This paper designs an integrated two-safety fusion assessment model for the functional safety and information security of power monitoring devices. The model combines the fault tree model and the attack and defense tree model, establishes the fault attack and defense tree model, and calculates the integrated weight by the least squares method to realize the quantitative analysis of the fault attack and defense tree. Aiming at the problem of difficult compatibility between security and lightweight of privacy data of electric power intelligent terminal during model data interaction, homomorphic encryption algorithm with mode power optimization and model compression algorithm based on pruning are designed, and a secure and lightweight data interaction mechanism based on mode power optimization is proposed. The embedding of the two security fusion assessment model can find the key nodes affecting the vulnerability of the system by analyzing the examples to guarantee the functional security and information security of the system. The encryption algorithm used in this paper has lower complexity of data encryption, higher encryption efficiency, the more sensitive algorithm stays above 50%, and the lightweight grouping structure occupies less storage, and the network’s resource occupancy rate is lower. In the electric power monitoring device system can be simultaneously and effectively take into account the security and lightweight
