Construction and Analysis of a Markov Chain-Based Quantification Model for Power System Regulation Capability

Abstract

With the large-scale integration of renewable energy, power system regulation capability faces severe challenges in maintaining operational balance, electromagnetic stability, and reliable interaction among power-electronic interfaces. This paper proposes a Markov chain-based quantification model for power system regulation capability. By constructing a state transition probability matrix, the model characterizes the evolution of system states under the participation of multiple regulation resources. The model considers the fluctuation characteristics of wind and photovoltaic output as well as the dynamic participation of energy storage and demand response. Validation using 15-minute operational data from the State Grid Northwest Regional Grid in 2023, where renewable energy accounted for 48% of installed capacity, shows that the proposed model can accurately assess regulation capability margins at different time scales, with an average error of 2.75% compared with actual operating data. The model provides a theoretical basis and practical tool for the optimal configuration and dispatch of regulation resources. It is also relevant to widearea electromagnetic sensing and communication-supported grid monitoring, where accurate state prediction is essential for secure operation in high-renewable power systems.