Predicting the flammability of epoxy resins from their structure and small-scale test results using an artificial neural network model

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