SYSTEMATIC ANALYSIS OF MULTIVARIATE SCENARIOS USING ADVANCED CLUSTERING METHODS

The continuous improvement of fuel cycle simulators in conjunction with the increase computing capacities have led to a new scale scenario studies. Taking into consideration multiple variable parameters and observing their effect on evaluation criteria, these studies regroup several thousands trajectories paving different possible values for operational parameters. If global methods like sensitivity analysis allow extracting useful information from groups trajectories, they only provide average values. In this work we present method analyze while keeping some localization information. Based principal component analysis, clustering been implemented order mathematically extract, ensemble simulated study, subgroups that similar behaviors. Typical representative subgroups, are then determined. application sample two output, total amount transuranic elements within number time MOX could not be built during time, is presented. comparison results between analyses shows allows good regular outputs but struggle discrete highly non-linear ones.