Ensemble Bagged Tree Based Classification for Reducing Non-Technical Losses in Multan Electric Power Company of Pakistan

• DOI: 10.3390/electronics8080860
• Publication Date: 2019-08-02T15:58:16Z
• Authors: Muhammad Salman S...; Mohd Wazir Mustafa; Usman Ullah Sheikh; Touqeer Ahmed Jumani; Nayyar Hussain Mi...
• Keywords: Ensemble Learning; Metering mode; Tree (set theory)

Non-technical losses (NTLs) have been a major concern for power distribution companies (PDCs). Billions of dollars are lost each year due to fraud in billing, metering, and illegal consumer activities. Various studies explored different methodologies efficiently identifying fraudster consumers. This study proposes new approach NTL detection PDCs by using the ensemble bagged tree (EBT) algorithm. The is an many decision trees which considerably improves classification performance individual combining their predictions reach final decision. relies on energy usage data identify any abnormality consumption could be associated with behavior. key motive current provide assistance Multan Electric Power Company (MEPCO) Punjab, Pakistan its campaign against stealers. model developed this generates list suspicious consumers irregularities further examined on-site. accuracy EBT algorithm found 93.1%, higher compared conventional techniques such as support vector machine (SVM), k-th nearest neighbor (KNN), (DT), random forest (RF)