Building energy models with morphological urban-scale parameters: A case study in Turin

With a growing awareness around the importance of the optimization of building efficiency, being able to make accurate predictions of building energy demand is an invaluable asset for practitioners and designers. For this reason, it is important to continually improve existing models as well as introduce new methods that can help reduce the so-called energy performance gap, which separates predicted from actual consumption values. This is particularly true for urban scale simulations, where even small scenes can be very complex and carry the necessity of finding a reasonable balance between precision and computational efforts. The scope of this work is to present two different models that make use of morphological urban-scale parameters to improve their performances, taking into account the interactions between buildings and their surroundings. In order to do this, two neighbourhoods in the city of Turin (IT) were taken as case studies. The buildings studied present similar characteristics but are inserted in a different urban context. Several urban parameters were extracted using a GIS tool and used as input, alongside the building-scale features, for two different models: i) a bottom-up engineering approach that evaluates the energy balance of residential buildings and introduces some variables at block-of-buildings scale, ii) a machine learning approach based on the bootstrap aggregating (bagging) algorithm, which takes the same parameters used by the previous model as inputs and makes an estimation of the hourly energy consumption of each building. The main results obtained confirm that the urban context strongly influences the energy performance of buildings located in high built-up areas, and that introducing simple morphological urban-scale parameters in the models to take these effects into account can improve their performance while having a very low impact on the computational efforts.