Toward Hybridized Parameter Estimation for Boundary-Based Control of Synchronous Machines

Hybridized parameter estimation methods that fusedeterministic accuracy with data-driven adaptability remainrare in practice. Deterministic techniques provide physicallyinterpretable parameters but struggle to capture nonlinear,time-varying effects, while heuristic approaches adapt well tochanging conditions but often lack transparency and machineinvariance. This work proposes a hybridized estimationframework that unites these strengths to generate constraintaware,behavior-based parameters directly linked to EfficiencyMaximization (EM), Optimized Capability (OC), and ProtectionResilience (PR) limits. These parameters quantify deviationsbetween idealized machine models and real-world performanceunder disturbances, enabling real-time forecasting ofoperational trajectories. Embedded within a boundary-basedcontrol scheme, the approach supports adaptive setpointoptimization via nonlinear control, maintaining safe, reliable,and efficient operation across machine types, ratings, andmodes. The proposed framework provides a scalable foundationfor advanced control architectures in synchronous machineapplications.