Constructal design associated with genetic algorithm (GA) is employed to optimize the geometry of an isothermal Y-shaped cavity embedded into a solid conducting wall. The structure has four degrees of freedom (DOF). The main purpose is to minimize the maximum excess of temperature between the solid and the cavity by means of GA and exhaustive search (simulating every geometry combinations). Results showed that GA was well succeeded to find the best shapes which minimize the maximal excess of temperature with a number of simulations strongly lower than that required with exhaustive search, allowing the optimization of cavity under new constraint conditions.

The work documents the design of Earth-Air Heat Exchangers based not only on sensible heat transfer, but also on latent heat exchanges. We compare the impact of the climate of Brazil and south of France on the relevance of such systems. The duct length is determined in order to obtain maximum underground heat exchanges. A time dependent model combined to actual weather data is developed to show when an underground heat exchanger becomes a good option in a tropical climate. The three-dimensional version of the model accounts for heat transfer in the soil and for heat and moisture transfer along the underground pipe. The comparison with a 1D model allows to propose a straightforward approach to assess the cooling/heating potential of different climatic regions.

No abstract

No abstract