KL-Regelungstechnik-Seminar / 19. Juli 2022, 16:00 Uhr - 17:00 Uhr
Hints on the Model Predictive Control (MPC) Using Long Short Term Memory (LSTM) As Predictors
Referenten: Dr. Paulo da Costa Mendes und Marvin Jung (beide Fraunhofer ITWM)
Abstract der Präsentation:
[nur in Englisch verfügbar]
Hints on the Model Predictive Control (MPC) using Long Short Term Memory (LSTM) as predictors (Hinweise zur modellprädiktiven Steuerung (MPC) unter Verwendung von Langzeitspeichern (LSTM) als Prädiktoren)
The prediction model is the most important part of a MPC strategy. The accuracy of such model influences the quality of predictions and control performance of the algorithm. In some practical cases, a model based on physical equations is not available, or is not easy to get all parameters, or its complexity could affect the real-time computation of the control signal. For this reason, the use of black-box models within a MPC framework becomes attractive, since to fit such models only input and output data are needed.