Plastics are nowadays one of the most used materials in several industries. From domestic tools up to the automotive industry, there is an enormous number of possible applications. The pressures imposed by the market have led to the development of new strategies capable of answering the required demands. Neural networks have revealed high potential in a wide range of situations and have been successfully applied in fault detection and diagnosis systems. In this paper we intend to clarify, in part, the different diagnostic methodologies and, on the other hand, we suggest a neural network approach for monitoring the plastic injection moulding process. Future work will use the developed neural monitoring scheme for process fault diagnosis with the aim of industrial quality management.
[1]
Ali Alawi,et al.
Real Time Fault Diagnosis
,
1997
.
[2]
David Kazmer,et al.
Incorporation of Phenomenological Models in a Hybrid Neural Network for Quality Control of Injection Molding
,
1999
.
[3]
Timo Sorsa,et al.
Neural networks in process fault diagnosis
,
1991,
IEEE Trans. Syst. Man Cybern..
[4]
Stuart Bennett,et al.
Dynamic System Fault Diagnosis Based on Neural Network Modelling
,
1997
.
[5]
Jeffrey L. Elman,et al.
Finding Structure in Time
,
1990,
Cogn. Sci..
[6]
Heikki N. Koivo,et al.
Artificial neural networks in fault diagnosis and control
,
1994
.