Estimation-Based Predictive Control of Nonlinear Processes Using Recurrent Neural Networks

Abstract

Abstract Machine learning techniques have demonstrated their capability in capturing dynamic behavior of complex, nonlinear chemical processes from operational data. As full state measurements may be unavailable in chemical plants, this work integrates recurrent neural networks (RNN) within extended Luenberger observers to develop data-based state estimators. Then, an output feedback model predictive controller is designed based on state estimates provided by the RNN-based estimator to stabilize the closed-loop system at the steady-state. A chemical process example is utilized to illustrate the effectiveness of the proposed state estimation approach.