With the advent of the era of big data in power

higher requirements have been placed on the accuracy of power load forecasting. Accurate power load forecasting is of great significance for the safe and stable operation of power systems and reducing costs. Aiming at the characteristics of short term power load showing different load operation rules in different time periods

the similarity of load behavior in different time periods is calculated in the daily range. Based on the consideration of the weather dimension and the time dimension

the behavior dimension is added. The behavior similarity factors of different time periods are introduced into the long short term memory network model

and the future load data is predicted based on the historical load data. Experimental simulation proves that considering the characteristics of load behavior in different time periods can effectively improve the accuracy of load prediction.