The Application of Artificial Neural Network (ANN) Technique to Formulation Design of Flame Retardant Polymers-II. Halogen-Free Flame Retardant polymeric Composites

摘要

The application of artificial neural network (ANN) to formula- tion design of the natural color, halogen-free flame retardant materials was conducted by means of flame retardant expert system (FRE 2.0). This work involves an ANN model with three objects. The optimized outputs, such as LOI, TS and EL were shown to be in good accordance with experiments. An intuitive Graphical analysis led to dealing with the correlation between performance and Components in formulations, where normally subjected to serious nonlinearity And/or multiobjects. It should be emphasized that together with the Three-layered BP (back-propagation) network-based modeling, one can do what Can hardly be done before with conventional techniques.