The power outage monitoring in the low voltage platform area of the distribution network is an important part of the operation and management of the power system. The traditional power outage monitoring methods mainly rely on manual fault reporting and on-site inspections, which have problems such as slow response speed and inaccurate fault location. This article proposes an active monitoring method and system for power outages in low voltage substations in distribution networks based on multi-source data fusion. By accessing a big data platform, it integrates multi-source information such as medium voltage line tripping data from distribution network scheduling, power outage alarm data from distribution transformer operation terminals, customer fault reporting data, medium and low voltage distribution network structure data, and distribution transformer operation data to achieve active identification and alarm of various types of power outage events in low voltage substations in distribution networks. This method does not require additional hardware installation. Through data modeling, automatic monitoring and analysis of voltage and current faults, and confirmation of power outage event collection, it improves the accuracy and efficiency of power outage monitoring, reduces investment costs, and enhances the reliability of user electricity consumption.