Abstract: In order to improve identification accuracy,the paper proposes a non-invasive household load decomposition method based on GRNN and attention mechanism model. Batch standardization is adopted to reduce the coupling between the layers of neural network,and GRNN’s powerful time series feature representation ability is used to extract the relationship between the total electricity consumption measured at power load access point and the power consumption of each electrical appliance. Meanwhile,the attention mechanism is used to reduce the weight parameters of the model. Finally,a numerical example is provided to verify the feasibility and superiority of the algorithm.