Reducing the energy consumption of buildings in the public sector is an important component in our efforts towards reaching our sustainability goals. In this context, a decisive prerequisite for administrations and policy makers is a tool for estimating the effectiveness of measures to reduce energy consumption. Estimating the impact of planned investments in building technology at scale, however, remains challenging, mainly for two reasons. For one, accurate physical modeling requires detailed building data, which can be difficult to obtain. Second, adapting established building models to novel measures aiming at energy consumption reduction is difficult. Hence, modeling building consumption patterns after retrofitting is a non-trivial task, and more research is needed to improve modeling techniques as well as to assess their effectiveness across a wide range of application scenarios. Modeling tools need to be generic enough to enable modeling of a variety of building types, they should ideally require as few input features as possible and they should allow for a high degree of automation in the selection and calibration of building modeling tools. Here, we propose a novel machine learning approach that does not require detailed building data and can automatically adapt to retrofitting measures. We evaluate our method on a data set of 113 public buildings in 4 building categories in Berlin, Germany. The data set contains energy consumption data in the initial state and after implementation of a weather-predictive heating control system. Despite being fully automated and requiring only minimal information about the building, our model can reliably predict the energy consumption of large public buildings better than established methods. All code and data are publicly released.

With the increasing need to tackle climate change, energy efficiency and reduced CO₂ emissions are proving to be one of society’s greatest challenges. Special consideration should be given to heating systems as they are prone to inefficiency due to non-optimal controller configurations and the shortage of experts or qualified technicians to optimize the operating behavior. Especially for residential heating systems, more often than not, the target metric is the achievement of specific heating and hot water temperatures by manual adjustments with limited sensor information and with little regard to efficiency. This presents potential for computer-aided optimization based on artificial intelligence techniques. In this paper, we presented a Decision Integration System that is interfaced with a data acquisition infrastructure and allows for the analysis of measured heating system data, the generation of recommended measures for efficiency improvement, and the simulative validation of recommended controller parameter changes. We presented different parts of the Decision Integration System, the interfaced data acquisition infrastructure, as well as the non-invasive sensor appliances used. We analyzed the measured data of real heating systems and evaluated our approach by generating the recommended measures based on rules created by heating system experts, which were then partially applied to the physical heating systems and partially evaluated in simulation. Finally, we compared long-term energy consumption data against the latest monitoring period after implementing the measures. Our results showed an average reduction in energy consumption of 24.52% across all considered buildings, corresponding to an approximate reduction of 8.12 tons of CO₂ emissions.