Robustness of multi-objective optimization of building refurbishment to suboptimal weather data

The interest in energy refurbishment has been growing in the last few years, since noticeable energy savings can be achieved through energy saving measures (ESMs) applied to the existing building stock. In this respect, one of the best opportunities to promote the energy renovation of the existing buildings is to define cost-effective solutions through multi-objective optimization techniques, which can rely on genetic algorithms (GA), and building energy simulation (BES). Although this topic has been widely discussed in the literature, little is yet known about the robustness of the optimal solutions obtained from GA multi-objective optimizations to the variation of the quality of the used weather data. This information could also be relevant to understand if the optimal solution obtained with historic weather data could be undermined by future climate changes. Aiming to provide objective confidence levels of the multi-objective optimization, in this work we investigate the extent to which the weather data used for BES can affect the optimal solutions. With this purpose, several multi-objective optimizations have been carried out with reference years obtained with different lengths of historic series for the location of Trento, North Italy. The results show changes to both Pareto’s fronts and optimal retrofit solution.