<p>Template energy calculation models that have been produced by the Building Energy End-use Study (BEES) team are used to quickly and reliably model commercial buildings and calculate their energy performance. The template models contain standardised equipment, lighting, and occupancy loads; cooling and heating requirements are calculated using an ideal loads air system. Using seven buildings, Cory et al. 2011a have demonstrated that the template models have the potential to closely match the monthly energy performance of detailed (individually purpose built) models and the real buildings. Three of these models were within the ±5% acceptable tolerance to be considered calibrated. The four template models that were not within the acceptable tolerance have been identified to have complex Heating, Ventilation, and Air Conditioning (HVAC) systems that the ideal loads air systems could not replicate. Because HVAC systems consume one of the largest proportions of energy in commercial buildings, this has a significant impact on the reliability of the template models. To address this issue, a set of detailed HVAC systems were needed to replace the ideal loads air systems. Due to HVAC system parameters not being collected by the BEES team and the lack of published modelling input parameters available, it is unknown what values are reasonable to use in the models. This study used a Delphi survey to collect real building information of the commonly installed HVAC systems in New Zealand commercial buildings. The survey formed a consensus between HVAC engineers that determined what the most commonly installed systems are and their associated performance values. The outcome of the survey was a documented set of system types and modelling input parameters that are representative of New Zealand HVAC systems. The responses of the survey were used to produce a set of HVAC system templates that replace the ideal loads air systems. The HVAC template models updated the software default parameter values with values that are representative of commonly installed systems in New Zealand. The importance of the updated input values was illustrated through a comparison of the calculated monthly energy consumption. The resulting difference in energy consumption using the updated parameter values is typically <5% monthly; at worst it is 75% for Variable Air Volume (VAV) system in the Wellington climate during June.</p>
Model Predictive Control Optimization via Genetic Algorithm Using a Detailed Building Energy Model