TY - JOUR
TI - Linear correlation and regression between the meteorological data and the electricity demand of the Dubrovnik region in a short-term scale
AU - Falkoni Anamarija
AU - Krajačić Goran
JN - Thermal Science
PY - 2016
VL - 20
IS - 4
SP - 1073
EP - 1089
PT - Article
AB - Intermitted patterns of solar and wind generation cause insecurity in the  power system supply since they depend on weather conditions. The aim of this  work is to perform linear regression and correlation between solar radiation,  wind speed, air temperature and electricity demand of the Dubrovnik region.  Gained results could help in system energy planning with a high share of  renewable energy sources in the electricity production. All of the data are  collected for consecutive three year period, years 2012, 2013 and 2014, in  the 10 minute time step. Results of the correlation of each of the parameters  individually in between the years show slightly variations between the  distributions, providing representative line-ar relation in between the years  for all the data, except the wind speed data. Cor-relations are also done  between all of the parameters for each year separately, based on the mean  monthly values. Result showed good relation with negative correlation between  solar radiation and wind speed, as well as good relation with positive  correlation between solar radiation and electricity demand. The same  correlations are done in the 10 minute time step and including time system  delay. The results indicate significant decrease in correlation coefficient  value and it is less possible that they can be pronounced with linear  regression line. Calculations of the correlation and regression, based on the  10 minute time step for summer and winter period separately, gained slightly  better results in relation between parameters than the ones including the  whole year data.