Non-destructive neural identification of the moisture content of saline ceramic bricks

• The identification of the moisture content of saline ceramic bricks was presented.
• The bricks were placed in aqueous nitrate, chloride and sulphate environments.
• Neural identification of the moisture content of saline ceramic bricks is possible.
• Broyden-Fletcher-Goldfarb-Shanno (BFGs) learning algorithm was applied.
• Presented approach can be especially useful in historical buildings.

The use of the artificial neural network (ANN) for the non-destructive identification of the moisture content of saline ceramic bricks is presented in the following article. A database was built based on conducted laboratory tests of solid ceramic bricks using dielectric, resistive and microwave methods. The database consists of parameters determined with the above-mentioned methods and also parameters describing the type and concentration of salts in the bricks. As a result of carried out research and analysis it was proved that neural identification of the moisture content of saline ceramic bricks is possible.