Preserving data redundancy in state estimation through a predictive database

This paper presents strategies for preserving data redundancy in state estimation through forecasting-aided state estimators (FASE). Forecasted states/measurements are obtained by an artificial neural network-based model. Many aspects of the pseudomeasurement provision problem are considered, regarding the use of forecasted measurements. The following questions emerge naturally. Numerical results covering the application of the proposed strategies under different levels of redundancy deterioration are presented and discussed.