Classification of impact damage on a rubber-textile conveyor belt using Naïve-Bayes methodology

Abstract

The selection of the proper materials is an integral component of most technological processes, and the selection of durable conveyor belt materials is a principal example. The overall quality of a conveyor belt and its service life under impact loads are very important factors. Accordingly, the purpose of the present work is to classify the types of impact damage that occur in rubber-textile conveyor belts. Under laboratory conditions, different types of conveyor belts (unused, used, used with a subsequent top cover layer renovation) were tested. They were subjected to variables such as the drop height and the type of impacting material. The severity of the conveyor belt damage was categorised using four degrees. An examination of the effects of the selected factors on the degree of damage was carried out using probability theory. In particular, an evaluation of the experimental test data and predictive modelling was conducted using the Naïve Bayes Classifier.