Optimization of 3D printed drone performance using synergistic multi algorithms

Abstract

The optimization of 3D-printed drone performance remains a significant challenge in achieving both efficiency and cost-effectiveness. Traditional single-algorithm approaches often fail to address the complex interplay between structural dynamics and energy efficiency requirements. Current approaches lack integrated optimization techniques that could complement each other's strengths in addressing these multifaceted challenges in drone design and performance. This paper presents a novel multi-algorithm optimization framework integrating Response Surface Methodology with Box-Behnken Design (RSM-BBD), Artificial Neural Networks (ANNs), and Osprey Optimization Algorithm (OOA). The approach systematically optimizes three critical parameters: frame length (20-30 cm), DC brushless motor (1250-1750 kV), and flight duration (3-5 minutes). Finite Element Analysis (FEA) validates the structural integrity of the optimized components under operational stresses. The integrated framework demonstrates complementary strengths across methods. RSM-BBD achieved superior results in battery consumption optimization with R² = 0.9801, while OOA excelled in stability enhancement by determining optimal parameters (20 cm frame length, 1456.47 kV brushless motor, and 3.69 minutes flight duration). ANNs effectively captured overall system dynamics with R² = 0.97362, providing comprehensive performance prediction. This synergistic approach establishes a new paradigm for drone optimization, maintaining manufacturing costs at $115.79 while delivering significant performance improvements. The framework provides a foundation for future developments in cost-effective, high-performance drone manufacturing and design optimization.