Data-Driven Economic Predictive Control for Nonlinear Systems
Analysis
This paper presents a novel data-driven control approach for optimizing economic performance in nonlinear systems, addressing the challenges of nonlinearity and constraints. The use of neural networks for lifting and convex optimization for control is a promising combination. The application to industrial case studies strengthens the practical relevance of the work.
Key Takeaways
- •Proposes a data-enabled economic predictive control method for nonlinear systems.
- •Uses neural networks for lifting to approximate nonlinearities.
- •Formulates the control problem as a convex optimization problem.
- •Demonstrates effectiveness through industrial case studies (water treatment, carbon capture).
Reference
“The online control problem is formulated as a convex optimization problem, despite the nonlinearity of the system dynamics and the original economic cost function.”