![]() There was push back against energy efficiency improvements because of concerns about durability.”Ĭould energy efficiency and durability be achieved at the same time? In 1998, ORNL began its partnership with the Fraunhofer Institute for Building Physics (IBP) to develop the capability to use computer simulation to evaluate the moisture durability of building envelope assemblies common in the US. “A general concern arose that as building envelopes became more airtight and energy efficient, they were also losing their drying potential, with moisture-related failures being the unintended consequence. These suits centered on moisturerelated failures of energy efficient building envelopes,” explained Andre Desjarlais, Group Leader, Building Envelope Research Group at ORNL. “We became involved in moisture, and its effects on efficiency, shortly after a rash of lawsuits were filed in the late 1990’s. These current studies are grounded in a history of building envelope moisture durability research at ORNL, sponsored by the US Department of Energy (DOE) Building Technologies Office and industry partners. To better understand energy efficiency features and their interaction with moisture and ultimately durability, Oak Ridge National Laboratory (ORNL) has established a new experimental facility and initiated several studies, including an examination of the relationship between air tightness and moisture durability, hygrothermal performance of below grade construction as a function of soil type, and moisture levels associated with cool roofs. But what happens when solar-reflective exterior surfaces, and airtight and high-R envelope assemblies are thrown into the mix? This question becomes more significant when anticipating expanded retrofit activity to improve existing building envelopes. When moisture accumulation exceeds the ability of the assembly materials to store the moisture without significantly degrading performance or long-term service life, moisture problems result. Miscalculating the impact of environmental factors like rain, solar radiation, temperature, humidity, and indoor sources of moisture can cause significant damage to many types of building envelope components and materials, and also can lead to unhealthy indoor living environments.Īs described by Joseph Lstiburek, B.A.Sc., M.Eng., Ph.D., P.Eng., principal of Building Science Corporation, moisture accumulates in the building envelope when the rate of moisture entry into an assembly exceeds the rate of moisture removal. Finally, we introduce 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.Moisture, and its accompanying outriders – things like mold, corrosion, freeze damage, and decay – present powerful threats to the durability and long-term performance of a building envelope. Models developed with both methods showed adequate prediction performance (root mean square error of 0.195 and 0.209, respectively). We tested Artificial Neural Network (ANN) and Gradient Boosted Decision Trees (GBDT) for their applicability and model accuracy. The simulation results are used for training and validation of machine learning tools for predicting wall durability. ![]() We present how machine learning can help predict durability data, such as mold growth, while minimizing the number of simulations needed to run. The number of potential wall systems results in millions, too many to simulate all of them. The BSA consists of decision rules based on expert opinions and thousands of parametric simulation results for selected wall systems. An expert system, the Building Science Advisor (BSA), has been developed to predict the performance and select the energy-efficient and durable building envelope systems for different climates. The inputs required for hygrothermal simulations are extensive, and carrying out simulations and analyses requires expert knowledge. Hygrothermal simulations provide insight into the energy performance and moisture durability of building envelope components under dynamic conditions.
0 Comments
Leave a Reply. |