Artificial neural networks applied to polymer composites: a review

Inspired by the biological nervous system, an artificial neural network (ANN) approach is a fascinating mathematical tool, which can be used to simulate a wide variety of complex scientific and engineering problems. A powerful ANN function is determined largely by the interconnections between artificial neurons, similar to those occurring in their natural counterparts of biological systems. Also in polymer composites, a certain amount of experimental results is required to train a well-designed neural network. After the network has learned to solve the material problems, new data from the similar domain can then be predicted without performing too many, long experiments. The objective of using ANNs is also to apply this tool for systematic parameter studies in the optimum design of composite materials for specific applications. In the present review, various principles of the neural network approach for predicting certain properties of polymer composite materials are discussed. These include fatigue life, wear performance, response under combined loading situations, and dynamic mechanical properties. Additionally, the ANN approach has been applied to composite processing optimizations. The goal of this review is to promote more consideration of using ANNs in the field of polymer composite property prediction and design.