Conflation of computational fluid dynamics and building thermal simulation
The present work is a contribution towards the integration of building simulation tools in order to better represent the complexity of the real world. It attempts to overcome certain shortfalls of contemporary simulation applications with respect to indoor air flows. As a result, the evaluation of building energy consumption and indoor air quality is expected to be improved. Advanced fluid flow models (as employed within Building Thermal Simulation - BTS - and Computational Fluid Dynamics - CFD) with different degrees of detail were investigated and their modelling deficiencies identified. The CFD technique which defines the fluid flow on a micro scale was integrated into BTS in which fluid flow is described in a larger scale. The resulting combined approach strengthens the modelling potential of each methodology by overcoming their specific deficiencies. BTS's inability to predict air flow property gradients within a single space was surmounted and the difficult of estimating CFD boundary conditions are now supplied by BTS. The conflation approach is expected to be employed where gradients of indoor air flow properties can be considered crucial to the evaluation of thermal comfort and energy consumption. The BTS environment, ESP-r, was elected to perform the current work and a new CFD program, dfs, was specifically developed for the analysis of three-dimensional, turbulent, transient air flow. Finally, the two approaches were integrated. The integration work focuses on the CFD boundary conditions where the interactions of BTS and CFD take place; these occur at the inside zone surfaces and at the zone openings. Three conflation approaches were devised addressing different degrees of complexity and sophistication. The first one, involving the two types of zone boundaries, corresponds to a simple approach where the BTS and CFD systems exchange information without any direct interaction. The second approach consists of three other schemes to handle the thermal coupling at the internal zone surfaces. The third approach comprises coupling between the nodal network approach as employed by the BTS environment, and the continuity and momentum equations in the CFD technique. A validation methodology consisting of analytical validation, intermodel comparison and empirical validation is described and applied. The technique is shown to be adequate for modelling indoor air flows when compared to existing models. Three situations, covering the different types of air flows encountered within buildings are discussed to demonstrate the combined method's applicability when compared with the nodal network approach. Finally, general conclusions are presented and some possible future work is identified showing that the developed methodology is very promising.