Abstract: Extreme climate conditions pose severe threats to the safe operation of substations.Traditional separate environmental monitoring and equipment monitoring approaches fail to effectively capture the coupling interaction patterns between these two aspects.To address this challenge,an environment-equipment coupling monitoring and intelligent early warning system for substations has been developed.The system synchronously collects meteorological parameters,equipment operating parameters,and infrastructure status through multisource heterogeneous sensors.A hierarchical progressive architecture is employed to achieve efficient data transmission and processing.An environment-equipment coupling interaction model has been established,deepening the understanding of equipment response mechanisms under extreme climate conditions.Building upon this foundation,data fusion,feature extraction,and machine learning technologies are utilized to identify and classify extreme climate events,establishing a multi-parameter correlation-based early warning rule base.Field application at a 220kV substation demonstrates that the system achieves 87%accuracy in identifying equipment failures induced by extreme climate,with an early warning lead time exceeding 35minutes.The system effectively enhances the safe operation and maintenance capability of power systems under complex climate