Prediction of aerodynamic characteristics of an aircraft model with and without winglet using fuzzy logic technique

• Main Authors: Hossain, Altab, Rahman, Mohammed Ataur, Hossen, Jakir, Iqbal, Parvez, M.I., Zahirul
• Format: Article
• Language: English
• Published: Elsevier 2011
• Subjects: TL500 Aeronautics
• Online Access: http://irep.iium.edu.my/7597/1/Journal_of_Aerospace_Science_and_Technology_%28ISI%29_Listed_%5B15%282011%29%5D.pdf; http://irep.iium.edu.my/7597/; http://www.sciencedirect.com/science/article/pii/S1270963810001586
• Institution: Universiti Islam Antarabangsa Malaysia

This paper describes the potentials of an aircraft model without and with winglet attached with NACA wing No. 65-3-218. Based on the longitudinal aerodynamic characteristics analyzing for the aircraft model tested in low subsonic wind tunnel, the lift coefficient (CL ) and drag coefficient (CD) were investigated respectively. Wind tunnel test results were obtained for CL and CD versus the angle of attack α for three Reynolds numbers Re (1.7 × 105, 2.1 × 105, and 2.5 × 105) and three configurations (configuration 1:without winglet, configuration 2: winglet at 0◦ and configuration 3: winglet at 60◦). Compared with conventional technique, fuzzy logic technique is more efficient for the representation, manipulation and utilization. Therefore, the primary purpose of this work was to investigate the relationship between lift coefficients and drag coefficients with free-stream velocities and angle of attacks, and to illustrate how fuzzy expert system (FES) might play an important role in prediction of aerodynamic characteristics of an aircraft model with the addition of winglet. In this paper, an FES model was developed to predict the lift and drag coefficients of the aircraft model with winglet at 60◦. The mean relative error of measured and predicted values (from FES model) were 6.52% for lift coefficient and 4.74% for drag coefficient. For all parameters, the relative error of predicted values was found to be less than the acceptable limits (10%). The goodness of fit of prediction (from FES model) values were found as 0.94 for lift coefficient and 0.98 for drag coefficient which were close to 1.0 as expected.