Title :
Building a hybrid pultrusion Derakane 440/40 process model using neural networks and process data
Author :
Wright, David T. ; Williams, David J.
Author_Institution :
Dept. of Manuf. Eng., Loughborough Univ. of Technol., UK
Abstract :
Previous extensive laboratory pultrusion trials and materials analysis yielded a rich data set. This paper details various representations of this data set in alternate artificial neural networks (ANN). The authors examine the effectiveness of the various ANN process models. A number of existing mathematical pultrusion models are discussed. Further, hybrid process models were developed which gave novel insight into various relationships between inter-process variables
Keywords :
glass fibre reinforced composites; knowledge representation; manufacturing processes; modelling; neural nets; process control; hybrid pultrusion Derakane 440/40 process model; materials analysis; neural networks; process data; Artificial neural networks; Data engineering; Knowledge representation; Materials testing; Mathematical model; Mechanical sensors; Neural networks; Object oriented modeling; Pulp manufacturing; Resins;
Conference_Titel :
Systems, Man and Cybernetics, 1995. Intelligent Systems for the 21st Century., IEEE International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-2559-1
DOI :
10.1109/ICSMC.1995.538371
