.png){kind=icon}
.png){kind=icon}
Return
Data-driven robust moving horizon estimation for nonlinear systems using deep Koopman operators
.png){kind=icon}
PRE
AI
DOI：10.1016/j.compchemeng.2026.109862.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A stable Koopman model is learned from noisy data to form a linear surrogate model for a class of nonlinear processes.
• Robust Koopman-based MHE avoids non-convex optimization in nonlinear estimation.
• Stability conditions are derived to ensure robust global asymptotic convergence of estimates.
• Case studies confirm good modeling and estimation over EDMD-based Koopman estimation method.
Keywords:
Moving horizon estimation
Koopman modeling
Deep learning
Robust state estimation
Journal
C
IF:
3.9
Papers:
191
Citations:
0