Predicting the flammability of epoxy resins from their structure and small-scale test results using an artificial neural network model
Ákos Pomázi, Andrea Toldy
Abstract
Abstract Developing optimal flame retardant polymer compositions that meet all aspects of a given application is energy and cost-intensive. To reduce the number of measurements, we developed an artificial neural network-based system to predict the flammability of polymers from small-scale test data and structural properties. The system can predict ignition time, peak and total heat release, and mass residue after the burning of reference and flame retarded epoxy resins. Total heat release was predicted most accurately, followed by the peak heat release rate. We ranked the input parameters according to their impact on the output parameters using a sensitivity analysis. This ranking allowed us to establish a relationship between the input and output parameters taking into account their physical content.