A rational method to estimate the energy baseline of a building is indispensable in performing such as retro commissioning, on going commissioning and ESCO (Energy Service Company) projects. Usually a statistical regression method is used to estimate a baseline. But the limitation of this method is that it cannot evaluate the effect of changes in operational conditions such as HVAC system and occupancy schedule, room temperatures, amount of fresh air intake in HVAC systems, weather conditions, etc. by which energy use is affected. In these cases an estimation method based on simulation would be a ood option. However, few studies are found about verification of the method based on experiments using a real building. Therefore we carried out an experiment for changes of room air temperatures and the fresh air intake volume in a real building and verified the accuracy of the baseline estimation method developed in our previous study. The results show that the model can estimate the baseline change with acceptable accuracy.