Analysis of machine degradation using a neural network based pattern discrimination model

Abstract

Abstract Machines degrade as a result of aging and wear, decreasing performance reliability and increasing the potential for faults and failures. The effect of machine faults and failures on factory productivity is an important concern for manufacturing industries. Economic effects relating to machine availability and reliability, as well as corrective (reactive) maintenance costs, have prompted facilities and factories to improve maintenance techniques and operations to monitor machine degradation and detect faults. This paper presents an innovative methodology to change maintenance practice from breakdown reaction to breakdown prevention, thereby reducing maintenance costs and improving productivity. To analyze machine behavior quantitatively, a pattern discrimination model (PDM) based on a cerebellar model articulation controller (CMAC) neural network was developed. A PUMA 560 robot was used to study the feasibility of the developed technique. Experimental results have shown that the developed technique can analyze machine degradation quantitatively. This methodology can help operators set up machines for a given criterion, determine whether the machine is running correctly, and predict problems before they occur. As a result, maintenance hours can be used more effectively and productively.