Redesign of morphing UAV's winglet using DS algorithm based ANFIS model

Purpose The purpose of this paper is to present a novel approach based on the differential search (DS) algorithm integrated with the adaptive network-based fuzzy inference system (ANFIS) for unmanned aerial vehicle (UAV) winglet design. Design/methodology/approach The winglet design of UAV, which was produced at Faculty of Aeronautics and Astronautics in Erciyes University, was redesigned using artificial intelligence techniques. This approach proposed for winglet redesign is based on the integration of ANFIS into the DS algorithm. For this purpose, the cant angle ( c ), the twist angle ( t ) and taper ratio (λ) of winglet are selected as input parameters; the maximum value of lift/drag ratio ( E max ) is selected as the output parameter for ANFIS. For the selected input and output parameters, the optimum ANFIS parameters are determined by the DS algorithm. Then the objective function based on optimum ANFIS structure is integrated with the DS algorithm. With this integration, the input parameters for the E max value are obtained by the DS algorithm. That is, the DS algorithm is used to obtain both the optimization of the ANFIS structure and the necessary parameters for the winglet design. Thus, the UAV was reshaped and the maximum value of lift/drag ratio was calculated based on new design. Findings Considerable improvements on the max E are obtained through winglet redesign on morphing UAVs with artificial intelligence techniques. Research limitations/implications It takes a long time to obtain the optimum E max value by the computational fluid dynamics method. Practical implications Using artificial intelligence techniques saves time and reduces cost in maximizing E max value. The simulation results showed that satisfactory E max values were obtained, and an optimum winglet design was achieved. Thus, the presented method based on ANFIS and DS algorithm is faster and simpler. Social implications The application of artificial intelligence methods could be used in designing more efficient aircrafts. Originality/value The study provides a new and efficient method that saves time and reduces cost in redesigning UAV winglets.