Abstract：Many methods are used to control three-phase unbalance in traditional distribution stations, but some adverse effects are ignored, such as the effect of commutation to loads, the count of use of commutation switch and the economy of distribution station area, etc. To solve above problems, a method based on load forecasting and commutation strategy is proposed to adjust loads. First, K-means is used to cluster historical daily load, then it forecasts short-term load based on historical datum which were counted by using Support Vector Machines, and calculates the degree of unbalance of three-phase load current when distribution transformer works. Finally, it establishes optimal commutation mathematical model which targets the least of unbalance degree of three-phase current and the least of usage frequency of commutation switch, and gets the best commutation scheme through genetic algorithm. This method can reduce line loss and three-phase load imbalance effectively, alleviate the problem of three-phase load imbalance in distribution area.
王春蘅, 韩笑, 罗维真. 基于负荷预测的配电台区三相不平衡治理研究[J]. 电气技术, 2019, 20(9): 10-13.
Wang Chunheng, Han Xiao, Luo Weizhen. Study on three-phase unbalanced load based on prediction in distribution area. Electrical Engineering, 2019, 20(9): 10-13.
2019, Vol. 20 
