ROBUSTNESS ASSESSMENT METHODOLOGY FOR THE EVALUATION OF BUILDING PERFORMANCE WITH A CLIMATE UNCERTAINTIES
Proceedings of BS2015: 14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.
ROBUSTNESS ASSESSMENT METHODOLOGY FOR THE EVALUATION OF BUILDING PERFORMANCE WITH A VIEW TO CLIMATE UNCERTAINTIES
Giorgia Chinazzo1 , Parag Rastogi1 , Marilyne Andersen1
1Ecole Polytechnique Fédérale de Lausanne (EPFL), Interdisciplinary Laboratory of Performance-Integrated Design (LIPID), Lausanne, Switzerland
This paper describes a new methodology to assess the robustness of building performance in the long term with a probabilistic approach. The aim is to include uncertainties related to climate change predictions as well as the intrinsic uncertainties in weather files describing them. A case study focussing on refurbishment strategies of a realistic building in Turin is presented to demonstrate the methodological steps. The main outcome is that it is advisable to have outcomes in terms of ranges of energy consumption instead of single output values to evaluate energy efficient design solutions in both present and future years.
The complex relationship that tightly binds climate conditions and buildings makes it necessary to use building simulation techniques coupled with weather data to calculate energy performance and make design decisions. Conventionally, building energy performance is evaluated with a deterministic approach by using a single input weather file referring only to historical weather conditions (characterized by a TMY file).
Hence, the choice of a particular design strategy is based on a single energy usage referring to current weather conditions. However, since buildings have a life span of 50 to 100 years, they must perform satisfactorily under both current and future climate (Ascione et al., 2014; Kaklauskas et al., 2005; Wilde et al., 2008), which according to the IPCC report is going to be warmer mostly due to man-made emissions of greenhouse gases (GHGs) (IPCC, 2007). For this reason, the assessment of different design strategies must take into account weather files referring to both present and projected climate conditions in future years. Climate change adaptation of buildings has been investigated in some studies, which have calculated the impact of climatic changes on energy performance (Camilleri et al., 2001; Frank, 2005; Gaterell and McEvoy, 2005; Guan, 2009; Zmeureanu and Renaud, 2008). However, all of this previous work is deterministic and uses just one input weather file (Tian and de Wilde, 2011). In other words, they underestimate the uncertainties related to climate change projections and the intrinsic uncertainties of weather files describing both present and future climate, due to different years of record, morphing method and weather variables recorded. Using a single weather file in building simulations, regardless of its source or generative algorithm, could lead to inaccurate energy consumption forecasts, and therefore wrong design decisions. Building on the work of Tian and de Wilde (2011) on sensitivity analysis in the prediction of the thermal performance of buildings under climate change, this study illustrates a new methodology for the evaluation of building robustness using probabilistic energy performance results.
The impact of using multiple input weather files and the methodological steps to interpret the results are explored. The methodology is demonstrated by means of a case study simulated with eighteen weather files coming from different sources, referring to many future years and IPCC scenarios (Solomon et al., 2007). The case study selected is an existing dwelling with twenty-two refurbishments in Turin, Italy. The retrofit solutions focus on the thermal properties of the envelope by varying Uvalue, solar heat gains, thermal mass and air tightness of the envelope. The methodology is divided into two steps: first the energy usage ranges of different refurbishments are calculated and represented by an index (RI), then the energy saving due to refurbishments in each year (in comparison with the non-refurbished building) are evaluated and compared using a second index (ESI). It is important to note that the proposed work-flow is built so as to be able to accommodate changing climate predictions and new findings from the IPCC, updating the results as more information becomes available or models improve. The methodology could also be used to modify or at least revisit building energy codes to better evaluate energy savings for new constructions or refurbishments.
The structure of this paper is as following. First, the intrinsic uncertainties of weather files are briefly presented by means of two preliminary studies. Then the methodology steps of the robustness assessment evaluation are explained, and the case study is described. Finally, the results of the simulations and the methodology are explored.
