Machine learning for the detection of electricity theft in emerging markets: Patrick Glauner скачать в хорошем качестве

Machine learning for the detection of electricity theft in emerging markets: Patrick Glauner

Electricity theft is a particular issue in emerging markets such as Brazil, India or Malaysia and may range up to 40% of the total electricity distributed. The world-wide annual losses to electric utilities are estimated to be USD 100 billion. Electricity theft is caused for example by meter tampering, bypassing meters or manipulation of transformers. In this talk, we will show how we use machine learning for the detection of electricity theft in a joint university-industry research project. Next, we will discuss the challenges we encountered when applying machine learning to real-world data and how we solved them. Last, we will also show how the output of our project creates real value for utilities.