Combining multi-objective evolutionary algorithms and descriptive analytical modelling in energy scenario design

Environmental and security concerns urge energy planners to design more sustainable energy systems, reducing fossil fuel consumptions in favour of renewable solutions. The proposed scenarios typically rely on a mixing of different energy sources, thereby mitigating the availability and intermittency problems typically related to renewable technologies. Optimizing this combination is of crucial importance to cope with economic, technical, and environmental issues, which typically give rise to multiple contradictory objectives. To this purpose, this article presents a generalized framework coupling EnergyPLAN – a descriptive analytical model for medium/large-scale energy systems – with a multi-objective evolutionary algorithm – a type of optimizer widely used in the context of complex problems. By using this framework, it is possible to automatically identify a set of Pareto-optimal configurations with respect to different competing objectives. As an example, the method is applied to the case of Aalborg municipality, Denmark, by choosing cost and carbon emission minimization as contrasting goals. Results are compared with a manually identified scenario, taken from previous literature. The automatic approach, while confirming that the available manual solution is very close to optimality, yields an entire set of additional optimal solutions, showing its effectiveness in the simultaneous analysis of a wide range of combinations.