Analysis of a non-calorimetric method for assessment of in-situ thermal transmittance and solar factor of glazed systems

Abstract

The performance of glazing systems is usually assessed through the thermal transmittance and the solar factor, two metrics characterised either through calorimetric laboratory tests or calculations. In this paper, the analysis of the performance of a non-calorimetric method for obtaining the in-situ thermal transmittance and solar factor of glazing systems is presented. This method, developed as a trade-off between accurate (and expensive) laboratory tests (which characterise the systems under standardised, "averaged" conditions), and easy and less expensive tests on systems installed in real buildings ( under real operative conditions), has been previously adopted for the characterisation of different glazed systems, but never presented and discussed in full detail.The method, suitable for full-scale glazing systems installed in buildings or in test cells, is based on the acquisition of temperature, heat flux, and solar irradiance values. Experimental data is then processed through simple equations and linear regressions to determine the thermal transmittance and the solar factor under real boundary conditions. In this paper, a detailed description of the method, the experimental test rig, and the related expected accuracy is reported. The method is then applied to a case study (a conventional double glazed unit) to give an example of the proposed procedure and to validate it.The results of the case study show the capability of the assessed in-situ thermal transmittance and solar factor to replicate the thermophysical behaviour of the glazing system within a satisfactory degree of accuracy. An in-depth discussion on the observed outcomes from the case study deepens the understanding of the method's performance and the results' significance. *Unmarked Revised Manuscript Click here to view linked References

Empirical validation of co-simulation models for adaptive building envelopes

“…These two parameters are usually measured under steady-state conditions either by laboratory tests or by software tools integrating databases of glass panes. However, due to the simplified approach in evaluating these parameters, they may not correspond to the actual thermal and solar performance (Goia & Serra, 2018). For example, there may be a discrepancy between the boundary conditions registered in situ and the standardised conditions used during laboratory characterisation.…”

“…Therefore, the simplified approach can lead to significant differences between the calculated and in-situ energy performance of the glazing system. Since there are no calorimetric conditions in non-controlled, non-calorimetric test facilities to assess the performance of the building envelopes, this framework used the in-situ characterisation based on the work by Goia and Serra (2018). The framework measures the U-and g-values through empirical measurements under non-calorimetric conditions.…”

“…Further details of the experimental setup of the in-situ characterisation are listed in Table 10. Following Goia & Serra (2018), the U-value U was obtained by linear regression using the ordinary least squares method and calculated as: Equation 3Equation 4…”

Calibration of DSF model for real-time control

“…By comparing the measured data with those simulated by the model, an anomalous distribution of the surface temperatures of the single skin was found: the temperature layers facing the cavity was too high while those facing outward has a lower temperature. This could be related to the assumption made for the development of the model in which the doubleglazing unit have been model as a single layer with fixed cavity resistance values independent from the boundary conditions [6]. Therefore, to overcome this simplification, the convective exchange coefficient inside the DGU gap has been increased as follow ( In step 5 instead, the air gap temperature simulated was higher than measured, so the contribution of convection inside the cavity has been increased as follow (Table 1, step 5):…”

“…Indeed the numerical model has very short computational times, allowing the model to be integrated in an embedded single board controller within the DSF itself, and the models outputs to be seamlessly integrated in a building level supervisory system. The simplified model has been validated and tested with the measured data acquired during the experimental campaign carried out on a DSF prototype installed in the South-exposed façade of the outdoor test facility at Energy Department of Politecnico di Torino [6].…”