PRESENTATION
2017 - Users Conference - London - Transmission & Distribution
Data Science and Predictive Analytics in Virtual Power Plant Environment
The foundation of the virtual power plant built at the Faculty of Electrical Engineering of Czestochowa University of Technology is the PI System. The virtual power plant uses data from the weather stations, electricity meters, air sensors, energy storage and photovoltaic panels. One of the most important functionality of the virtual power plant is the ability to analyze the building’s electricity consumption profile. Using the available software within the PI System infrastructure we created statistical analysis profile of the electricity consumption. Additionally, electricity consumption forecasts were made utilizing different time spans and scheduling of building utilization by students and staff. The direct business impact of this functionality of the virtual power plant is that this has led to cost reduction in purchased electricity as the result of optimal tariff selection for the campus.
Industry
- Transmission & Distribution
Company
Czestochowa University of Technology
Speaker
Piotr Szelag
Company
Czestochowa University of Technology
Speaker
Sebastian Dudzik
Sebastian Dudzik Ph.D., Associate Professor, works at Czestochowa University of Technology, Faculty of Electrical Engineering.
His research interests include the applications of active infrared thermography, artificial neural networks and neuro-fuzzy models of heat exchange and non-destructive testing, as well as the application of machine learning algorithms in data mining systems. He participates in the virtual power plant project developed at Faculty; especially in the implementation of data processing algorithms for prediction purposes.
He has been the Head of the Institute of Optoelectronics and Measurement Systems at the Faculty of Electrical Engineering, CUT since March 2017.