Disclosed are an AC/DC bulk power grid dynamic security risk situation rolling prospective early warning method and system. The method comprises: constructing an original feature set of a TTC rapid evaluation model; generating a training sample set of the TTC rapid evaluation model on the basis of a power grid topological structure and prediction information in a future period of time; building the TTC rapid evaluation model on the basis of an SDAE and an ELM; generating a future state operating scenario, and on the basis of the TTC rapid evaluation model and a heuristic search method, calculating a prevention and control measure type required for ensuring system security; and carrying out hierarchical early warning on the operating scenario according to an operating state category of a power grid and the required prevention and control measure type. On the basis of deep learning technology, dynamic security risk situation rolling early warning can be rapidly carried out, a final hierarchical early warning result can reflect security of a system operating state more comprehensively, and effective guidance information can be provided for next prevention and control decision.
展开▼
