Open AccessTransfer learning for the design of a digital twins-based automatic relay production line
Author(s) -
Rui Yang,
Qinglong Mo,
Zijuan Huang,
Yu Zhang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1682/1/012028
Subject(s) - production line , field (mathematics) , intersection (aeronautics) , industrial production , production (economics) , industrial revolution , industrial engineering , computer science , industry 4.0 , manufacturing engineering , digital transformation , transfer of learning , artificial intelligence , engineering , transport engineering , mechanical engineering , mathematics , economics , macroeconomics , world wide web , law , keynesian economics , pure mathematics , political science , embedded system
At present, the world is experiencing a new round of scientific and technological revolution and industrial transformation. It has entered a historic intersection with the transformation and upgrading of the manufacturing industry, bringing new opportunities to the industry. Digital twins are considered as an effective way to realize the interaction and integration of the manufacturing physical world and the information world, attracting great attention from the relevant academic and business sectors at home and abroad. Specifically, they use virtual models and simulation technology to design industrial production lines and predict the future operation of equipment, which contributes to the efficient operation of intelligent production processes. Nonetheless, the key learning problem of accurate modeling should be solved, so as to adapt to the random and dynamic changes in industrial equipment design. Transfer learning, an innovative learning paradigm in machine learning, is developed to solve challenging learning problems with only a few or no labelled samples in the target field.