Finite-difference methods, when applied to the differential equations modelling energy flows in buildings, give rise to a system of non-linear difference equations. A frequently used direct solution method involving linearisation is analysed and a related method is proposed. These and one other connected method are compared using a building-related test problem prepared for this purpose. The proposed numerical method is found to produce least error, about 30% less than the commonly used method, and it is recommended for inclusion in building energy simulation software. A fundamental method for estimating the pre-conditioning period of a building, arising from the work, is discussed.
Get full access to this article
View all access options for this article.
References
1.
Clarke J.A.Energy simulation in building design (Bristol: Adam Hilger) (1985)
2.
Lambert J.D.Numerical methods for ordinary differential systems: the initial value problem (Chichester: Wiley) (1991)
3.
Pinney A. and Parand F.The effect of computational parameters on the accuracy of results from detailed thermal simulation programsProceedings of Building Environmental Performance (University of Kent , Canterbury: BEPAC) pp 207-222 (Apr 1991 )
4.
Ozisik M.N.Heat conduction2nd edition ( New York: Wiley) (1993)
5.
Crowley M.E. and Hashmi M S JEvaluation of implicit numerical methods for building energy simulationProc. Inst. Mech. Eng., Part A, Journal of Power and Energy212(A5) 331-342 (1998)
6.
User's guide, Afathcad PLUS 6.0 (Cambridge, MA: Mathsoft Inc) (1995)
7.
Press W.H., Flannery B.P., Teukolsky S.A. and Vetterling W.T.Numerical recipes in C, the art of scientific computing2nd edition (Cambridge: Cambridge University Press) (1992)
