Implementation of GF-HOG Technique for Effective Commercial and Industrial Load Clustering and Classification for Better Demand Response

Abstract

With increased penetration of renewables since last decade has evolved measures from the regulator for robust distribution network mostly catering to residential load. With increasing future demand of commercial and industrial load (CIL) due to aspiring GDP growth and insistence of popular governments to encourage renewable use, large chunk of the CIL will be served by renewables. So increasing the robustness alone on the renewable supply side will be in vain unless effective Demand response with rationalized tariff system for CIL which is more profitable energy market than subsidized residential tariff in India .The present method employed for clustering and classifying Load profile of CIL loads with wild swings will impede effective demand response implementation due to tariff cartelization in favor of not so popular energy systems. The method of GF-HOG technique with SVM will adopt cluster and classification of pattern of Load profiles for better recognition generated out from large CIL data. This will help produce more vibrant tariff structure for effective CIL load management. The algorithms were tested on a local network which 70% -30 % of load in favor of CIL and found to generate better DR.