Using Machine Learning Method to Discover Hygrothermal Transfer Patterns from the Outside of the Wall to Interior Bamboo and Wood Composite Sheathing

To identify hygrothermal transfer patterns of exterior walls is a crucial issue in the design, assessment, and construction buildings. Temperature relative humidity, as sensor monitoring data, were collected from outside wall to interior bamboo wood composite sheathing over year Huangshan Mountain District, Anhui Province, China. Combining machine learning method reservoir computing (RC) with agglomerative hierarchical clustering (AHC), novel framework was built for better extraction characteristics on time series data. The experimental results confirmed hypothesis that change temperature humidity (RHT12) dominated (RHT11). delay between two adjacent peaks 1 2 h, while 4 h sheathing. There no significant difference peak April July. 50 120 min. However, 100 240 min April, whereas it 20 impact formed relatively linear relationship outdoor time. characterized effectively by delays. discovery using will facilitate development energy-efficient durable materials structures.