Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Mechanical and Aerospace Engineering


Jianshun Zhang

Second Advisor

John Grunewald

Subject Categories

Mechanical Engineering


In this research, a systematic approach was introduced to establish a high quality and comprehensive material database. The method was applied to a great number of materials in different material categories. The database provides a solid base for hygrothermal simulation and further research.

A method was developed to derive the generic material from the material cluster comprising specific materials with similar characteristics. A novel approach was developed by using the generic material to extrapolate less incomplete material data set to full data set that is suitable for the hygrothermal simulation. The approach extends the material database, and hence enhances the usability of existing hygrothermal simulation tools.

Moisture storage characteristics (i.e., the moisture retention function) are one of the most difficult aspects to measure in developing a high quality database. In this study, a method was developed to simplify the procedure for moisture storage measurement with the aid of statistical analyses. For the building brick and plaster/mortar categories, results show that properly selected three measurements in the overhygroscopic range and one measurement in the hygroscopic range were sufficient to get the knowledge of moisture storage characteristics.

A probabilistic approach based on the Monte Carlo method was developed and incorporated into a current hygrothermal simulation tool, to assess hygrothermal performance of building enclosure assembly against different performance criteria. The uncertainties from different sources, including material properties, boundary coefficients, indoor conditions, dimensions of the material layers, and orientation of the construction, were accounted for. The rank correlations of basic material parameters in different material categories were obtained and incorporated in the Latin hypercube sampling.

The probabilistic approach was then applied to assess the durability, thermal efficiency, and mold growth risk of a retrofitted wall assembly. The most influential input variables against the specific performance criterion were identified by sensitivity analysis.


Open Access