Adaptive Recognition and Control of Shield Tunneling Machine in Soil Layers Containing Plastic Drainage Boards

Abstract

The underground plastic vertical drains (PVDs) are a significant problem for shield machines in tunneling construction. At present, the main method to deal with PVDs is to manually adjust the parameters of the shield machine. To ensure that a shield machine autonomously recognizes and adjusts the control in soil layers containing PVDs, this study constructs a shield machine advance and rotation state-space model utilizing Bayesian decision theory for the judgment of excavation conditions. A Bayesian model predictive control (Bayes-MPC) method for the shield machine is proposed, followed by a simulation analysis. Finally, a validation experiment is conducted based on a Singapore subway project. Compared with traditional methods, the method proposed in this paper has better performance in the simulation, and it also has demonstrated effectiveness and accuracy in experiments. The research outcomes can provide a reference for the adaptive assistance system of shield machines excavating underground obstacles.