Network security situation awareness plays an important role in the power information network. The main functions of it include detecting and identifying network attacks in time, determining the status of the network, and predicting the future direction of network security status. The availability and reliability of the power information network system is closely related to the accuracy of the network security situation awareness platform and the function of multi-source data integration. This paper firstly completes the architectural design of network security situation awareness platform based on the current situation of power information network, researches the multi-source heterogeneous security data fusion technology, realizes the network security monitoring data convergence function of the awareness platform, and solves the problem of a single source of network security information data; Secondly, for the low accuracy of the existing network security situation awareness platform being used by power information network, we designed and constructed a much improved power information network security situation awareness platform based on the AM-LS-CNN algorithm to realize real-time sensing of the network security situation of the power information network, and enhance the ability to identify potential network security risks.