Low-Order Response Surface Modeling of Wind Tunnel Data Over Truncated Inference Subspaces

DeLoach, Richard; Erickson, Gary E.

doi:10.2514/6.2003-456

Cited by 16 publications

(18 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, dividing whole experiment range with some subspaces make each response surface model in subspaces more accurate. There is still many practical issues for how to divide subspaces for accurate response surface model [3]. The practical issues for applying subspace partitioning method shown in Table l.…”

Section: Subspace Partitioningmentioning

confidence: 99%

“…Hocking proposed simple method to resolve discontinuities between subspaces by fixing one variable making the response surface as a function of one variable [5], and the method applied in wind tunnel testing [3] but it is not applicable to making three-dimensional integrated response surface model. However, modeling aerodynamic coefficients in three-dimensional space is necessary to analyze aircraft effectively.…”

Section: Integrating Response Surfacesmentioning

confidence: 99%

“…DOE is actively applied to wind tunnel testing at NASA Langley Research Center for the purpose of reducing total operating time and cost [2][3][4]. In Korea, increased demand of wind tunnel testing for developing complicated flight vehicles and insufficient wind tunnel testing facilities also leads to utilize DOE actively.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Response surface smoothing for wind tunnel testing based on design of experiment with subspace partitioning

Lee

Ahn

2014

2014 14th International Conference on Control, Automation and Systems (ICCAS 2014)

View full text Add to dashboard Cite

This paper presents a new approach in resolving discontinuities between subspaces when the subspace partitioning method is applied to measure aerodynamic coefficients. The discontinuities between response surfaces in each subspace are hard to resolve in an analytical method, but it can be smoothened by overlapping between subspaces and integrating with proper interpolation. The traditional One Factor at a Time (OF AT) method is used for gathering aerodynamic force and moment coefficient data before applying the Design of Experiments (DOE) method for the purpose of later validation of the suggested method. In each subspace, the Central Composite Design (CCD) is used to reduce experimental points. The Separated Response Surface Model compared to the Integrated Response Surface Model. Quality of fit of the Developed Integrated Model calculated with OF A T data for validation. The test model base is the lambda wing configuration of the UCAV 1303. The study was carried out at the low speed wind tunnel of the Agency for Defense Development (ADD-LSWT).

show abstract

Section: Subspace Partitioningmentioning

confidence: 99%

Section: Integrating Response Surfacesmentioning

confidence: 99%

See 1 more Smart Citation

Response surface smoothing for wind tunnel testing based on design of experiment with subspace partitioning

Lee

Ahn

2014

2014 14th International Conference on Control, Automation and Systems (ICCAS 2014)

View full text Add to dashboard Cite

show abstract

“…13 proposes, as a rule-of-thumb, one-quarter to one-third of the number of points used to generate the response model. For both experiments previously discussed, one third of confirmation points were chosen which makes it 31 and 35 confirmation points, making a total of 153 and 172 data points for Experiments 1 and 2, respectively.…”

Section: Number Of Confirmation Pointsmentioning

confidence: 99%

Aircraft Wind Tunnel Characterization using Modern Design of Experiments

Dias

Suleman

Broughton

2013

54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference

View full text Add to dashboard Cite

The present work focused on the full characterisation of a blended-wing-body UAV airframe recurring to a wind tunnel testing approach known as Modern Design of Experiments (MDOE). The results of this project are going to be applied in the design of a new control system for the aircraft. The tests were completed in a low-speed wind tunnel facility at Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa. The model was prepared for the wind tunnel tests, which included structural analysis, reinforcement and instrumentation. As part of the MDOE technique, the design of experiment of the test program was considered and analysed prior to the testing itself. The tests were then executed and the data collected and analysed. The final results consist in 12 models that fit the data for the six aerodynamic responses (2 per response): Lift, Drag, Side Force, Pitching Moment, Rolling Moment and Yawing Moment coefficients. All the models were extensively analysed using confirmation points. An example of significant systematic variation for the drag coefficients was detected and corrected. A study of the bias and tolerance trade-off resulting from the data correction for the wind tunnel influences was also conducted.

show abstract

“…As such, the cumulative binomial probability distribution determines the number of successes that can be expected out of a total number of trials, given a probability of success of each trial and the Type I inference risk error probability for the experiment. 23 This number of expected successes is known as the critical binomial number and is used as an objective criterion for determining whether or not a response model is valid. If the number of successful confirmation trials is greater than or equal to the critical binomial number, than the response model is said to be valid -otherwise it is not.…”

Section: Confirmation Of the Response Modelmentioning

confidence: 99%

Hypersonic Wind Tunnel Calibration Using the Modern Design of Experiments

Rhode

DeLoach²

2005

41st AIAA/ASME/SAE/ASEE Joint Propulsion Conference &Amp;amp; Exhibit

Self Cite

View full text Add to dashboard Cite

A calibration of a hypersonic wind tunnel has been conducted using formal experiment design techniques and response surface modeling. Data from a compact, highly efficient experiment was used to create a regression model of the pitot pressure as a function of the facility operating conditions as well as the longitudinal location within the test section. The new calibration utilized far fewer design points than prior experiments, but covered a wider range of the facility's operating envelope while revealing interactions between factors not captured in previous calibrations. A series of points chosen randomly within the design space was used to verify the accuracy of the response model. The development of the experiment design is discussed along with tactics used in the execution of the experiment to defend against systematic variation in the results. Trends in the data are illustrated, and comparisons are made to earlier findings. = stagnation pressure, measured in settling chamber of the wind tunnel P T,2 = post-normal-shock stagnation pressure, or pitot pressure Re 2 = Reynolds number based on static conditions behind a normal shock t = time from start of data acquisition for a given run t α = t-statistic associated with the Type I inference error probability t β = t-statistic associated with the Type II inference error probability T T,1 = stagnation temperature, measured in the settling chamber of the wind tunnel X = axial location in the test section, referenced from upstream edge of injection opening y' = spanwise distance along pitot rake, referenced from centerline γ = ratio of specific heats λ = precision requirement for response prediction ρ 2 = static density behind normal shock ρ ∞ = freestream density σ = standard deviation of measurement

show abstract

Low-Order Response Surface Modeling of Wind Tunnel Data Over Truncated Inference Subspaces

Cited by 16 publications

References 10 publications

Response surface smoothing for wind tunnel testing based on design of experiment with subspace partitioning

Response surface smoothing for wind tunnel testing based on design of experiment with subspace partitioning

Aircraft Wind Tunnel Characterization using Modern Design of Experiments

Hypersonic Wind Tunnel Calibration Using the Modern Design of Experiments

Contact Info

Product

Resources

About