Renewable Energy Communities (RECs) are local aggregations of energy users who can share electricity generated from Renewable Energy Sources (RES) to increase onsite self-consumption. The resulting increased use of RES makes the local balance between energy generation and demand a more challenging task. Indeed, RES have a strong variability associated with their spatial and temporal availability, and an inherent uncertainty associated with variations in weather conditions. This uncertainty can affect any choice on the design and operation of a REC, depending on which of all possible scenarios will occur. This paper aims at evaluating the impact of uncertainty in weather variables (i.e., solar irradiance and ambient temperature) on the optimal life cycle cost (i.e., investment and operation costs) of a REC. The novelty consists of assessing the accuracy of the optimal solutions under uncertainty for different locations of the REC, corresponding to Italian cities with different climatic conditions (i.e., Padova and Palermo). First, the 'best' set of daily stochastic scenarios (i.e., the most representative) of the weather variables is obtained by applying a clustering technique on a 'past' dataset (2005-2008). A Stochastic Programming (SP) model with the 'best' set of scenarios is then used to find the design and operation of the system that minimizes its life cycle cost in a 'future' dataset (2009-2010). The optimal 'stochastic forecasts' solutions are compared with the state-of-the-art 'deterministic forecasts', obtained by solving a Mixed-Integer Linear Programming (MILP) model for average seasonal days. Results show the higher accuracy of the 'stochastic forecasts' in predicting the life cycle costs (errors of 3.5% and 5% for Padova and Palermo, respectively, versus errors of 30% of the 'deterministic forecasts'), taking as reference the 'perfect forecasts' obtained by solving the MILP model with 'perfect knowledge' of data occurred (i.e., for 365 days of a year)
How Uncertainty Affects the Optimal Design and Operation of Renewable Energy Communities: An Italian Case Study
volpato gabriele
;rech sergio;andrea lazzaretto
2024
Abstract
Renewable Energy Communities (RECs) are local aggregations of energy users who can share electricity generated from Renewable Energy Sources (RES) to increase onsite self-consumption. The resulting increased use of RES makes the local balance between energy generation and demand a more challenging task. Indeed, RES have a strong variability associated with their spatial and temporal availability, and an inherent uncertainty associated with variations in weather conditions. This uncertainty can affect any choice on the design and operation of a REC, depending on which of all possible scenarios will occur. This paper aims at evaluating the impact of uncertainty in weather variables (i.e., solar irradiance and ambient temperature) on the optimal life cycle cost (i.e., investment and operation costs) of a REC. The novelty consists of assessing the accuracy of the optimal solutions under uncertainty for different locations of the REC, corresponding to Italian cities with different climatic conditions (i.e., Padova and Palermo). First, the 'best' set of daily stochastic scenarios (i.e., the most representative) of the weather variables is obtained by applying a clustering technique on a 'past' dataset (2005-2008). A Stochastic Programming (SP) model with the 'best' set of scenarios is then used to find the design and operation of the system that minimizes its life cycle cost in a 'future' dataset (2009-2010). The optimal 'stochastic forecasts' solutions are compared with the state-of-the-art 'deterministic forecasts', obtained by solving a Mixed-Integer Linear Programming (MILP) model for average seasonal days. Results show the higher accuracy of the 'stochastic forecasts' in predicting the life cycle costs (errors of 3.5% and 5% for Padova and Palermo, respectively, versus errors of 30% of the 'deterministic forecasts'), taking as reference the 'perfect forecasts' obtained by solving the MILP model with 'perfect knowledge' of data occurred (i.e., for 365 days of a year)
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
