Electrical load profile analysis and peak load assessment using clustering technique

Load profile analysis in different regions is very useful to power utilities for managing the load requirements in economic and efficient manner. For the demand side management and grid operation, the variation in demand is to be known. In this paper, classical k-means clustering approach is used for finding similar types of profiles of a practical system for demand variation analysis and energy loss estimation. For different zones, typical load profiles based on similar consumption are obtained. Primarily, the load factor represents feeder demand variation, and loss factor helps average energy loss estimation in distribution power system without load flow studies. In this paper, a concept is proposed for analysis of electricity consumption pattern on different days in particular zones based on cluster load factor and cluster loss factor. Normal and abnormal peak load requirements in cluster of similar types of profile of days of different zones are identified. Cluster loss factor helps in identifying the energy loss variation due to different load patterns.