DEGGENDORF, GERMANY - Modern societies heavily depend on electrical energy. This dependency increases with diminishing reserves of fossil fuels and resulting changes to energy consumption, such as a shift to electric mobility. For emerging markets, energy consumption has been substantially increasing in recent years due to increased economic wealth.
As some customers want to illegally reduce their electricity bill, electric utilities all around the globe have been facing the issue of electricity theft. In emerging markets, however, electricity theft is a particular issue and ranges up to 40 percent of total electricity distributed.
In addition to the loss of profit, thieves may also damage power infrastructure, creating extra costs for electric utilities. This article will show how artificial intelligence (AI) helps to detect electricity theft and how explainable machine learning (ML) can help when making potentially expensive automated decisions. We will also discuss why it will take more than just AI to permanently reduce electricity theft.
Losses in Power Systems
When producing and distributing electric power, some of it gets lost and offers no contribution to the profit margin of electric utilities. These transmission and delivery losses can be divided into technical losses and non-technical losses.
Technical losses occur mostly due to power dissipation, which is caused naturally by internal electrical resistance in generators, transformers and transmission lines. In general, technical losses are around 1-2 percent of the total electricity distributed in efficient systems, such as in Western Europe. In less efficient systems in emerging markets, they can range up to 10 percent.
In turn, non-technical losses are caused by external entities and are usually the major share of the overall lossThe content herein is subject to copyright by The Yuan. All rights reserved. The content of the services is owned or licensed to The Yuan. The copying or storing of any content for anything other than personal use is expressly prohibited without prior written permission from The Yuan, or the copyright holder identified in the copyright notice contained in the content.