A Network Partitioning Method for Real-Time Simulation of Power System Based on State-Space and Branch Cutting

Abstract

With the sustainable development of the power system, the dynamic characteristics of the system are becoming increasingly complex. The electric network partitioning method is an effective approach for implementing real-time simulation of complex power systems. This paper proposes a novel electric network partitioning method based on state space and branch cutting, aiming to enhance efficiency and precision of real-time simulation. In the proposed method, the entire system is partitioned into multiple subsystems through branch cutting. Each cut branch is modeled based on Kirchhoff's voltage law and the characteristics of components. Each subsystem is solved using the state-space method, and all subsystems can be simulated in parallel. Compared with traditional network partitioning methods, the proposed method effectively reduces the computational cost resulting from integrating state-space and branch cutting methods. Additionally, the proposed method eliminates the need for time delay, which can ensure high precision. Finally, the advantages of the proposed method are verified through several real-time simulations of an RLC circuit, a grid-connected inverter system with an LCL filter, and a boost-inverter system.