Neural networks used as a soft-modelling technique for quantitative description of the relation between physical structure and mechanical properties of poly(ethylene terephthalate) yarns

摘要: Abstract The use of artificial neural networks (ANNs) to develop a model the relation between physical structure and mechanical properties poly(ethylene terephthalate) yarns is investigated. was studied on set 295 drawn yarn samples, produced under an exceptionally large variety applied process conditions. ANN, consisting three layers neurons, trained using property measurements these as input output patterns, respectively. All ten were fitted within variance that acceptable for quantitative use. This paper demonstrates ANNs may be technique in solving real-world problem. resulted improved understanding complicated properties.