Evaluating the impact of input variables on thermal load estimation in office spaces using validated models

The design and optimisation of energy efficiency measures in buildings are often based on simulation models, which must be validated with experimental measurements to ensure a reliable representation of the building’s actual behaviour. One of the main challenges in this process is minimizing uncertainties associated with input variables, as any deviation can lead to significant discrepancies from the design case. The main objective of this study is to quantify the impact of input variable variations on the heating and cooling loads of an office model. For this purpose, a validated dynamic simulation methodology has been developed and implemented in several phases: monitoring of the office space and boundary conditions; dynamic modelling of the office; validation of the simulation model using experimental data; and sensitivity study through parametric and influence analysis. As a result of this methodology, the most influential variables affecting three annual series of thermal demand have been identified. In this case study, it is concluded that accurate definition of seasonal set-point temperatures, climate files, and infiltration rates are crucial for obtaining accurate thermal demand estimates. Conversely, variations in occupancy schedules and ground temperature profiles have a lower impact, allowing for less accuracy in their input definition without compromising the reliability of the results.