Further development of the change-point model – Differentiating thermal power characteristics for a residential district in a cold climate

The building and service sector accounts for nearly 40% of total energy use in Sweden. existing, historic stock large part this comprises an important the national pursuit to increase efficiency. One main problem decreasing existing is lack data describing thermal performance characteristics. This paper presents a novel development change-point model predicting buildings using selected time periods based on time-dependent variations climate user behavior. predicted power characteristics include specific heat losses (Qtotal), hot water circulation (HWC) tap (HTW), balance temperature. A residential district with 73 Linköping, Sweden, has been used as study object. developed shown be effective robust performance. average R2 was 0.70 predictions losses. sensitivity analyses conclude that steps months correspond highest value. variation width prediction temperature 0.9 °C interquartile range three-year comparison hourly heating supply data. Moreover, from property owner perspective, useful identifying deviating can easily get overview district.