In order to provide methodological guidance for the load optimal distribution of thermal power units under the condition of deep peak regulation

the objective functions of optimal load distribution under medium-high load and low load are established respectively

and the load optimal distribution under different conditions is studied.The k-means clustering algorithm is used to process the data and the coal consumption characteristic curves of two units under medium-high load and low load are fitted. The objective function of load distribution is established to compare the coal consumption before and after optimization under static load distribution

incremental load distribution and continuous variable load conditions. After setting the total load instructions of the two units under the condition of deep peak shaving

the coal consumption of the two schemes of load average distribution and optimal distribution is compared. The results show that the clustering-based objective function models of optimal distribution under medium-high load and low load established by the coal consumption characteristic curve are reliable

whose optimization effect is remarkable under the condition of deep peak regulation. The k-means clustering algorithm can be used to optimize the load distribution of thermal power units and provide reference for the study of load optimal distribution under the condition of deep peak regulation of thermal power units.