A non-destructive method of the evaluation of the moisture in saline brick walls using artificial neural networks
Source abstract
Abstract
Abstract The article presents a method of non-destructive evaluation of the moisture content in saline brick walls. This method is based on the use of artificial neural networks (ANNs) that are trained and tested on a database that was built for this purpose. The database was created based on laboratory tests of sample brick walls. The database contains over 1100 sets of results. Each set consists of two parameters that describe the dampness of the tested sample walls, which were determined using dielectric and microwave methods, and also three parameters that describe the concentration of salts in these walls. The ANN with back propagation error and the Broyden–Fletcher–Goldfarb–Shanno learning algorithm (BFGS) was used. It was shown that the proposed method of assessment allows reliable results to be obtained, which was confirmed by the high values of the linear correlation coefficient for learning, testing and experimental validation.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Geophysical Methods and Applications · Soil Moisture and Remote Sensing · Building materials and conservation
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References
Reference metadata was not supplied for this record.