Frequency Control of Decoupled Synchronous Machine Using Koopman Operator Based Model Predictive

TL;DR

This paper proposes a data-driven Koopman operator-based model predictive control method for frequency regulation in decoupled synchronous machines, demonstrating its effectiveness in a two-area power system.

Contribution

It introduces a novel Koopman operator framework for nonlinear control of decoupled synchronous machines without requiring detailed device models.

Findings

Effective damping of frequency oscillations demonstrated

Control approach validated on Kundur two-area system

Offers a practical solution when device models are unavailable

Abstract

Conventional generators have been retired or replaced by renewable energy because of the utility long-standing goals. However, instead of decommissioning the entire plant, the rotating mass can be utilized as a storage unit to mitigate the frequency issues due to these changes in the grid. The goal is to design a control utilizing the retired machine interfaced with the grid through a back to back converter referred to as decoupled synchronous machine system (DSMS) to damp frequency oscillations. However, in a practical setting, it is often not possible for a utility to obtain access to the detailed state equations of such devices from the vendor making the addition of another layer of control a challenging problem. Therefore, a purely data driven approach to nonlinear control design using Koopman operator based framework is proposed for this application. The effectiveness of the proposed system is demonstrated in the Kundur two-area system.