Response Surface Method to Calculate Energy Savings Associated with Thermal Comfort Improvement in Buildings

Abstract

In developed countries, a large part of the building stock in 2050 will consist of currently existing buildings. Consequently, in order to achieve the objectives in terms of energy efficiency in the building sector we must consider not only new infrastructures but also the old ones. A reduction in energy consumption for climate control of between 50 and 90% can be achieved by rehabilitation and the implementation of different energy efficiency measures. Currently, these measures to reduce energy consumption and associated CO2 emissions can be modelled using computer tools. However, high precision and detail of thermal behaviour models through simulations can mean a great computational cost for companies, which results in a blockage of servers and workers. In this paper, the Response Surface Methodology (RSM) is presented as an innovative methodology for the simplification of models for calculation of the energy savings associated with thermal comfort improvement in buildings. A single-family house model, located in three different climates, is presented as a case study in order to validate the proposed methodology. Different scenarios were simulated, addressing heating and cooling temperature set points and external wall insulation represented by the transmittance (U-value). Results obtained from energy simulation using Design Builder were contrasted against those estimated from the simplified model extracted from the RSM analysis. The results revealed a deviation lower than 3% when comparing both methods. Therefore, the simplified mathematical prediction models are demonstrated to be suitable for the study of the energy performance of buildings, saving computational time, costs and associated human resources.

Applying the response surface methodology to predict the energy retrofit performance of the TABULA residential building stock

“…Both floor and orientation factors should be included in future studies. The importance of occupants' activity regarding energy consumption in residential buildings may be an important factor [48,49] that was also neglected in this and previous studies using RSM in both residential and non-residential buildings [15][16][17][18][19][20][21]. It is therefore necessary to include the factors related to occupant activity for a more accurate analysis.…”

“…In this paper, four design factors are selected that affect the building energy consumption, based on data extrapolated from the national TABULA project [9]. Compared to the previous studies utilizing RSM [15][16][17][18][19][20][21] for both residential and non-residential buildings, the inclusion of the overall heating and cooling system efficiency as a design factor is a novelty. It was shown that for both climate regions in Bosnia and Herzegovina (North and South) the cooling system efficiency had the strongest influence on the energy consumption for cooling when compared to the impact of SHGC, heat transfer coefficient of external walls, and roofs, while the heating systems efficiency had the second highest impact on the energy consumption for heating among the four design parameters selected.…”

“…The RSM method offers an opportunity for building designers to optimize the design response by using a sequence of designed experiments (DOE) that will determine the relationship between input building parameters and the building design response. This approach has been successfully applied to model building consumption for improved energy efficiency in several studies, including EE retrofit optimization of schools [15], university buildings [16,21], office buildings [17], residential dwellings [18,20], and apartment buildings [19]. A summary of the studies utilizing RSM to predict building energy consumption is shown in Table 1.…”

“…For example, the RSM simulations of energy consumption in education buildings [15,16,21] and offices [17] identified insulation thickness of the internal [16,21] and external walls [15,21], the heat transmission coefficient of the roofs [15], solar heat gain coefficient (SHGC) [15][16][17]21] and heat transfer coefficient of the external windows [15][16][17]22], roof heat transfer coefficient [16], internal and external shading coefficients [17,21] and window to wall ratio [15] as the key parameters affecting building energy efficiency. Similarly, for residential buildings the RSM simulations [18][19][20] pointed out heating and cooling system set-points [18,20], insulation thickness [18], SHGC [19], air infiltration rate [19], and insulation heat transfer coefficient [18][19][20] as the main contributors to energy savings.…”

“…The selection of DOE depends on the choice number of design points, i.e., building parameters and model running time [22]. The RSM is also usually applied in combination with traditional experimental designs for calibrating linear models: fractional factorial design (FFD) [15,19], central composite design (CCD) [20], and Box-Behnken design (BBD) [16,18,21], and D-optimal [17] to address non-linear models. The BBD is less expensive because it needs fewer runs compared to the other non-linear counterpart CCD, but the BBD design may contain regions of lower prediction quality due to a lower number of design points [22].…”