Stochastic simulation of building envelope performance: methodology and implementation

It is well recognized that climate changes will have an impact on building performance in different aspects, such as the performance with respect to whole building energy consumption, to indoor thermal comfort as well as to the hygrothermal performance of building envelope. Climate-resilient building design has become crucial for constructing new buildings or rehabilitating existing buildings when adapting to the changing climate. NRC Construction Research Centre has carried out a Climate-Resilient Built Environment (CRBE) project: “Decision Support Tools for Building Envelope Performance Assessment” (2021-2026), intended as means to develop decision support tools, including guides and models for the design of resilient new buildings, and the rehabilitation of existing buildings to ensure that existing and future climate loads and extreme weather events are considered. To achieve these goals, the NBC Part 9 walls will be investigated through hygrothermal simulation and stochastic simulation to take the uncertainties of climate change into consideration. In the CRBCPI project carried in the past five years (2016-2021), an extensive set of hygrothermal simulations were carried out to evaluate the mould growth performance of different types of wood-frame building envelopes under historical and future climatic conditions across a number of major cities in Canada. The hygrothermal simulations were performed by following the approach described in Guideline on Design for Durability of Building Envelopes1, and the simulation results were presented in the report Results from Hygrothermal Simulations and on Durability and Resilience of Wall Assemblies to Climate Change2. To consider the uncertainties of the input parameters, such as the material properties, rain water penetration moisture loads and cladding ventilation rates, stochastic simulations need to be performed in the CRBE project (2021-2026). The stochastic simulation procedure should be generalized, and the generalized procedure can be applied to the CRBE project related to hygrothermal performance analysis and moisture damage risk assessment of building envelopes. The information provided in this report describes the process of rationalisation of the procedure used for hygrothermal performance analysis and probabilistic risk assessment of moisture related damage of building envelopes using stochastic simulation. It is intended for use by expert practitioners who have knowledge of hygrothermal simulation, and require hygrothermal performance results to assess climate resilient building envelope design, considering the uncertainties in input parameters. A complete stochastic simulation procedure requires the following steps: • The quantification of stochastic variables • Applying proper sampling techniques to generate stochastic models • Implementing stochastic simulations • Data visualization of stochastic results Generally, the stochastic simulation needs to be implemented in a third-party programming environment, in which a hygrothermal simulation engine can be called, and the stochastic models with randomly assigned variables can be repeatedly launched. In this report, the hygrothermal simulations were executed using Delphin 5.9.6, a commercial hygrothermal simulation program. Python 3.4 software was used for generating probability distributions of stochastic variables, implementing advanced sampling techniques to generate stochastic models, launching stochastic simulations and visualizing the stochastic results. A typical 2x6-inch wood-frame wall was used to demonstrate the stochastic simulation procedure; the stochastic simulations were performed under the historical and future climatic conditions of Ottawa. Although the stochastic simulation framework was developed for the purpose of hygrothermal performance analysis and moisture damage risk assessment of building envelopes, this procedure can also be adapted for summer-time overheating risk assessment.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.