Input-Output Selection for LSTM-Based Reduced-Order State Estimator Design

In this work, we propose a sensitivity-based approach to construct reduced-order state estimators based on recurrent neural networks (RNN). It is assumed that mechanistic model available but too computationally complex for estimator design and only some target outputs are of interest should be estimated. A can estimate the sufficient address such problem. We introduce an sensitivity analysis determine how select appropriate inputs data collection data-driven development desired accurately. Specifically, consider long short-term memory (LSTM) network, type RNN, as tool train model. Based it, extended Kalman filter, estimator, designed outputs. Simulations carried out illustrate effectiveness applicability proposed approach.