Algorithms for multi-dimensional heat transfer in buildings

A first order correction to uni-directional heat transfer is proposed, so that multi-dimensional heat transfer effects can be accounted for with only a moderate increase in storage and CPU timerequirements. The model has been implemented into the ESP building energy simulation program and is shown to be able to predict the order of magnitude of changes due to corner effects and therml bridges.These effects are shown to be non-negligible even in full scale buildings, especially if one isinterested in an accurate prediction of internal surface temperatures. As expected, the effects are particularly significant in scaled test cells, for which use of the algorithm results in an improved agreement with some experimental results.