Prognostic by Classification of Predictions Combining Similarity-Based Estimation and Belief Functions

摘要: Forecasting the future states of a complex system is paramount importance in many industrial applications covered community Prognostics and Health Management (PHM). Practically, can be either continuous (the value signal) or discrete (functioning modes). For each case, specific techniques exist. In this paper, we propose an approach called EVIPRO-KNN based on case-based reasoning belief functions that jointly estimates values signal modes. A real datasets used order to assess performance estimating break-down where combination both strategies provide best prediction accuracies, up 90%.