Structural Optimization of Composite Rotor Blade for Unmanned Helicopter using Variational Asymptotic Beam Sectional Analysis and Multi-Objective Bat Algorithm

This work presents a structural design optimization of composite rotor blades for a typical unmanned helicopter. Multiple parameters that influence the stiffness and weight of the rotor blade are considered as design variables. These include the number of composite plies, the thickness of composite plies, and the web location as a fraction of the rotor blade airfoil chord. To further improve the stiffness, carbon nano-tubes are doped into the composite material and the percentage of doped carbon nano-tubes are considered as an additional design variable. The structural analysis of the rotor blade is carried out using principle of the Variational Asymptotic Method (VAM). Using VAM, the 3D elastic rotor blade problem is decoupled into two reduced-dimension sequential problems of 2D cross-sectional analysis and 1D beam analysis. Variational Asymptotic Beam Sectional Analysis (VABS) tool is used for the 2D cross-section analysis to obtain the blade stiffness. Orthogonal array based design of experiments approach is followed to efficiently explore the design space. A novel orthogonal array, named MGB-4P OA which has 18 sample design points, is used to construct the design of experiments. VABS is used for cross-sectional analysis at the 18 sample design points of the MGB-4P orthogonal array. Second order polynomial nonlinear response surface models are generated for the blade stiffness and blade weight using their respective values from the orthogonal array. The response surface models predict the output of the structural analysis from VABS and the estimated blade weights accurately, and are therefore, considered as surrogate models for the objective functions during optimization. A recently developed nature inspired meta-heuristic Multi-Objective Bat-Algorithm (MOBA) optimization routine is employed in conjunction with Pareto analysis for optimizing the design variables for the mutually conflicting objectives of maximizing the rotor blade stiffness and minimizing the blade weight.