In this paper, a trained fuzzy neural network (FNN) control system for gravure printing is presented. Firstly, the gravure printing process and its problems that need to be solved are introduced. Then, a FNN controller is proposed, which integrates fuzzy IF-THEN rules into a neural network structure based on the characteristics of gravure printing. The gradient descent with momentum learning method is utilised to train the FNN. The learning ability and effectiveness of the proposed FNN control system is demonstrated with experimental data. Finally, conclusions are drawn and further research summarised.
|Publication status||Published - 2004|
|Event||Control 2004 - Bath, UK United Kingdom|
Duration: 6 Sept 2004 → 9 Sept 2004
|Country/Territory||UK United Kingdom|
|Period||6/09/04 → 9/09/04|