Aging life prediction system of polymer outdoors constructed by ANN. 1. Lifetime prediction for polycarbonate

摘要

By the application of artificial neural network (ANN), the concept of constructing polymers' aging life prediction system has been proposed for the first time in the present work. Such a system can be designed via integrating several independent ANN modules playing different roles: the S module for calculating environmental stress level, the F module for outputting the feature of aging life distribution, the S_α module for computing the accelerated environmental stress level, the α_T module for studying the correspondence between life distribution of certain accelerated and natural aging tests, and the structure module for converting polymer structures into structure parameter matrix. Here polycarbonate, as a widely used engineering thermoplastic with complex aging behavior and various aging mechanisms, has been researched to constructing its aging life model with the help of the systematic ANN method above, which well reflects huge potential of ANN's application in the field of polymers' aging life prediction for practical application.