Uncertainty-based computational robust design optimisation of dual-thrust propulsion system

摘要

This paper proposes a robust design optimisation methodology for performance evaluation of high-fidelity propulsion systems under both aleatory and epistemic uncertainties. In this paper, uncertainty-based robust design methodology is applied using first-order orthogonal design to estimate worst case of the uncertainties. The effectiveness and computational efficiency of proposed formulation is obtained by decoupling computation of both types of uncertainties in the optimisation process loop. The system robustness is measured by directly minimising the variance of target mean and standard deviation and adhering to performance constraints. A parametric sensitivity analysis of performance parameters due to input parameters variations is also performed to identify the most significant design variables. A hybrid genetic algorithm and simulated annealing approach are used as an optimiser. By using the proposed method, the objective function to maximise the booster phase to sustain phase thrust ratio and total impulse could be achieved while adhering to the motor performance constraints to provide an insensitive solution.