Modelling the Torque with Artificial Neural Networks on a Tunnel Boring Machine

Abstract

The performance of earth pressure balanced tunnel boring machines (EPB-TBM) is dependent of a variety of parameters. Moreover, these parameters interact in a rather challenging way, making it difficult to adequately model their behaviour. Artificial neural networks have the aptitude to model complex problems and have been used in a variety of construction engineering problems. They can learn from existing data and then be used to predict the results, which makes them adequate for modelling problems where large amount of data is generated. In this work, a multilayer feedforward artificial neural network has been used to predict the torque at the cutter head of an EPB-TBM. A time series neural network has been used, where torque was predicted as a function of the measured torque and the volume of the injected foam on previous time steps. Results indicate that feedforward artificial neural network can be used to predict the torque at the cutter head in a EPB-TBM