Research On Three-phase Load Unbalance Based On Clustering And Combinatorial Optimization

The problem of three-phase load imbalance is common in my country’s low-voltage distribution network,and the long-term imbalance of three-phase load will cause a series of problems such as reduced power supply quality,economic losses,and safety hazards to the power system.With the access of high-power household loads such as new energy vehicles,the frequency of three-phase load imbalance in my country’s distribution station area and the difficulty of governance have increased.Among them,the improvement of governance difficulty is mainly reflected in the diversity of users’ electricity consumption behaviors caused by different household loads.However,there are few three-phase load imbalance treatment technologies that can be better designed for the diversity of users’ power consumption behaviors.To this end,this paper proposes a phase sequence optimization allocation algorithm based on the clustering results of electricity consumption behavior,which improves the traditional uniform allocation algorithm,and adjusts the phase sequence of users through cluster-by-cluster optimization to achieve a balanced state for three-phase loads.Firstly,the user’s long-term load curve is selected as the clustering object,and the load curves with similar shapes and magnitudes are divided into one cluster to achieve the purpose of distinguishing the electricity consumption behavior of different users.Then based on the clustering results,the phase sequence optimization allocation algorithm aims to minimize the three-phase load unbalance,and adjusts the phase sequence of users through the idea of class-by-class phase sequence adjustment and phase-by-phase phase sequence optimization,thereby reducing the three-phase load unbalance..The results show that only the phase sequence adjustment for high power consumption users can basically make the three-phase load imbalance in the distribution station area basically reach the grid standard.At the same time,in order to meet the different needs in practical applications,the algorithm can adjust the optimization target in real time according to the adjusted user phase sequence,and continue to adjust the phase sequence of users in the category with low average power consumption,at the cost of more phase sequence switching times,the unbalanced degree of three-phase load is further reduced.In the example analysis,the average three-phase load unbalance in the distribution station area is 20.2%,and only need to adjust the phase sequence of 15%of the users with high power consumption can reduce the three-phase load unbalance to 9.8%,lower than 15% of the grid standard.Further adjustments to the phase sequence of all low-power users can reduce the average three-phase load imbalance to 3.1%,but the phase sequence of 62.1% of users needs to be adjusted.In actual engineering,staff can flexibly choose a phase sequence adjustment scheme according to their own needs.