Nowadays, people spend most of their time inside the buildings, thus special attention has to be paid to the formation of indoor climate. Indoor climate quality can be affected by many factors. One of them is a proper design of mechanical ventilation, which is nowadays widely used mainly in non-residential buildings, where it provides sufficient air exchange and exhaust of pollutants.

The design of mechanical ventilation in present days can be considered as a “prescription based design“– where prescriptions and regulations are used for its design in different types of buildings. The proper distribution elements are then chosen according to the technical specifications from manufacturers or again, according to the prescriptions. But is the prescription based design the optimal design? This question can be answered using “the performance based design” based on surrogate modelling and optimization. Many input variables (flow rate, usage of the room, size of the distribution elements, etc.) have to be taken into account in the design phase of the mechanical ventilation, which offers a vast amount of possible combinations. In this case, the customized surrogate model for design optimization can help to find the proper design of the ventilation system in the room level.

Objective of this report is to investigate possibility of using CFD simulations and surrogate models for design optimisation of mechanical ventilation diffusers. Numerical simulations of isothermal airflow in office room are used to predict mean age of air in given space. Surrogate models are built using data obtained from these simulations.

Final project report