Application of data mining methods in energetics

OData support
Supervisor:
Dr. Vokony István
Department of Electric Power Engineering

Modern database management systems are already able to cope with the task of collecting and storing data, and can keep up with the growing demand. There is almost no question why it is worth retaining data that is constantly cumulating, as it is self-evident for everyone to have a vast amount of valuable information in an ever-expanding data stream. The real question is rather how to access this valuable information hidden in the data sea.

To solve this challenge data mining, a new discipline of the modern business world, was created. Data mining is a technology that evaluates knowledge or compliance through a database. To be effective, the analysis is very useful for maintenance and service management. In such an environment data mining is one of the key technologies for managing a large database. As a result, it is necessary to develop a regular computer-assisted method for managing the complexity of the data and drawing conclusions.

My thesis focuses on the potential of data mining. In my work I examine what opportunities lie ahead in the field of energy to exploit the technology based on technical as well as economic and business considerations.

First of all the technological diversity, analytical requirements and methodological practices of data mining solutions are reviewed. Then it is shown what the distribution network companies do and what data are available to perform these tasks. Later, based on these, sub-areas are highlighted where it is worthwhile to use data mining technologies and to evaluate the most significant ones in terms of feasibility. Finally, AMR customers, whom connected to the distribution networks of E.ON Hungária Zrt., are clustered in order to better describe the consumption habits of these users.

Downloads

Please sign in to download the files of this thesis.