New methodologies for constructing steady-state security regions of electric power systems

Abstract

This paper presents two alternative methodologies of the Steady-State Security Region (SSSR) construction process. This tool is widely used for security assessment of electric power systems, especially in the real-time environment. In this context, the main objective of this work is to improve the methodology for obtaining SSSRs, which is used in several commercial programs, such as the network analysis software ANAREDE, developed by the Electric Energy Research Center (CEPEL), in Brazil. The first methodology presented in this paper consists on adding a voltage stability indicator to the conventional SSSR, based on the Thévenin and load impedance computations. The SSSR simultaneously evaluate a significant amount of physical and operational limits of the network, such as voltage level, thermal limits on transmission lines and transformers, among others. However, the current methodologies for obtaining the SSSR do not specifically possess an indicator of the voltage stability condition of the analyzed system. On the other hand, electric power systems currently operate near their operating limits. This is due to the continuous increase in load and the restrictions (economic, environmental etc.) that limit proper expansions, bringing systems closer to voltage instability conditions. Therefore, adding a voltage instability indicator in the SSSR becomes a useful and necessary feature for the tool. The second technique proposed in this work aims to significantly reduce the computational effort of the conventional methodology, creating a new procedure based on the Particle Swarm Optimization (PSO) algorithm. A series of modifications and adjustments to the parameters used in the PSO algorithm are required to adapt it to the construction of SSSRs. The results show that the proposed methodology is reliable, since the impact of the randomness used in the PSO algorithm does not have a significant impact on the quality of SSSRs, and also demonstrate that the proposed methodology presents a significant computational gain when compared to the conventional methodology used for obtaining the SSSRs.