Sage Journals: Discover world-class research

Abstract

Analytical tests are based on physically simple cases for which the exact solution is known. A set of recommended analytical tests is described for the validation of conduction calculations in dynamic thermal models of buildings. They are one product of a collaborative research project on model validation. The principal advance is the explicit recognition of the need to design tests differently for (a) verification and debugging, in which not more than one source of error should contribute to the difference observed between model predictions and exact solution, and (b) tests to estimate the likely accuracy in use, in which ultimately all sources of error should be active in the way that they would be in a normal building simulation. The chief obstacles to creating such clear-cut tests are the discretisation errors arising from input sampling and reconstruction, and other deliberate approximations used in the modelling. It is shown that it is possible to construct tests that have all the necessary properties, without having recourse to tests of isolated sub-models. It is strongly recommended that analytical tests be made available as a software facility accessible to users, and be applied: (i) to programs by their authors, both when new, and as part of routine software maintenance, (ii) to established programs by independent validators.

Get full access to this article

View all access options for this article.

References

Allen E. , Bloomfield D. , Bowman N. , Lomas K. , Allen J. , Whittle J. and Irving A.

Analytical and empirical validation of dynamic thermal building models

Proc. Int. Building Energy Simulation Conf, Seattle pp 274-280 (August 1985)

Allen E. and Bloomfield D.

Improving confidence in thermal calculation procedures

Proc. CLIMA 2000, Copenhagen Vol. 2 pp 351-356 (August 1985)

Jones L.

The analyst as a factor in the prediction of energy consumption

Proc. 2nd Int. CIB Symp. Energy Conservation in the Built Environment , Danish Building Research Institute, Copenhagen pp 313-321 (1979)

Bloomfield D.P.

The influence of the user on the results obtained from thermal simulation programs

Proc. 5th Int. Symp. Use of Computers for Environmental Engineering Related to Buildings pp 161-180 Bath (July 1986)

Judkoff R. , Wortman D. , O'Doherty R. and Burch J. A methodology for validating building energy analysis simulations. Draft Report SERI/ TR-254-1508 (Golden, Colorado : National Renewable Energy Laboratory) (1983)

Bland B.H. Analytical tests of conduction An investigation into analytical and empirical validation techniques for dynamic thennal models of buildings BRE/SERC collaborative research project final report Volume II ( Garston: Building Research Establishment) (December 1988)

Bland B.H. and Bloomfield D.P.

Validation of conduction algorithms in dynamic thermal models

Proc. CIB 5th International Symp. Use of Computers for Environmental Engineering Related to Buildings, Bath pp 22-35 (1986)

Bland B.H.

Experience with analytical tests for the validation of dynamic thermal models

Proc. USER1 Conf, Ostend , Belgium pp 124-129 (1988)

Maloney D. A verification of the use of heat conduction transfer functions as used in the program BLAST (Urbana, IL: BLAST Support Office, Department of Mechanical and Industrial Engineering, University of Illinois) (1985)

10.

Maloney D.

Notes on conduction transfer functions

BLAST News 4(2) ( Urbana,-IL : BLAST Support Office, Department of Mechanical and Industrial Engineering, University of Illinois) (February 1986)

11.

Carslaw H.S. and Jaeger J.C. Conduction of heat in solids 2nd edn ( Oxford: UP ) (1959)

12.

Keeble E.

Modelling building performance

Building Serv. CIBSE J. 10(1) p 59 (January 1988)

13.

Bloomfield D.P. and Fisk D.J.

Room thermal response and control stability

Building Serv. Eng. Res. Technol. 2(2) 88-92 (1980)

Conduction in dynamic thermal models: Analytical tests for validation

Abstract

Get full access to this article

References