Disturbance-exploitation based RMPC with application to multi-objective load frequency control in power systems

Abstract

This paper introduces a robust dual-model predictive control strategy for enhancing load frequency control in networked power systems. By seamlessly integrating a disturbance observer with a nonlinear model predictive control framework, our approach offers two main advantages: incorporating disturbance estimation into the optimization process and separately addressing robustness and performance. A nonlinear disturbance observer is developed to estimate drift from external power imbalances and unknown dynamics, updating the prediction model for accurate estimation. The dual-model nonlinear predictive control strategy is then introduced, ensuring input-to-state stability (ISS) to enhance frequency regulation and lower power generation costs. The performance of the proposed control method has been evaluated using the Speedgoat real-time simulator and compared with various control strategies under diverse scenarios. Experimental results indicate that the proposed method provides superior dynamic response and robust characteristics.