11:18am - 11:21am
A review on coupled building physics analyses
TU Vienna, Germany
With the help of building physics analyses, an improvement in comfort, energy consumption and economic efficiency of buildings or quarters can be achieved. Although there is a variety of existing calculation methods and tools for different analyses in planning, operation or product development, most of the tools address a specific problem, which due to its complexity is considered in a limited computational domain or is simplified to reduce the calculation time.
An isolated consideration of problems is often not adequate to generate sufficiently precise results or to investigate several interdependent problems, which is why calculation methods or tools must be coupled in an interdisciplinary manner. But such a coupling includes several problems like incompatible or proprietary data models, the high effort for the creation of the complex calculation models, the calculation performance or the exchange of data. For a broad and economic application of building physics analyses it is important to be able to carry out detailed analyses easily and quickly. In order to apply optimisation algorithms or to develop predictive operating strategies by means of machine learning, for example, automated model generation and high-performance calculation is necessary.
With this background an overview of calculation methods and tools for coupling analyses with the focus on energy-, thermal- and fluid mechanical problems as a common application is given. Furthermore, requirements regarding the necessary information for the simulations, the data models and the coupling are identified. Possibilities of automated generation of simulation models, data exchange and the performance of existing multi physics simulation models are analysed and limiting factors are discussed.
11:21am - 11:24am
Selecting durable building envelope systems with machine learning assisted hygrothermal simulations database
Oak Ridge National Laboratory, United States of America
Hygrothermal simulations provide insight into the energy performance and moisture durability of building envelope components under dynamic conditions. However, the inputs required for hygrothermal simulations are extensive, and carrying out simulations and analyses requires expert knowledge. An expert system, the Building Science Advisor (BSA), has been developed to assist in predicting the performance and selecting the energy-efficient and durable building envelope systems in different climates. The BSA consists of decision rules based on expert opinions and thousands of parametric simulation results for selected wall systems. The BSA user could select from many different materials and parameters (such as ventilation in the cavity) for the layers in the wall system. The number of potential wall systems results in millions, too many to simulate all of them. This paper presents how machine learning can help predict durability data, such as mold growth, while minimizing the number of simulations needed to run.
The paper shows how parametric hygrothermal simulations are combined with advanced machine learning methods to compute results for constructions that were not part of the parametric analysis. The simulation results are used for training and validation of machine learning tools for predicting wall durability. We will test different classifications and regression methods for their applicability, model accuracy, and bias/variance trade-off. As a result, we will identify the most appropriate machine learning methods and build models for performance predictions.
The expert system, the BSA, consisting of expert opinions in the form of a rules database and simulation results supplemented with the machine learning models, guides practitioners to help achieve more robust conclusions and design better building envelopes. We conclude with an introduction on how the information supports guidance for envelope design via an easy to use web-based tool that does not require the end-user to run hygrothermal simulations.
11:24am - 11:27am
A New Wall Function for Indoor Airflow with Buoyancy Effect
1School of Environmental Science and Engineering, Tianjin University, Tianjin 300072, China; 2School of Mechanical Engineering, Purdue University, West Lafayette, IN 47905, USA; 3Department of Mechanics, School of Mechanical Engineering, Tianjin University, Tianjin 300072, China
Convective heat transfer on interior surfaces of building envelope is of significance in predicting building energy consumption. Buoyancy effect is one of the factors that affects it. Reynolds averaged Navier-Stokes (RANS) models combined with standard wall function is often used to simulate convective heat transfer coefficient of an interior wall. However, since the effect of buoyancy is not considered in the standard wall function, the convective heat transfer is often underestimated, which in turn will affect the prediction of indoor air velocity and temperature distributions. Some researchers modified standard wall function by adjusting the wall Prandtl number. But, the ad hoc adjustment does not reflect actual physical mechanism. This investigation developed a new wall function to take the effect of buoyancy on heat transfer into account. The proposed new wall function was developed by adding a buoyancy source term to the Navier-Stokes equation near the wall. The new wall function varied with buoyancy and y+. The constant term in the new wall function changes with the ratio of buoyancy and inertia forces and is linear to the logarithm of Richardson number. Then four typical indoor flows were simulated to test the accuracy of the new wall function. In the two-dimensional natural convection case, the predicted velocity, temperature and local Nu number were 30.3%, 12.8% and 19.7%, respectively, better than those with the standard wall function. In the two-dimensional mixed convection case, the temperature prediction was 59.7% more accurate than that with the standard wall function on a horizontal wall and 30.1% on a vertical wall. In the three-dimensional mixed convection, the performance of the new wall function in predicting air temperature was 51.5% better than the standard wall function. The results verified that the new wall function performed better in flow with buoyancy than the standard wall function.
