A co-simulation framework for scale-adaptive coupling between heterogeneous computational codes

We present the scale-adaptive coupling between heterogeneous computational codes and experimental facilities for human-centred indoor thermal performance analysis. Currently, the analysis comprises human thermoregulation, convective and radiative heat transfer between a virtual manikin model and the environment as well as models for thermal sensation and comfort perception. The developed software framework presented in this paper enables the modular adaptation of existing and future simulation software and experimental devices for common integration into a uniform co-simulation environment. A middleware layer provides the runtime infrastructure which facilitates loose coupling of numerical sub-models on distributed computing platforms. It provides services to visualize intermediate results and to interact with the simulation during computation. Multiple scales are addressed by the simulation framework as the level of detail can be adapted, depending on the trade-off between accuracy and computational costs. Built-in numerical smoothing improves the quality and the stability of the simulation results. The architecture further supports the use of parallel computing resources like clusters or graphics hardware to increase the overall simulation performance.