A risk-averse approach for the planning of a hybrid renewable energy system

We propose a risk-averse two-stage stochastic programming for a hybrid renewable energy system planning problem, where we model the risk-aversion using Conditional Value at Risk (CVaR). The aim of this study is to find the optimal capacities of the system components in a cost e ective way while considering the risk-aversion of the decision maker. Renewable energy sources that are utilized in our hybrid system are solar and hydro, while the diesel fuel is used as a backup source. We assume that the water in ow to the reservoirs is uncertain, therefore, based on historical stream ow data for Mediterranean Region of Turkey, we generate scenarios for stream ow by using a modified k-nearest neighbor (k-NN) algorithm. We solve our model for di erent levels of risk-aversion and compare the optimal solutions. For models with large number of scenarios, we propose a multi-cut scenario-wise decomposition algorithm as an exact solution method. In order to evaluate the performance of our algorithm, we compare it with CPLEX. We conclude that, for a large number of scenarios, our algorithm is more efficient than CPLEX.