Research on State Evolution of Power Grid Based on Random Matrix Theory Indicators

With fast development of the ubiquitous power internet of things, the limitations of model-based methods of power grid state evolution are gradually revealed, which are always of high time complexity and highly depend on expert experience. By mining the evolution characteristics of the grid operation data, this paper proposes three random matrix theory indicators considering three perspectives to evaluate the operation state of the power grid. Firstly, the nodal voltage is used instead of the line load ratio to construct a high-dimensional data matrix; with the use of random matrix theory, the evaluating indicators are constructed to correctly map the factors affecting the evolution of the grid state to data changes. Finally, through the simulation example of the IEEE39-bus system, the proposed random matrix indicators and the evaluation method can effectively characterize the state of the grid, which can meet the requirements of real-time dynamic monitoring of power grid.