Stepwise Regression Analysis of MDOE Balance Calibration Data Acquired at DNW

DeLoach, Richard; Philipsen, Iwan

doi:10.2514/6.2007-144

Cited by 6 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The details will not be described here in any depth, but they can be found in standard references on response surface modeling [2][3][4] . Specific details for strain-gage balance calibration applications can also be found in the literature [5][6][7][8][9] . The rejection/retention criterion for a given term is typically associated with some threshold probability that the estimated response would significantly change if that term were rejected, given the level of random error in the calibration data.…”

Section: Nomenclaturementioning

confidence: 99%

A Statistical Theory of Bidirectionality

DeLoach¹,

Ulbrich²

2013

AIAA Ground Testing Conference

Self Cite

View full text Add to dashboard Cite

Section: Nomenclaturementioning

confidence: 99%

A Statistical Theory of Bidirectionality

DeLoach¹,

Ulbrich²

2013

AIAA Ground Testing Conference

Self Cite

View full text Add to dashboard Cite

“…6,7 It should be noted that MDOE design matrix optimization is also well-suited for automated balance calibration machines. 8,9 Such machines are capable of applying multiple loading combinations without certain physical constraints inherent in a traditional dead-weight calibration system, although at the expense of some additional complexity in alignment. 10 Independent of the MDOE activity at Langley, NASA Ames Research Center has investigated a key aspect of the design matrix optimization problem with respect to balance calibration; namely, the selection of math models that can be supported by a given load schedule and that also minimize the residual variance when the model is fitted to the calibration data.…”

Section: Introductionmentioning

confidence: 99%

“…where ( ) X X X X H ′ ′ = −1 (8) and X is the design matrix, as before. The variance in the vector of response estimates is computed as follows:…”

Section: Introductionmentioning

confidence: 99%

A Comparison of Two Balance Calibration Model Building Methods

DeLoach

Ulbrich

2007

45th AIAA Aerospace Sciences Meeting and Exhibit

Self Cite

View full text Add to dashboard Cite

Simulated strain-gage balance calibration data is used to compare the accuracy of two balance calibration model building methods for different noise environments and calibration experiment designs. The first building method obtains a math model for the analysis of balance calibration data after applying a candidate math model search algorithm to the calibration data set. The second building method uses stepwise regression analysis in order to construct a model for the analysis. Four balance calibration data sets were simulated in order to compare the accuracy of the two math model building methods. The simulated data sets were prepared using the traditional One Factor At a Time (OFAT) technique and the Modern Design of Experiments (MDOE) approach. Random and systematic errors were introduced in the simulated calibration data sets in order to study their influence on the math model building methods. Residuals of the fitted calibration responses and other statistical metrics were compared in order to evaluate the calibration models developed with different combinations of noise environment, experiment design, and model building method. Overall, predicted math models and residuals of both math model building methods show very good agreement. Significant differences in model quality were attributable to noise environment, experiment design, and their interaction. Generally, the addition of systematic error significantly degraded the quality of calibration models developed from OFAT data by either method, but MDOE experiment designs were more robust with respect to the introduction of a systematic component of the unexplained variance. = Difference between total sum of squares and explained sum of squares Significance = Risk of erroneously rejecting a null hypothesis SVS = Single Vector System t-Limit = Minimum acceptable probability that a given term is significant t-Value = Measured quantity expressed as multiple of standard error in measurement Total SS = Total sum of squares. Measure of variability in data set VIF = Variance Inflation Factor/ A measure of multicollinearity

show abstract

“…In recent years, several studies have been conducted based on SVS. Deloach and Philipsen (2007) applied the Stepwise Regression Analysis on data obtained from SVS. They compared the obtained results with OFAT.…”

Section: Introductionmentioning

confidence: 99%

Development and Evaluation of Calibration Procedure for a Force-Moment Balance Using Design of Experiments

Nouri

Mostafapour

2016

Lat. Am. j. solids struct.

View full text Add to dashboard Cite

To measure the aerodynamic and hydrodynamic forces and moments acting on scaled models in water and wind tunnel tests, multi-component strain gauge force and moment balance are usually used. Their performance and accuracy largely depends on the rig and calibration method, including the load table (design matrix) and analysis of experimental data. In this article, for the calibration of a six-component balance, a calibration procedure using the BoxBehnken design (BBD) of experiments was developed. In the suggested design matrix, in addition to all possible combinations of the two-component load of the six component load (Test vectors with two active factors), the pure loads (test vectors with one active factor) are also used. The implementation of the design matrixwas done using a calibration rig, which has the ability to perform formal experimental design techniques completely. The obtained experimental data were fitted with second-order equations using regression analysis. The statistical significance of independent variables and interactions was tested using the analysis of variance (ANOVA) with 95% confidence (α = 0.05). The results of residuals indicate that the suggested model sufficiently predicts the responses as a function of input factors. The comparison between the Modified Box-Behnken design (MBBD) and BBD calibrations indicates that the MBBD method estimates the data more accurate. The results show that the MBBD method is the most appropriate method compared to the existing methods for calibrating balance in this paper.

show abstract

Stepwise Regression Analysis of MDOE Balance Calibration Data Acquired at DNW

Cited by 6 publications

References 8 publications

A Statistical Theory of Bidirectionality

A Statistical Theory of Bidirectionality

A Comparison of Two Balance Calibration Model Building Methods

Development and Evaluation of Calibration Procedure for a Force-Moment Balance Using Design of Experiments

Contact Info

Product

Resources

About