Answers are grounded in this paper and cite their sources.
State estimation (SE) is the most critical part of power systems management and control centers because correct data from the equipment in the network is needed before any operation. Power systems in the past were less complex than today's systems, so simple methods were sufficient to solve the SE problem. As power systems have developed and distribution systems have become more interconnected to enhance reliability and handle growing loads and uncertainties, the methods for solving the SE problem have evolved over time. Many methods have been employed so far to address the state estimate problem; each of these techniques has the potential to be helpful in particular situations; therefore, identifying the strategies and getting familiar with their features can be crucial. Based on this issue, SE methods are divided into two categories. In one category, their dynamic and static characteristics are specified, and another category is based on the application of methods. Most of the methods have been studied, and the advantages and disadvantages of each have been thoroughly investigated to identify their strengths and weaknesses. The methods based on artificial intelligence (AI) can have good potential in solving SE problems, so they have been specifically investigated. This category of SE methods can be beneficial in solving future problems. Based on this, the existing challenges for the future of SE have been discussed. As a case study, we demonstrate how AI techniques, such as transfer learning (TL), can address one of these challenges—specifically in handling network reconfiguration in a 118-bus system. This example can guide those interested in the field to tackle similar challenges and provide direction for future research.
No takes yet. Share an insight, caveat, or question.
Ranjbar et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: