A new multidisciplinary approach to RLVs design
- Paper number
IAC-05-D2.4.01
- Author
Mr. Davide Bonetti, DEIMOS Space S.L., Spain
- Coauthor
Mr. Massimiliano Vasile, Politecnico di Milano, Italy
- Year
2005
- Abstract
Fully reusable launch vehicles (RLV) will probably replace expendable ones in the next years: today’s expendable launch vehicles have reached their technology plateau and a next generation of new solutions is required in order to reduce the cost of access to space by an order of magnitude or more. In particular new methodologies are required to optimally designe the next generation of transportation vehicles. This paper describes a multidisciplinary approach to the design of RLVs. In particular the analysis is focused on two stage to orbit RLVs with a waverider-based configuration and a rocket H2/O2 propulsive system. The target of a vertical takeoff from the Kourou European spaceport is a circular 400 km orbit. Most of the published waverider and sharp leading edge vehicles studies aim at the maximization of a single performance for a fixed design point. Just a few of them try to optimize a complete vehicle design taking into account even a simplified model for all the disciplines and nearly none try to simulate a mission to achive the best configuration for a given design task. Although in a simplified way, in this paper, trajectory control, thermal protection, propulsive system and aerodynamics analysis are performed for both launch and reentry phases for many different configurations. Since the different disciplines involved aim at maximizing their own objective function, a considerable optimization effort is required at subsystem level. Moreover in order to find a globally optimal solution at system level an additional effort has to be made to blend all the subsystems together. The aim of this paper is to search for optimal configurations in a wide hyperspace of design variables with minimal sampling of the constraint and objective functions associated to each discipline. All the disciplines are highly coupled and their merit functions, defining the optimality of the solutions, are in conflict. For this reason a multiobjective analysis is necessary to look for what it is called a Pareto frontier of optimal solutions. In order to quickly explore the design space and to reconstruct the Pareto set, metamodels have been developed for each different discipline. The aerothermodynamic analysis is simplified to longitudinal bidimensional wedge sections. The aerodynamic metamodel is an interpolation of a wide database of bidimensional flow analysis performed for many configurations in different Mach number and angle of attack conditions. The thermal protective system is the result of reentry trajectory optimization and since sharp leading edges produce extreme thermal loads, new advanced ceramic materials are involved. Sharp leading edges are one of the greatest technological challenge but they are also a key factor to improve performances of actual launch vehicles. Based on the results of the aerothermodynamic models, the design problem is then reduced to two phases: a simulation of vehicle dynamics during ascent to orbit and a simulation of the reentry path. Kriging metamodels have been chosen to represent these two phases. This seems to be a promising choice because with respect to more classical radial basis response surfaces, Kriging offer the great advantage to provide an estimation of the global error produced by representing the original problem with a metamodel. This allows a online adaptation of the response surface to better fit the problem. Then an evolutionary search, driven both by the local value of the response surface and by the estimated approximation of the design space, is preformed. This gives the possibility to incrementally make the metamodel as close as possible to the physical problem during optimization. Solutions found with this approach are then used to incrementally reconstruct the Pareto frontier for the launch and reentry objective functions
- Abstract document
- Manuscript document
IAC-05-D2.4.01.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.