Robust Optimization with Tchebysheff Decomposition for Mars Entry Probe Design
An evidence based robust design optimization method with Tchebycheff decomposition is proposed for micro Mars probe design under epistemic uncertainty. Super-formula based super-ellipse is used for the probe geometric configuration instead of the conventional sphere-cone configuration. Evidence based multi-objective optimization(MOO) method is used to optimally design the probe. The MOO problem is casted into a set of scalar optimization problems with Tchebycheff decomposition. Individuals are grouped with an adaptive clustering algorithm. In each cluster,individuals are analyzed with Proper Orthogonal Decomposition(POD) technique, and sorted by the ”energy” levels occupied by the components. A new population is produced by projecting the cluster centroid to the principal component vectors, modeling the distribution and reproducing new individuals. A strategy similar to steepest descend method in single-objective optimization is implemented for reproducing the new population, pushing forward the front to the true Pareto front. Performance and efficiency of the new algorithm are tested on a set of standard benchmark test problems. To reduce computational cost of evidence computation, an Evolutionary Binary Tree (EBT) algorithm and response surface model is employed. Finally, numerical simulation of a Mars micro probe heat shield design with the proposed optimization algorithm is presented.
Keywordsrobust optimization Tchebysheff decomposition Mars entry probe
Unable to display preview. Download preview PDF.
- 2.Wright, M.J., Tang, C.Y., Edquist, K.T., Hollis, B.R., Krasa, P., Campbell, C.A.: A review of aerothermal modeling for mars entry missions. AIAA Paper 443, 4–7 (2010)Google Scholar
- 5.Vasile, M.L.: A behavioral-based meta-heuristic for robust global trajectory optimization. In: IEEE Congress on Evolutionary Computation, CEC 2007, pp. 2056–2063. IEEE (2007)Google Scholar
- 7.Vasile, M., MInisci, E., Wijnands, Q.: Approximated computation of belief functions for robust design optimization. In: 14th AIAA Non-Deterministic Approaches Conference (2012)Google Scholar
- 9.Anderson, J.D.: Hypersonic and high temperature gas dynamics. Aiaa (2000)Google Scholar
- 10.Reagan, F.J., Anandakrishnan, S.M.: Dynamics of atmospheric re-entry. Aiaa (1993)Google Scholar
- 12.Zuiani, F., Vasile, M.: Multi agent collaborative search based on tchebycheff decomposition. In: Computational Optimization and Applications, pp. 1–20 (2013)Google Scholar