The invention discloses a high-speed train multi-parameter intelligent threshold monitoring method, comprising the steps: firstly collecting the working condition parameters and state monitoring dataof the high-speed train, and dividing the state monitoring data according to the working condition parameters; analyzing the state monitoring data by using fuzzy kernel clustering to obtain various center coordinates and intra-class distances; and finally, calculating a regression relationship among the center coordinate, the intra-class distance and multiple working condition parameters, furtherintelligently calculating thresholds under various working conditions, and carrying out intelligent alarm on the high-speed train. The high-speed train multi-parameter intelligent threshold monitoringmethod is simple and feasible, and is suitable for determining the intelligent threshold of the high-speed train monitoring data under variable working conditions.