Varying horizon learning economic MPC with unknown costs of disturbed nonlinear systems

Key Points

• Control strategies in nonlinear systems can be optimized through varying horizon learning, enhancing economic outcomes.
• Key evidence shows that improved performance is achieved under specific cost disturbance conditions, improving efficiency.
• This observational analysis assesses how unknown costs affect performance in nonlinear systems using economic model predictive control techniques.
• Findings support the notion that tailoring learning algorithms significantly influences control outcomes in disturbed systems.