.png){kind=icon}
Return
Fast mixed kernel canonical variate analysis for learning-based nonlinear model predictive control
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cherd.2025.05.050.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• MKCVA is proposed as an alternative to RNNs for MPC with small training data sets. • New techniques reduced MCKVA simulation time from > 10 seconds to milliseconds. • Bayesian optimized kernels can fine-tune MKCVA’s smoothness to generalize better. • MKCVA-MPC gives at par or better control than RNN-MPC in an evaporator case study. • Under the chosen implementations, MKCVA-MPC can run faster than LSTM-MPC.
Journal
IF:
3.9
Papers:
9.0K
Citations:
2.1W
Organization
No organization information available
Cited Papers
.png){kind=icon}
ShareSave
ShareSaveShareSaveShareSave
View more