Detection of faults in a high speed packaging machine using amultilayer perceptron (MLP)

摘要

In high speed manufacturing systems, continuous operation isdesirable, with minimal disruption for repairs and service. Anintelligent diagnostic monitoring system, designed to detect developingfaults before catastrophic failure, or prior to undesirable reduction inoutput quality, is a good means of achieving this. Artificial neuralnetworks have already been found to be of value in fault diagnosis ofmachinery. The aim here is to provide a system capable of detecting anumber of faults, in order that maintenance can be scheduled in advanceof sudden failure, and to reduce the necessity to replace parts atintervals based on mean time between failures. Instead, parts will needto be replaced only when necessary. Analysis of control information inthe form of position error data from two servomotors is described