11:27am - 11:30am
SELECTING EXTREME WEATHER FILE TO ASSESS OVERHEATING IN RESIDENTIAL BUILDING
1Univ. Bretagne Sud, UMR CNRS 6027, IRDL, Lorient, France; 2Tribu, Lyon, France; 3Cerema, Equipe-Projet BPE, Nantes, France
Climate change is great challenge for current and newly built buildings. Therefore, it should be taken into account when building performance simulations are run in the design phase. Nowadays, TMY weather file can be easily generated with software like Meteonorm following the IPCC scenarios. Nevertheless, since these data are extrapolated stochastically from monthly mean values, they do not show a real pattern and do not include extreme events like heat waves. In order to get more representative data, we propose in this work a methodology to select measured files from a large database in light of heat waves and climate change.
This methodology is applied to the city of Lyon, for which 26 years of weather data are available. Three measured weather files projected for the time slices 2020, 2050 and 2080 are selected. These files are used in building performance simulation of residential building with low or high thermal inertia. Summer overheating is analysed through two different comfort indicators: adaptative comfort chart (according to EN 16798) and Givoni chart. Results indicates that summer overheating risk is obviously increased with future weather files. When compared to usual TMY files, this risk is also enhanced by using weather file including extreme events like heat waves. Last, we note that discomfort is mainly encountered during this extreme events.
11:30am - 11:33am
A multiple Linear Regression Model to predict indoor temperature trend in historic buildings for book conservation: the case study of “Sala del Dottorato” in Palazzo Murena, Italy.
1Department of Engineering – University of Perugia, Via G. Duranti 93 (Perugia, Italia); 2CIRIAF - Interuniversity Research Centre, University of Perugia, Via G. Duranti 63 (Perugia, Italia)
The indoor climate of historic buildings is governed by the aim to preserve them, their interiors and to ensure human comfort. For the preservation of cultural and artistic heritage, relative humidity and temperature are very important parameters, including their amplitudes and changes rate in time. For this reason, a monitoring campaign is fundamental to evaluate these aspects. In the present study, a two-year experimental campaign of indoor temperature and relative humidity was carried out inside of “Sala del Dottorato”, located on the first floor of Palazzo Murena, which is the headquarters of the University of Perugia since 1810. In this room, a great number of rare and ancient books are preserved. The paper deals with the study and the evaluation of the correlation between outdoor and indoor microclimate conditions in the room, in order to ensure the proper conservation of the book heritage; it is aimed at understanding how the two parameters follow outdoor variations and how the hygrothermal inertia of the building, with massive walls, can mitigate these variations. The analysis is carried out for temperature, which is the most critical issue, because of the high values in summer and daily variations during the cold season (with discontinuous heating), which can accelerate the rate of the degradation process of the books. Thanks to a continuous monitoring system for indoor and outdoor thermohygrometric parameters, it will be possible to notify, in real-time, when the microclimatic indoor conditions are out of the range for conservation. This monitoring makes it possible to develop a Multiple Linear Regression model to predict and analyse the indoor temperature trend which is closely related to outdoor temperature and relative humidity. This model could permit to estimate a forecast of this parameter and could make it possible to predict in advance critical conditions for correct conservation.