Monitoring the status of power load equipment is crucial for the normal operation of the smart grid,howev-er,the clarity of the equipment is often low,making real-time warning challenging,especially in densely deployed states.We proposed an intelligent alarm method for load equipment based on multi-source time series feature fusion(IAM-MTSF).Utilizing equipment monitoring images and multi-source semantic information with temporal features,vi-sual feature reconstruction is guided by deep learning to classify faults in power load equipment.Experimental simu-lations demonstrate that the proposed method can achieve an accuracy of 93.6％in determining fault types,thereby proving its effectiveness in load equipment alarms and grid operation and maintenance.
维普
万方数据