Investment Evaluation under Multiple Uncertainty: Optimal Sizing and Configuration of an Integrated Energy Production System by Renewable Sources

In this work, we present a methodology to evaluate long-term investments in integrated energy production systems by renewable sources (also Hybrid Renewable Energy Systems). Unavoidably, the investment decision occurs under multiple uncertainty, which affects the environmental, technical, and economic-financial variables: availability of renewable sources (climatic data), end-user behaviour, technology efficiencies and costs, discount rate for the project, inflation of energy prices. Thus, we adopt a simulation-based optimization procedure in order to provide the decision maker with useful information for the investment choice and the analysis of the output reliability. All the involved subsystems (photovoltaics, small wind turbines, solar thermal, heat pump, electric and thermal storages, end-user system and demands, traditional back-up systems) and their mutual interactions are dynamically simulated. A multiobjective optimization finds the Pareto frontier that maximizes net present value (NPV) and minimizes CO2 emissions. The optimal design parameters of the energy system are determined by a controlled elitist genetic algorithm. Once the optimal design configurations are identified, results are furtherly investigated by a sensitivity analysis, in order to quantify uncertainty associated with the input variables and its propagation to the final results. The process would be a first step in order to guide investors to a more in-depth analysis and characterization of the critical variables and possibly towards a more robust design choice among the Pareto set. A case study of a small hybrid production system for an autonomous (off-grid) building is finally presented with a demonstrative aim.