Small-Signal Synchronization Stability Enhancement of GFL-Based Renewable Energy Generation Using the Koopman Operator

Recent reports have highlighted small-signal synchronous instability issues in grid-following converter-based renewable energy generation systems connected to weak power grids. This study introduces a Koopman operator-based approach for system state prediction and small-signal synchronization stability enhancement through a data-driven strategy. Initially, system dynamics are identified and forecasted using measured data via delay embedding and the Koopman operator, effectively transforming the original nonlinear system dynamics into a more manageable linear framework in a higher-dimensional space. Subsequently, a supplementary control loop is implemented, and control variables are calculated employing model predictive control within the elevated Koopman space. This proposed technique is independent of the ‘white box’ model structure and parameters, thereby offering adaptability to changes in operational conditions. The effectiveness of this method has been confirmed through case studies conducted on a modified IEEE 39-bus system, demonstrating its viability.