Risk assessment of an On Orbit Servicing mission based on space tug concept

This Master Thesis aims to perform a Risk Assessment of an On Orbit Servicing mission based on space tug concept during a conceptual design phase. FMECA and FTA methods are used as complementary techniques in order to identify all critical components and events during on orbit operations. After a brief introduction about history and definition of risk analysis, a literature review about on orbit servicing is developed. It focuses on three aspects: historical ancestors, modern technological demonstration missions and servicing opportunities and economic feasibility. After that, the space tug design concept is described as a derivation from existing projects with few additions. Risk assessment lays its foundation on a functional analysis, consisting of : functional tree, FFBD and functions/components matrix. These methods are inputs for a functional FMECA, which identifies potential failure modes, causes and effects and assigns scores to Occurrence, Severity and Mitigation in order to evaluate which failure should be prioritized during risk management thanks to a RPN. All critical functions are listed in a CFL that is the result of this method. FTA is based on the breakdown of the Top Event into Intermediate Events until reaching Basic Events which failure probability can be evaluated. A Standard Fault Tree is developed in order to describe mission operations. Then three analysis are performed: qualitative, quantitative and sensitivity. The first one aims to identify Minimal Cut Sets of Events that cause TE to occur. The second one consists of evaluate TE failure probability, which is a combination of BEs failure probabilities. The third one is based on importance measures, FIM and FUIM, which represent how much a MCS influence respectively TE failure probability or uncertainty. A core aspect of this work is the use of fuzzy set theory to evaluate each BE failure possibility. This theory is a well-established method used in several industrial applications, such as nuclear power plants or ships, and it can be used in this space application thanks to its general nature. So failure data are not derived from handbooks or historical data but from experts’ elicitation: space sector experts are selected and asked their opinion on each basic event failure possibility by using linguistic expressions. These expressions are translated into Triangular Fuzzy Numbers and TE failure possibility is calculated and then converted into probability, which is of 10^(-3) order of magnitude. Sensitivity analysis reveals that most of critical MCS are first order ones, which are MCS with a single event that can cause TE failure. In particular, both FMECA and FTA identified propulsion subsystem, docking/berthing equipment and refueling tools as the most impacting and weakest design point of all mission. So, their mitigation should be prioritized during following design phases and both methods should be iterated in order to constantly control risks.