Multiple Regression for UAV Aerodynamic Model based on Experimental Data

Grounding the completeness of experimental aerodynamic model can be done with statistical criteria (determination criterion or Durbin - Watson criterion). This article offers a schematic analysis of the aerodynamic model by statistical synthesis of structural and parametric design solutions based on multiple regression. It provides current development of the method of structural-parametric (two-level) selection of design solutions for Unmanned Aerial Vehicles (UAV) based on a statistical synthesis. The experimental model reconstructed from the "aerodynamic channel" is evaluated in terms of its completeness using the statistical criteria. The method of statistical synthesis of structural and parametric design decisions is based on models and methods of the theory of selforganizing complex systems. This paper applies this method to select a new circuit design analysis of the aerodynamic model of UAV. We show the results of the analysis of the aerodynamic model of UAV by channels C_x and C_z. We estimated the multicollinearity of the multiple regression model using the determinant of the matrix of pairwise correlation coefficients between the factors.

References

[1] Balyk V.M., Kalutskiy N.S., Kulakova R.D. Strukturno-parametricheskaya samoorganizatsiya slozhnykh tekhnicheskikh system. [Structural-parametric self-organization of complex engineering systems]. Delovaya slava Rossii [The Inter-Industry Almanac], 2007, iss. 2, pp. 58-63.

[2] Lysenko L.N., Koryanov V.V., Raykunov K.G. The theory of stochastic orbit correction applied in the planning design and ballistic support of interplanetary missions. Tr. MGTU im. N.E. Baumana № 606: Aktual’nye problemy razvitiya raketno-kosmicheskoy tekhniki i sistem vooruzheniy [Proc. of BMSTU No. 606: Actual problems of development of rocket and space technology and weapon systems]. Moscow, MGTU im. N.E. Baumana Publ., 2013, pp. 94-115 (in Russ.).

[3] Kalugin V.T., Mordvintsev G.G., Chernukha P.A. Chislennyy metod rascheta aero-dinamicheskikh kharakteristik letatel’nykh apparatov pri sverkhzvukovom obtekanii [Numerical method for calculating aerodynamic characteristics of aircraft in supersonic flow]. Moscow, MGTU im. N.E. Baumana Publ., 2007. 34 p.

[4] Tarasov E.V., Balyk V.M. Metody proektirovaniya letatel’nykh apparatov [Methods of aircraft design]. Moscow, MAI Publ., 2006. 96 p.

[5] Tarasov E.V., Balyk V.M., Loginov A.B. Vybor oblikovykh kharakteristik sistem LA [Selecting aircraft system conceptual characteristics]. Moscow, MAI Publ., 1998.

[6] Balyk V.M. Statisticheskiy sintez proektnykh resheniy pri razrabotke slozhnykh system [Statistical synthesis of design choice in the complex system development]. Moscow, MAI Publ., 2011. 278 p.

[7] Suslov V.I., Ibragimov N.M., Talysheva L.P., Tsyplakov A.A. Ekonometriya [Econometrics]. Novosibirsk, SO RAN Publ., 2005. 744 p.

[8] Tarasov E.V., Balyk V.M., Ustinov S.A., Shipov O.V. Metody optimizatsii oblikovykh kha-rakteristik tekhnicheskikh ob"ektov na primere LA i JICA [Methods of optimization of technical object conceptual characteristics on the example of aircrafts and JICA]. Moscow, MAM Publ., 1992.

[9] Anatolyev S. Durbin - Watson statistic and random individual effects // Econometric Theory (Problems and Solutions), 2002-2003.

[10] Nguyen K.T., Nguyen V.T. Numerical solution of unmanned aerial vehicles (UAVs) according to the experimental data. Naukoemkie tekhnologii [Science Intensive Technologies], 2016, no. 3, pp. 11-20 (in Russ.).