Neural network model for resilient modulus of emulsified asphalt mixtures

Abstract

This paper explores the potential use of neural networks (NNs) in the field of emulsified asphalt mixtures. A neural network model is developed for predicting, with sufficient approximation, relationship between the factors affecting resilient modulus (inputs: curing time, cement addition level, and residual asphalt content) and the resilient modulus (output) of emulsified asphalt mixture. A backpropagation neural network of three layers is employed. First resilient modulus data are obtained by conducting laboratory resilient modulus tests on emulsified asphalt samples, and then the results are used to train the neural network. The effectiveness of different neural network configurations is investigated. Effect of parameters such as curing time, cement addition level and residual asphalt content that influence the resilient modulus is also explored. Results indicate that NN predicts the resilient modulus with high accuracy. It is also demonstrated that NN is an excellent method that can reduce the time consumed and can be used as an important tool in evaluating the factors affecting resilient modulus of emulsified asphalt mixture at the design stage.