Sensitivity Analysis of Missile Model

In this report, a sensitivity analysis of a simplified model of a surface-to-airmissile (SAM) is presented. The purpose of the work is twofold. First, a literature study in the area of sensitivity analysis for aircraft and missile models is conducted, where the state-of-the-art in this area is examined and summarized. Secondly, a sensitivity analysis of a given missile model is performed with respect to a number of different input parameters. The focus of this analysis is on how much of the model variation that can contributed to the individual parameters, and to identify the parameters that have the greatest impact on the model behavior. The purpose of this work is to understand how perturbations in the parameters affect the model's behavior, and the results from this study are intended to aid the development of new missile models. The sensitivity analysis in the report is made with respect to the missile performance. The range of the missile is used as performance measure, both in the form of launch diagrams as well as the range in a single direction. Launch diagrams provide information on the range in all different directions around the launch site, but it is computationally expensive to estimate. The range in a single direction were therefore evaluated as a less computationally expensive alternative. Various methods for sensitivity analysis, with different requirements of computational capacity, were also studied. All the studied methods show consistent results, where the drag coefficient was identified as the single most influential parameter. This is followed by various engine parameters, such as specific impulse, burn time, and fuel mass. The only exception is the burn time of the booster, which indicates that the power development in the booster is more important than for how long time the rocket stage burns. The results are compiled in a separate section, in order to facilitate the use in future model development. For better understanding, the results are also visualized in different launch diagrams and scatter plots. The work in this report is based on a previous analysis, where launch diagrams were also used as performance measure. This work is a continuation and extension of that analysis, and the aim is to clarify and visualize the results as well as examining less computationally demanding objective functions.