Distributionally Robust Frequency Constrained Low-Carbon Scheduling Considering Flexibility of Carbon Capture Power Plants

Abstract

The large-scale integration of renewable energy source (RES) and the adoption of carbon capture technologies provide enhanced economic benefits and flexibility in scheduling low-carbon power systems (LCPS). However, this is accompanied by concerns regarding the significant uncertainty associated with RES and the frequency stability of LCPS. This paper proposes a distributionally robust frequency constrained low-carbon scheduling (DRFC-LCS) method, achieving coordinated optimization of coal-fired power plants (CFPP), carbon capture power plants (CCPP), energy storage systems (ESS), and RES within LCPS, ensuring post-disturbance frequency security while minimizing economic costs and carbon emissions. The dynamic frequency response of LCPS following a power imbalance is analyzed, incorporating frequency support from various generation resources into the frequency constraints. The complex frequency nadir constraint is converted into a linear surrogate expression using a support vector machine (SVM). A sample-independent distributionally robust chance-constrained (SI-DRCC) method based on the Wasserstein metric is introduced to manage the dual uncertainties associated with RES in power generation and primary frequency response (PFR) reserve provision. The effectiveness of the DRFC-LCS method is validated using the IEEE-39 bus and IEEE-118 bus test systems. Results indicate that incorporating frequency constraints in the scheduling of LCPS is crucial, and fully utilizing the regulation capacity of CCPP ensures secure operation of LCPS while delivering significant economic benefits.

Keywords: low-carbon power system, frequency constraint, distributionally robust chance constraint, carbon capture power plants