The filters tuning is a crucial issue due the need to quantify the accuracy of the model in terms of the process noise covariance matrix for process characterized by structural uncertainties which are time-varying. Thus, approaches to time-varying covariances were studied and included to a traditional EKF and an optimization-based state estimators constrained EKF (CEKF) formulations. The results for these approaches have shown a significant improvement in filters performance. Furthermore, the performance of these estimators as a transient data reconciliation technique has been appraised and the results have shown the CEKF suitability for this proposes.