Dynamic Modeling, Predictive Control and Performance Monitoring: A Data-driven Subspace Approach

摘要: I Dynamic Modeling through Subspace Identification.- System Identification: Conventional Approach.- Open-loop Closed-loop Identification of Matrix and Noise Model Using Data.- II Predictive Control.- Control: Data-driven Approach to III Control Performance Monitoring.- Loop Assessment: State-of-the-art MPC MIMO Feedback Assessment.- Prediction Error Assessment with LQG-benchmark from Data.