CEU Electronic Theses and Dissertations, 2019
| Author | Pásztor, Benedek |
|---|---|
| Title | Day-ahead prediction of photovoltaic power plant production |
| Summary | This capstone summary presents the work conducted as a capstone project at E.ON Energiatermelő Kft. The focus of the work is on the bonus production of a Phovoltaic Power Plant (PVPP), Szigetvár Gamma. Bonus is given to a PVPP owner by the electricity market operator depending on how punctually the owner is capable of predicting the electricity production of the power plant by 10 AM, previous day (scheduling a power plant). After a comprehensive analysis of possibly available data, its utilization on the domain, an automatically refreshing machine learning (ML) algorithm has been put in production. The ML gets refreshed each day with weather and production data of the previous day. The newly built ML shows a learning curve from its first time in production in March 2019 until June 2019. Compared to the original scheduling process of the company, the ML in production has been able to reach 25% higher bonus, giving extra revenue to the company. |
| Supervisor | Pál, Jenő |
| Department | Economics MSc |
| Full text | https://www.etd.ceu.edu/2019/pasztor_benedek.pdf |
Visit the CEU Library.
© 2007-2021, Central European University