Abstract: In view of the accuracy and efficiency of fault identification in substation inspection, this paper proposes a substation joint inspection fault identification system supported by multi-source fusion data. The system constructs a multi-source heterogeneous data acquisition module of "space-time-domain" in one, realizes data standardization and efficient storage through data preprocessing and storage modules, integrates multi-source information with the help of multi-dimensional data fusion modules (data layer, feature layer, and decision layer fusion), relies on intelligent fault diagnosis modules to complete multilevel fault diagnosis, and initiates hierarchical disposal through fault response and decision modules. At the same time, the key technologies such as multi-source data collection and collaboration strategies, data fusion algorithms and models, intelligent fault diagnosis mechanisms, and real-time response and collaborative decision-making are elaborated in depth. The performance test shows that the fault identification accuracy of the system is 98.2%, the false alarm rate and false alarm rate of 0.02times/day and a missed alarm rate of 0.007times/day, and the average response time is only 8.7s, which is significantly better than the traditional method, which provides a strong guarantee for the safe and stable operation of the substation.