most robust pressure port locations were 45 degrees along the sphere from the front of the probe, which agrees with current probe designs.With the method proven on the simple hemispherical multi-hole probe case it was then extended to the case of a flush air data system on a NACA 0015 airfoil, utilizing the XFOIL panel code at first and then moving to an experimental study. Three pressure ports were required as a measure of angle of attack was desired independent of the effects of dynamic and static pressures. Two of these pressure ports were "measurement'' ports, with one "reference'' port. Using the relative sensitivity parameter, it was found with the XFOIL data that the measurement ports should be between 10% and 60% chord on the pressure and suction surfaces of the airfoil. The pressure at the trailing edge of the airfoil was almost completely invariant with angle of attack, and so was selected as the location of the reference pressure port. A pressure coefficient that could be correlated to angle of attack was defined using these three pressure measurements. The validity of a 3-port flush air data system based on the chord of an airfoil or wing was verified using the XFOIL data. While testing flush air data system configurations it was found that the pressure coefficient function isn't necessarily monotonic with respect to angle of attack, highlighting another item to be considered when designing a flush air data system.A NACA 0015 wing model was then tested in the University of Colorado Boulder low speed wind tunnel. The model had 31 pressure ports distributed chordwise on the pressure and suction surfaces to both build a pressure profile of the airfoil and to allow for choosing any 3 pressure ports as a flush air data system. The wind tunnel tests were run at angles of attack from 0 degrees to 10 degrees at speeds of 15 m/s, 20 m/s, and 25 m/s to replicate Reynolds numbers often seen on small UAVs (150,000, 200,000, 250,000). The wind tunnel tests confirmed the trends found in the XFOIL cases. An example FADS configuration was then investigated in more detail, with the pressure coefficient function being correlated to angle of attack using a fourth order polynomial fit. The resulting system was able to estimate angle of attack to an accuracy of approximately 0.2 degrees. The same system was also calibrated using a polynomial fit to the XFOIL inviscid model predictions, and it was found that calibrating using the XFOIL model resulted in angle of attack estimations accurate to within 1 degree for small angles.
590 $a School code: 0051.
650 4 $a Aerospace engineering. $3 1002622
650 4 $a Atmospheric sciences. $3 3168354
653 $a Flush air data system
653 $a Multi hole probe
653 $a Multi-hole probe
653 $a Relative sensitivity
653 $a Wind tunnel tests
653 $a Angle of attack
690 $a 0538
690 $a 0725
710 2 $a University of Colorado at Boulder. $b Aerospace Engineering. $3 1030473
773 0 $t Masters Abstracts International $g 81-03.
790 $a 0051
791 $a M.S.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13864485