Big Data Analytics for the Future Energy System

Dang Ha The Hien, Data Scientist at eSmart Systems, defended his PhD degree at the University of Oslo, Department of Informatics, November 23rd, 2018.

The thesis “Big Data Analytics for the Future Energy Systems” provides a deeper understanding of the current power grid system, its transition to the smart grid, and the role of big data analytics in the process.  After a well-executed dissertation in Oslo, Hien was congratulated by colleagues from eSmart Systems and representatives from the university.

About the the thesis

Big data Analytics for the Future Energy System

We are living in a world that is being shaped by algorithms, especially learning ones. Every one of us is experiencing the ever-increasing influence of learning algorithms on our daily behavior. For instance, 35% of Amazon product sales results from its recommender system, 75% of viewer activity in Netflix is driven by its recommendation engine, 70% of all trades in American markets is the result of algorithmic trading, and 40% of the prices of the 10,000 top products on Amazon is determined by a pricing engine that learns to set the product price based on customer demand, competitors prices, and other factors. The world will continue to be transformed by programs that are not written by us, but by learning machines that are trained on Big Data. With the appearance of these big data technologies, together with the recent development in networking infrastructure and sensor technologies, the human society is potentially witnessing the most significant change in the electricity value chain since its beginnings: the emergence of the future energy system, or better known as the smart grid.

This thesis provides a deeper understanding of the current power grid system, its transition to the smart grid, and the role of big data analytics in the process. By showing that the amount of data generated by a household in a smart grid is roughly equivalent to a Facebook user, or 0.3MB per day, the thesis emphasizes the big data challenge as well as how big data technology is needed to fuel the organic growth of a smart grid. This recognition leads to the proposal of Graph of Virtual Actors, a Big Data Analytics architecture for IoT (Internet of Thing) applications like smart grid. Furthermore, to demonstrate the potential of applying big data technologies in smart grid, the thesis proposes and experiments various novel and effective methods for automatic short-term load forecasting and automatic consumers segmentation.


  • Faculty: Mathematics and Natural Sciences
  • Department: Informatics
  • Area of Expertise: Data Analytics – Machine Learning
  • Supervisors: 
    • Professor Olaf Owe, Department of Informatics, University of Oslo
    • Associate Professor Jan Roland Olsson, Østfold University College

The Faculty of Economics and Management at University of Oslo has appointed the following adjudication committee to evaluate the thesis:

  • Professor Erik Dahlquist, Mälardalen University, Sweden
  • Professor Mohan Lal Kolhe, University of Agder, Norway
  • Senior Researcher Lizeth Tapia Tarifa, Department of informatics, University of Oslo