STATCOM optimal siting strategy based on probabilistic modeling of historical scenarios and EHPO-DBSCAN scenario reduction

The high penetration of renewable energy significantly increases the stochasticity and complexity of power system operating scenarios, imposing challenges on voltage stability assessment such as high data dimensionality, heavy computational burden, and insufficient pertinence of reactive power configuration. To address the voltage support optimization problem in current systems, this paper proposes a comprehensive STATCOM siting strategy based on historical scenario probabilistic modeling and improved DBSCAN scenario reduction. Firstly, a joint probabilistic model of wind speed, generation, and load is constructed based on historical data. The Copula function is employed to analyze the nonlinear dependence among multiple variables, and high-fidelity three-dimensional operating scenarios are generated via conditional resampling. Meanwhile, a two-layer architecture for the fault set is constructed. Secondly, a scenario reduction method fusing automatic parameter-tuning DBSCAN clustering with the Enhanced Hunter-Prey Optimization (EHPO) is proposed. Finally, an improved Mahalanobis distance to construct voltage control partitions and a preliminary screening process by synthesizing the dominance index and V-Q sensitivity are employed for lowering data dimensionality, and a joint steady-state and transient evaluation system based on the Transient Voltage Severity Index (TVSI) and L-index is established to determine the optimal siting scheme. Test results on the IEEE 39-bus system demonstrate that the proposed method can effectively extract typical scenarios and identify high-risk faults with high computational efficiency, ensuring that the optimized STATCOM configuration improves the transient voltage characteristics and steady-state security margin of the system.