A data-driven power system security evaluation technology

As the fields of science and technology advance and electricity consumption rises, the role of power system safety evaluation becomes ever more pivotal. This scholarly article introduces a novel power system security evaluation model that is data-driven. The methodology commences with an assessment of the static power flow, utilizing the PQ decomposition technique and the Newton-Raphson method. Subsequently, error features are meticulously extracted, followed by the application of a stack classifier, which is a novel contribution of this research, for the classification process. Ultimately, the efficacy of the proposed approach is substantiated through simulations using the IEEE 39-node model. The results demonstrate a notable enhancement in speed while maintaining a relatively low error margin.