Title: Prognosis of failures due to the abnormal temperature increase of the malt crusher, using ANFIS neuro-fuzzy approach: case study of the flour mill of the breweries of Cameroon

Authors: Timothee Kombe; Sandra Nzeneu

Addresses: Laboratory of Industrial Automatic, PhD Training Unit of the Sciences of the Engineer, University of Douala, P.O. Box 8698 Douala, Cameroon ' Laboratory of Industrial Automatic, PhD Training Unit of the Sciences of the Engineer, University of Douala, P.O. Box 8698 Douala, Cameroon

Abstract: In this article, we present a method for predicting malt grinder failures. The objective is to control the evolution of the temperature of the grinding chamber, in order to optimise the availability and reliability of the grinder. The methodological approach is based on the ANFIS neuro-fuzzy network, which offers, in a single tool, precision for nonlinear systems. The predicted temperature is classified according to a mode of operation of the equipment. The evaluation of the performance of the prediction and classification systems is characterised respectively by a learning error function of 0.1236 and a classification rate of 79.6%.

Keywords: artificial intelligence; hybrid neuro-fuzzy network; form recognition; failure; prognosis; Cameroon.

DOI: 10.1504/IJISE.2022.121047

International Journal of Industrial and Systems Engineering, 2022 Vol.40 No.2, pp.228 - 254

Received: 19 Aug 2019
Accepted: 14 Jan 2020

Published online: 23 Feb 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article