Electricity Consumption and Load Profile Segmentation Analysis for Rural Micro Grid Customers in Tanzania

Understanding electricity consumption behavior of different groups of customers is important to ensure appropriate microgrid designs. This paper assesses segmentation among microgrid customers in Tanzania according to electricity consumption behavior. Using the k-means clustering algorithm, 821 customers are placed into clusters by mean daily electricity consumption and mean normalized load profiles. We divided customers into three distinct segments by daily consumption level: <; 140, 140-450 and > 450Wh. We also found five distinct load profile clusters. Clusters with high daytime electricity use tend to be commercial customers that consume large amounts of electricity. Households exhibit large evening peaks and consume relatively small amounts of electricity. Four of five load profiles show distinct evening peaks in consumption and one shows high night time electricity use. Microgrid developers can use these clusters and customer segments either to optimize the load profile for a power system by mixing certain customer types, or optimize the power system design to a given customer base.