Abstract:Aiming at the problem of low accuracy of electricity sales prediction in different periods and modes, an accurate prediction method of electricity sales in distribution station area based on particle swarm optimization algorithm is designed. Establish grey model, mine data, input power load data, normalize, extract capacity utilization characteristics, quasi variable prediction problem expression, establish power sales growth trend model, calculate curve fitting problem, use particle swarm optimization algorithm to design the accurate prediction process of power sales in distribution station area, optimize weight transformation function and particle weight, and complete the accurate prediction of power sales in distribution station area. The experimental results show that this method has high accuracy in one month''s electricity sales, one year''s electricity sales, low-frequency modal prediction and high-frequency modal prediction, and meets the demand of accurate prediction of electricity sales in distribution station area.