Prediction of moisture content of brick walls of historical buildings in southern Fujian based on neural network
Year: 2025
Type: conference-paper
Abstract: historical buildings of Gulangyu Island in Xiamen is not only a typical sample of the historical buildings in southern Fujian, but also a significant part of the World Cultural Heritage of Gulangyu Island, showing rich history and culture. Based on the field investigation data, this paper uses artificial neural network model to numerically analyze and predict the mass moisture content of brick walls of 8 historical red brick buildings in Gulangyu Island. The experimental results show that the training, testing and verification of the data set based on the prediction method proposed in this paper all obtain a high R2 value. It shows that the prediction model constructed in this paper has good generalization performance and high prediction accuracy, and is suitable for the prediction of the moisture content of brick walls of historical buildings in southern Fujian Province.
Authors: Yiting Lin, Niwat Angkawisittpan, Xiaoli Fu, Guifen Lyu
Institutions: Xiamen Institute of Technology, Mahasarakham University
Language: English
Subfield: Building and Construction
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4414179708