General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms

No abstract

This article presents the application of nonlinear (simulation-based) and linear (structured singular value) worst-case tools to the VEGA launcher Verification and Validation (V&V) process, during atmospheric ascent. The simulation-based worst-case evaluation is performed by minimizing a set of cost functions that capture launcher performance objectives, using the Worst-Case Analysis Optimization Tool (WCAT) and a high-fidelity nonlinear simulator of VEGA. The linear worst-case search uses the structured singular value (µ) and a linear fractional transformation (LFT) model representing the yaw rigid-motion of the VEGA launcher but numerically evaluated using time simulation data from the VEGA simulator. In order to facilitate the analysis of the worst-case results as well as the comparison between the two analysis tools, a selection of the most critical uncertainties is performed using sensitivity analysis based on selected nonlinear simulator time responses. It is highlighted that the presented analysis tools are complementary to traditional Monte-Carlo approaches in that they strive to identify worst-case uncertainty combinations as opposed to providing probabilistic guarantees on performance metric satisfaction. In addition, as it will be shown, these approaches require only a fraction of the time required to perform a Monte Carlo campaign.

This article presents the application of optimization-based worst-case search approaches to the VEGA launcher during the P80 ascent phase. Four optimization algorithms are applied covering evolutionary and deterministic global methods as well as their hybrid versions, i.e. mixed with local approaches. A comparison with a traditional Monte Carlo campaign shows that the optimization-based algorithms are much amenable to worst-case search.

No abstract