Skip to Main content Skip to Navigation
Theses

OPTIMISATION MULTICRITERES DE L'EFFICACITE PROPULSIVE DE MINI-DRONES BIOMIMETIQUES A AILES BATTANTES PAR ALGORITHMES EVOLUTIONNAIRES

Abstract : Multicriteria optimisation of the kinematics of a flapping wing mini UAV, in cruise flight, has been investigated. The goal of this work is to find, for different families of kinematics and various advance velocities, solutions that simultaneously maximize the propulsive efficiency, minimize the departure of the lift from the weight and minimize the norm of the aerodynamic moment. We used evolutionnary algorithms to solve this multicriteria problem because of their simplicity of implementation, their flexibility and their efficiency in solving these kind of problems. As a consequence of the nature of the problem, there exist multiple solutions to it, which raises the question of finding a way to visualize, analyze and extract a unique solution from the whole group of Pareto optimal solutions. We identified, in the litterature, simple methods able to help us solve this issue and used them in our study. The \textit{Scatter-Plot Matrix Method} helped us visualizing the Pareto surfaces, while multiple regression allowed us to establish a quantitative link between the optimized criteria and the kinematical parameters. Futhermore, the method of level diagrams helped us finding a compromise solution and the technology of decision trees dealt with identifying the critical kinematical parameters to which the near neighborhood of this solution was sensible, whereas the use Kohonen maps uncovered the structure of this near neighborhood. Using these tools, for each family of kinematics (dihedral, dihedral and pitch, dihedral and pitch with two panels), we identified a compromise solution and the kinematical parameters that control its near neighborhood. The characteristics of each compromise solution obtained were successfully compared to power measurements and drag coefficients for birds in cruise flight and the legitimity of using a linearized flight physics model was asserted by computing reduced frequencies and Strouhal numbers for these solutions and noticing that the latters were favorable to such an approach. Finally, the comparison of the flapping frequency between our compromise solutions and a geometricaly similar bird showed that increasing the complexity of the kinematics allows to get closer to the flapping frequency of the bird, which is a quite encouraging result for our approach.
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00550437
Contributor : Mohamed Hamdaoui <>
Submitted on : Tuesday, December 28, 2010 - 1:10:50 AM
Last modification on : Friday, March 22, 2019 - 1:32:16 AM
Document(s) archivé(s) le : Tuesday, March 29, 2011 - 2:41:18 AM

Identifiers

  • HAL Id : tel-00550437, version 1

Citation

Mohamed Hamdaoui. OPTIMISATION MULTICRITERES DE L'EFFICACITE PROPULSIVE DE MINI-DRONES BIOMIMETIQUES A AILES BATTANTES PAR ALGORITHMES EVOLUTIONNAIRES. Sciences de l'ingénieur [physics]. Université Pierre et Marie Curie - Paris VI, 2010. Français. ⟨tel-00550437⟩

Share

Metrics

Record views

1241

Files downloads

2588