WEATHER FILE UNCERTAINTIES
This paper does not address climate change projection uncertainties, because they are mainly related to climate models and future scenarios (Nakicenovic and Swart, 2000), but it focus on weather file uncertainties. The latter can be illustrated by means of two preliminary studies, referring respectively to present and future years. In general, future weather files are associated with temporal uncertainty, and present weather files with spatial uncertainty. In previous work (Chinazzo et al., 2015), we illustrate the first preliminary study, which demonstrates that spatial uncertainties of present weather files are related to intrinsic variability that can neither be predicted nor avoided. To prove that, we simulate a building model with weather files referring to present climate conditions coming from two weather file sources in six different weather stations in the north of Italy. The main outcome of that study is that energy usage results are quite different even if the model is the same and it is situated in the same climatic area. The main differences can be observed between the results calculated with files coming from the two sources. These ones are the U.S. Department of Energy’s website (E+) and the METEONORM software (MN).
In the following, we describe the second preliminary study, which refers to future weather files and the associated temporal uncertainties. Future weather files are generated from the present ones by means of two different software. The first one is the software ‘CC WorldWeatherGen Climate change world weather file generator’, a Microsoft Excel based tool which generates climate change weather files for any location (Jentsch et al., 2013). It transforms ‘present-day’ .epw weather files into future .epw weather files by using a model from the IPCC 2013 report (HadCM3 A2 experiment ensemble) for three future time slices, the 2020’s, 2050’s and 2080’s. The second software used to generate climate change weather files is Meteonorm, for different scenarios (B1, A1B and A2), and for any year between 2010 and 2200 (Remund, 2014). In the second preliminary study, the two present weather files (from the two sources E+ and MN) refer to Milan. Figure 1 displays the four years on the x-axis and the energy usage in kWh/m2 on the y-axis for cooling. The energy usage for cooling has an increasing trend through the years for both sources and the different scenarios, due to the predicted warming of the earth. The worst projection is made by the E+ weather files, because the energy usage is always higher compared to the three MN data sets. In each scenario of the MN sets, the difference between the three results is higher the further the projection is in the future. One way to interpret this is to say that, the further a projection is in the future, the less precise the predictions of energy usage are. In general, the differences between the energy usage predicted by the two sources is due to different extrapolation algorithms and to different input data, which we demonstrated to vary even in the present. Due to the fact that the weather files refer to the future, we cannot assess which one is wrong and which one is correct. For this reason, all the weather files can be considered as probable future projections. The main conclusion of these two preliminary simulations is that two different weather files cannot be considered as ‘duplicates’ of the same point, even if they refer to the same climatic area. Instead, they can be counted as random inputs, or ‘replicates’ in a simulation of building performance where all other factors remain the same. In other words, our methodology is a sensitivity analysis (Lomas and Eppel, 1992; Saltelli et al., 2004) where the uncertainties are represented by the input weather files.
The objective of this paper is to describe a methodology to assess the robustness of building performance to uncertainties in weather file, which we illustrated in the two preliminary studies before. The methodology starts with the simulation of a building model with different weather files, coming from different sources and stations and representing many future years and scenarios. These weather files create a large ensemble of plausible future climates, where each member of the ensemble represents one equally probable guess about how the climate could be. In this way it is possible to analyse the behaviour of different design strategies under many plausible future climates and assess their robustness over climatic uncertainties. In general, the robustness is defined as the sensitivity of particular performance indicators of a building to errors in the design assumptions (Hoes et al., 2009). In our case the errors are represented by the weather files used as inputs, and a robust solution is insensitive to climate change uncertainties. The methodology can be divided into two main parts: the energy usage robustness evaluation and the energy saving evaluation. Each of them is characterised by a graphical part and by an index. In general, the methodology helps to compare various design strategies in terms of ranges of energy usage. This approach could ultimately help architects and engineers to make more informed energy efficient choice at an early design phase.
Energy usage evaluation
... READ the Article