Scenario aggregation and analysis via mean-shift methodology

A challenging aspect of dynamic methodologies, such as the Dynamic Event Tree (DET) methodology, is the large number of scenarios generated for a single initiating event. Such large amounts of information can be difficult to organize in order to extract useful information. The scenario dataset is composed of scenarios which contain information on the system components and the system process variables, such as values of pressures and temperatures for the reactor coolant system and the containment throughout the time period of the transient. In order to facilitate analysis, it can be fruitful to accomplish two tasks: i) identify the scenarios that have a "similar" behavior (i.e. identify the most evident classes), and, ii) decide, for each event sequence, to which class it belongs (i.e., classification). It is shown how it is possible to accomplish these two tasks using the Mean-Shift Methodology. The Mean-Shift methodology is a kernel-based, non-parametric density estimation technique that is used to find the modes of an unknown distribution, which corresponds to regions with highest data density. The methodology is illustrated by applying it to the DET analysis of a simple level controller.