Calibration of building energy simulation models for energy-efficient retrofitting: A residential case study in Samsun-Havza


Creative Commons License

Tozlu S., Coşkun A., Arslan Selçuk S., Çakıcı F. Z.

International journal of energy studies (Online), cilt.10, sa.2, ss.595-617, 2025 (Hakemli Dergi)

Özet

Considering today's technological developments, it can be said that academic interest in intelligent systems that provide data about human living spaces has increased significantly. In the energy sector, the growing availability of data from smart evaluation systems and advanced devices, combined with progress in energy modeling software, has notably enhanced the effectiveness of energy modeling and efficiency improvement efforts. Calibration of Building Energy Simulation (BES) models is crucial for ensuring the accuracy required for implementing and evaluating energy efficiency strategies. Organizations such as ASHRAE 14-2014, IPMVP and FEMP are developing model validation methods in this context. This study addresses methodological challenges and reduces uncertainties encountered during the calibration processes of BES models. The primary objective of the research is to contribute to optimizing energy efficiency strategies. Integrating systematic calibration approaches and uncertainty assessment methods is anticipated to enable more accurate energy performance analyses. Methodologically, the study presents an approach to resolve errors in validation measurements within calibration processes. On the empirical side, the applicability of the systematic calibration methodology was successfully tested using forty days of hourly recorded indoor temperature data and indoor temperature data obtained from the EnergyPlus program via DesignBuilder; and validated with N(MBE) and CV(RMSE) uncertainty indices.As a result of the analysis, it was determined that the total final energy consumption (heating, DHW, electricity) of the building in question was 128.31 kWh/m², and approximately 72% of this was heating energy. Calibration results indicated that N(MBE) was 1.68% and CV(RMSE) was 13.86%, both within the thresholds set by ASHRAE 14, FEMP and IPMVP. This result shows that in terms of applicability, the calibrated model can be a practical tool that can be successfully used in energy efficient retrofit proposal development and implementation research.