A new approach in multidisciplinary design optimization of upper-stages using combined framework

Optimum design of an upper-stage with bipropellant propulsion system consists of optimization of three major subsystems including thruster, feeding subsystem, and propellant tanks. Optimization of such a complex system involved in optimization of many disciplines including structure, heat transfer, aerothermodynamics, guidance and control, trajectory and propulsion. Hard coupling of the disciplines increase the optimization processing times. Multidisciplinary design optimization algorithm can derive the optimum configuration but more elapsed time is needed for single-level methods such as all at once (AAO) and lower feasibility occurred in multi-level methods such as collaborative optimization (CO). In this paper, a new multidisciplinary design optimization framework is proposed for such coupled disciplines with concentrating on the propulsion system. The proposed framework uses Combined Single-level and Bi-level Optimizations (CSBO) frameworks to minimize numbers of design variables and system constraints when feasibility is increased. For this goal, modeling of every discipline is introduced and the design algorithm validated by redesigning of two real bipropellant thrusters. Three MDO frameworks are applied for our problem including AAO, CO and CSBO. Comparisons between the results show that CSBO can find the optimum solution in shorter elapsed time with lower F-count. Therefore, CSBO is more efficient for complex systems with coupled disciplines.