Statistical Load Time Series Analysis for the Demand Side Management

The paper presents a brief summary of the study which was carried out as part of the demand response (DR) project in the scope of Slovenian-Japanese NEDO project. The purpose of this study was to examine the possible annual substation (SBS) peak load decrease before actual DR activation in order to assess the possible benefit of the future program. SBS load time series data were thoroughly examined with various types of statistical diagrams. The daily load profiles were analyzed with the unsupervised machine learning. With 50 hours of DR activation available per year, the annual peak could be decreased for around 5 %. Since the load is highly dependent on temperature, normalized daily peak load was calculated with supervised machine learning. It can be seen throughout the paper that advanced statistical diagrams and machine learning techniques allow better assessment of future the DR program.