NEUR AL CONTROLLER FOR THE SELECTION OF RECYCL ED COMPONENTS IN POLYMER - GYPSY MORTARS

• Main Authors: Grzegorz KŁOSOWSKI, Tomasz KLEPKA, Agnieszka NOWACKA
• Format: Article
• Language: English
• Published: Polish Association for Knowledge Promotion 2018-06-01
• Series: Applied Computer Science
• Subjects: neural networks; gypsum-polymers; rubber regranulate
• Online Access: http://www.acs.pollub.pl/pdf/v14n2/4.pdf

This study presents research on the development of an intelligent controller that allows optimal selection of rubber granules, as an admixture recycling component for polymer-gypsy mortars. Based on the results of actual meas-urements, neural networks capable of predicting the setting time of gypsum mortar, as well as determining the bending and compressive strength coef-ficients were trained. A number of simulation experiments were carried out, thanks to which the characteristics of setting times and strength of mortars containing different compositions of recycling additives were determined. Thanks to the obtained results, it was possible to select the rubber admixtures optimally both in terms of the percentage share as well as in relation to the diameter of the granules.