Vol.23, No.1, 2023, pp. 23–29
UDC:

FAULTS DIAGNOSTICS OF CEMENT DRAFT FAN USING ARTIFICIAL NEURAL NETWORK (ANN)

Noureddine Menasri*, Noureddine Aimeur

University of M’sila, Laboratoire de Matériaux et Mécanique des Structures (LMMS), ALGERIA

*email: noureddine.menasri@univ-msila.dz

 

Abstract

Fans are the key components of any cement manufacturing process. Without them, the process does not work very well, or it would not be effective. They can be subjected to a large number of damages (wear, unbalance, etc.) occurring during the operation and whose causes are multiple. One problem of great importance in industrial monitoring is performing fault detection and determining the faulty component, or at least the suspect area in the schema of the system. To address this issue, the diagnostics of most defects that may affect the fans is investigated in this work using spectral analysis of vibration which allows the construction of signatures defects. These signatures are dedicated to automating the diagnostics by artificial neural network.

Keywords: draft fans, bearings, diagnostics, spectral analysis, artificial neural network (ANN)

full article (568 kB)