Building Performance Based on Measured Data

S. Andersson
Department of applied physics and electronics, Umeå, Sweden

J-U Sjögren
Department of applied physics and electronics, Umeå, Sweden

R. Östin
NCC Ltd, Stockholm, Sweden

T. Olofsson
Department of applied physics and electronics, Umeå, Sweden

Ladda ner artikelhttp://dx.doi.org/10.3384/ecp11057899

Ingår i: World Renewable Energy Congress - Sweden; 8-13 May; 2011; Linköping; Sweden

Linköping Electronic Conference Proceedings 57:20, s. 899-906

Visa mer +

Publicerad: 2011-11-03

ISBN: 978-91-7393-070-3

ISSN: 1650-3686 (tryckt), 1650-3740 (online)


With increasing liability for builders; the need for evaluation methods that focuses on the building’s performance and thus excludes the impact from residents’ behavior increases. This is not only of interest for new buildings but also when retrofitting existing buildings in order to reduce energy end-use.

The investigation in this paper is based on extensive measurements on two fairly representative type of buildings; a single family building in Ekerö; Stockholm built 2000 and two apartment buildings in Umeå (1964) in order to extract key energy performance parameters such as the building’s heat loss coefficient; heat transfer via the ground and heat gained from the sun and used electricity.

With access to pre-processed daily data from a 2-month periods; located close to the winter solstice; a robust estimate of the heat loss coefficient was obtained based on a regression analysis. For the single family building the variation was within 1% and for the two heavier apartment buildings an average variation of 2%; with a maximum of 4%; between different analyzed periods close to the winter solstice.

The gained heating from the used electricity in terms of a gain factor could not be unambiguously extracted and therefore could only a range for the heat transfer via ground be estimated. The estimated range for the transfer via ground for the two apartment buildings were in very good agreement with those calculated according to EN ISO 13 370 and corresponded to almost 10% of the heating demand at the design temperature. For the single family building with an insulated slab and parts of the walls below ground level; the calculations gave slightly higher transfer than what was obtained from the regression analysis. For the estimated gained solar radiation no comparison has been possible to make; but the estimated gain exhibited an expected correlation with the global solar radiation data that was available for the two apartment buildings.


Regression analysis; Heat loss coefficient; Heat transfer via ground; Gained heat


[1] ASHRAE Handbook Fundamentals 2005; Atlanta; USA

[2] F.W. Yu and K.T. Chan; Energy signatures for assessing the energy performance of chillers; Energy and Buildings; 2005 Vol. 37 (7); pp. 739-746. doi: 10.1016/j.enbuild.2004.10.004.

[3] C. Ghiaus; Experimental estimation of building energy performance by robust regression; Energy and Buildings; 2006 Vol. 38(6); pp. 582-587 doi: 10.1016/j.enbuild.2005.08.014.

[4] A Rabl and R Rialhe; Energy signature models for commercial buildings: test with measured data and interpretation; Energy and buildings; 1992 Vol. 10; pp.143-154.

[5] S. Andersson; J-U. Sjögren; T Olofsson and R Östin; Prestanda- och beteendeuppföljning av byggnaders energianvändning. 2010; Internal report. Dept. of Applied Physics and Electronics; Umeå University

[6] S Wold; M Sjöström and L Erikson; PLS-regression: A basic tool of chemometrics; Chemometrics and intelligent laboratory systems; 2001 Vol. 58; pp 109-130 doi: 10.1016/S0169-7439(01)00155-1.

[7] T Olofsson; S Andersson and J-U Sjögren; Building energy parameter investigations based on multivariate analysis; Energy and Buildings; 2009 Vol. 41(1); pp. 71-80 doi: 10.1016/j.enbuild.2008.07.012.

[8] M Lundin; S Andersson; R Östin A.R. Further validation of a method aimed to estimate building performance parameters; Energy and Buildings; 2005 Vol.37 (8); pp. 867-871 doi: 10.1016/j.enbuild.2004.11.005.

[9] P.A. Gonza´lez; J.M. Zamarren˜o; Prediction of hourly energy consumption in buildings based on a feedback artificial neural network; Energy and Buildings 2005 Vol. 37(6) pp. 595–601. doi: 10.1016/j.enbuild.2004.09.006.

Citeringar i Crossref