Abstract
Many calculation procedures for determining the thermal transport through building walls are based on thermostat directed graph networks. A recent study showed that such design calculations were only accurate to the 50% level. Thermal transport processes can be represented in either a local constitutive convolution form or as a directed graph network. Both representations are based on response factors which can, in principle, be used to estimate dynamic thermal transmission of the building components. This paper shows that only the local constitutive equations yield, for all cases, consistent and accurate values for the physical properties of the materials under test. The directed graph network representations are unreliable because they are ill posed and are likely to lead to erroneous predictions for the performance of buildings. The constitutive representation allows the measurement of thermal conductivity and transmission of building components under real meteorological conditions. It is then used to measure the convective and radiative coefficients of a wall and the representation can be extended to the nonlinear thermal transport case. (7 figures, 2 tables). (Author)
Irving, A D;
[1]
Dudek, S;
Warren, G;
[2]
Dewson, T
[3]
- Rutherford Appleton Lab., Chilton (United Kingdom)
- Newcastle upon Tyne Univ. (United Kingdom). Dept. of Building Science
- Bristol Univ. (United Kingdom). Dept. of Mathematics
Citation Formats
Irving, A D, Dudek, S, Warren, G, and Dewson, T.
On the local constitutive and directed graph network representations of response factors.
United Kingdom: N. p.,
1993.
Web.
Irving, A D, Dudek, S, Warren, G, & Dewson, T.
On the local constitutive and directed graph network representations of response factors.
United Kingdom.
Irving, A D, Dudek, S, Warren, G, and Dewson, T.
1993.
"On the local constitutive and directed graph network representations of response factors."
United Kingdom.
@misc{etde_10124771,
title = {On the local constitutive and directed graph network representations of response factors}
author = {Irving, A D, Dudek, S, Warren, G, and Dewson, T}
abstractNote = {Many calculation procedures for determining the thermal transport through building walls are based on thermostat directed graph networks. A recent study showed that such design calculations were only accurate to the 50% level. Thermal transport processes can be represented in either a local constitutive convolution form or as a directed graph network. Both representations are based on response factors which can, in principle, be used to estimate dynamic thermal transmission of the building components. This paper shows that only the local constitutive equations yield, for all cases, consistent and accurate values for the physical properties of the materials under test. The directed graph network representations are unreliable because they are ill posed and are likely to lead to erroneous predictions for the performance of buildings. The constitutive representation allows the measurement of thermal conductivity and transmission of building components under real meteorological conditions. It is then used to measure the convective and radiative coefficients of a wall and the representation can be extended to the nonlinear thermal transport case. (7 figures, 2 tables). (Author)}
place = {United Kingdom}
year = {1993}
month = {Oct}
}
title = {On the local constitutive and directed graph network representations of response factors}
author = {Irving, A D, Dudek, S, Warren, G, and Dewson, T}
abstractNote = {Many calculation procedures for determining the thermal transport through building walls are based on thermostat directed graph networks. A recent study showed that such design calculations were only accurate to the 50% level. Thermal transport processes can be represented in either a local constitutive convolution form or as a directed graph network. Both representations are based on response factors which can, in principle, be used to estimate dynamic thermal transmission of the building components. This paper shows that only the local constitutive equations yield, for all cases, consistent and accurate values for the physical properties of the materials under test. The directed graph network representations are unreliable because they are ill posed and are likely to lead to erroneous predictions for the performance of buildings. The constitutive representation allows the measurement of thermal conductivity and transmission of building components under real meteorological conditions. It is then used to measure the convective and radiative coefficients of a wall and the representation can be extended to the nonlinear thermal transport case. (7 figures, 2 tables). (Author)}
place = {United Kingdom}
year = {1993}
month = {Oct}
}