Analysis of the electricity demand patterns of a building in a university Campus

The scope of this paper is to investigate the demand patterns of buildings of the Polytechnic School of the Aristotle University of Thessaloniki (AUTH) Campus in Greece. Electrical consumption data are obtained by the installed Supervisory Control and Data Acquisition (SCADA) system and the period of study is one year. Apart from offices, the building accommodates various laboratories for teaching and research purposes that incorporate a wide diversity of electrical equipment, such as electrical machinery, heating and cooling systems, small scale electronics, high voltage experimental laboratory devices, etc. The load profiling methodology is utilized for the purpose of extracting the representative daily load curves. The clustering of the load curves is done using the Kohonen Self-Organizing Maps (SOM) of various topologies. A detailed optimization of the SOM's parameters takes place for the purpose of the minimization of the clustering error.