Relatore: dott.ssa Francesca Marta Lilja Di Lascio - Facoltà di Economia, Libera Università di Bolzano

Abstract :

We perform a copula-based analysis of thermal energy demand and its complex relationship with meteorological variables, such as solar radiation and temperature. The gathered data concern the district heating system (DHS) of the Italian city Bozen-Bolzano and the weather station S. Maurizio during 2014-2017. Since modern DHSs are sustainable energy distribution services that exploit renewable sources and avoid energy waste, in-depth knowledge of thermal energy demand, which is mainly affected by weather conditions, is essential to enhance heat production schedules and contribute to reducing climate change. Hence, we perform a three-step analysis that consists in i) removing serial dependence in each time series using autoregressive integrated moving average models, ii) estimating copula models on residual time series, and iii) analytically deriving the conditional copula-based probability function of thermal energy demand given meteorological variables. The investigation of the probability law of energy demand given weather scenarios, especially extreme climatic phenomena, makes it possible to provide useful insights on the production management phase of local energy utilities.