사

Prediction of Yarn Properties Using Natural Computation.

  • 출판일1999.08
  • 저자
  • 서지사항
  • 등록일 2016.11.02
  • 조회수 257
Researchers used computation techniques adapted from nature to derive data relevant to the effect of process variations and structure on yarn properties. Artificial neural networks (ANNs) determined the relationships between process settings, physical structure, and thermal and mechanical properties of polyethylene terephthalate yarns. Researchers developed the SYGNA system consisting of a genetic algorithm in which an ANN is embedded as a knowledge base for optimizing process settings and yarn structures. Akzo Nobel used the genetic algorithm, trained and validated ANNs, and used background texts and data to create a user friendly software system called BESSY, which makes data concerning the relationship between process conditions, physical structures, and end uses widely accessible. 11 refs.
이전글 다음글 목록