Figure 2 - uploaded by Symeon Christodoulou
Content may be subject to copyright.
Source publication
Buildings account for around 40% of EU energy requirements and have been the focus of several initia- tives, the most notable of which has been the recast of the European Energy Performance of Buildings Di- rective (EPBD, Directive 2010/31/EU). The Directive requires that all member states ensure that new build- ings, as well as large existing buil...
Context in source publication
Context 1
... obtained simulation results are used to rank the influence of each input variable on a building’s total annual energy consumption and on the Building Energy Rating (BER). Figure 2 shows the probability distribution function (PDF) of the Building Energy 2 Rating [kWh/m /yr] for a single-storey dwelling, derived from various possible simulation combinations. The uncertainty distribution in BER for a single storey dwelling nearly follows a log-normal distribution with geometric mean approximately 2 254kWh/m /yr. Similarly, for two-, and three-storey dwellings the BER distribution nearly follows log- normal distribution with geometric mean 2 2 237kWh/m /yr and 221kWh/m /yr respectively. 3.1 Scenario Set (A) Storey count According to the results for the three dwelling types examined (single-, two-, and three-storey dwellings) the contribution of the 20 most important factors to BER [kWh/m2/yr] are tabulated by ranking in Tables 2-4. The greatest impact in energy use is at- tributed to the floor area followed by the external conditions, the dwell ing’s envelope u -value (roof, window, walls, floors), space heating system, ventilation, windows area, walls area etc. 3.2 Scenarios set (B): Risk Factor Grouping In order to reduce the number of input parameters by an order of magnitude, so as to reduce the computational complexity of the energy simulations, the analysis is then narrowed down to 12 stochastic groups of similar type and/or effect. Then, sensitivity analysis within each group is used to rank the significance of the factors constitute each group. Re- 3.3 Scenario Set (C): Variable Climate Profiles According to the results, the climatic external conditions are shown to be one of the most sensitive factors, having a significant contribution to the BER. Though, deeper consideration is needed so as to account for the fact that DEAP is more representative of northern European ...
Citations
... A building's energy performance is the outcome of a multi-dimensional parametric system which requires specialized, in-depth consideration. The energy performance of a building results from a non-linear system with multi-interrelationships between a wide variety of influencing factors, stemming from the building's architectural, structural, functional, operational, and location characteristics (Christodoulou and Chari 2014). The complexity of this system makes the modelling and prediction of the energy performance of buildings a difficult task. ...
... It was concluded that 50% of the factors contribute in total approximately 98% of BER, while the remaining 50% factors with a total minimum contribution of 2% to BER could be ignored. With this information at hand, a ANN model with 34 inputs was built, eliminating the remaining 34 most insignificant factors, as per the sensitivity analysis presented in Chari (2014 andChristodoulou et al. (2014a). In essence, the contribution of each input factor and of its variability to the predicted output factor (the BER classification) is assessed and the factors are ranked in descending order with regard to this contribution. ...
The energy in buildings is influenced by numerous factors characterized by non-linear multi-interrelationships. Consequently, the prediction of the energy performance of a building, in the presence of these factors, becomes a complex task. The work presented in this paper utilizes risk and sensitivity analysis and applies artificial neural networks (ANNs) to predict the energy performance of buildings in terms of primary energy consumption and CO2 emissions represented in the Building Energy Rating (BER) scale. Training, validation, and testing of the utilized ANN was implemented using simulation data generated from a stochastic analysis on the ‘Dwellings Energy Assessment Procedure’ (DEAP) energy model. Four alternative ANN models for varying levels of detail and accuracy are devised for fast and efficient energy performance prediction. Two fine-detailed models, one with 68 energy-related input factors and one with 34 energy-related input factors, offer quick and multi-factored estimations of the energy performance of buildings with 80 and 85% accuracy, respectively. Two low-detailed models, one with 16 and one with 8 energy-related input factors, offer less computationally intensive yet sufficiently accurate predictions with 92 and 94% accuracy, respectively.
