A comparison and evaluation of available techniques for monitoring bored tunnelling excavated spoil volumes

Abstract

With plans to increase the network of Singapore's transit system, the push to go underground utilising bored tunnelling is inevitable; a scenario mirrored in many conurbations across the world. Being a city, tunnelling works often have to be carried out under or adjacent to major buildings, roads and existing tunnels. A disruption to any of these may have catastrophic consequences in terms of economic damage and impact to the public. Close monitoring of ground and structural movements above the tunnelling works is critical. However, due to the nature of the construction, it is difficult to predict major ground loss, and as such, this only appears after the outcome of any such event. The first line of defense against any ground collapse during bored tunnelling lies in the scrutiny of the excavated spoil volume from the tunnel face. On Contracts 824 and 825 of the Circle Line Project (CCLP), different techniques are being used to calculate the amount of excavated material per ring. These include the traditional method of counting the number of spoil disposal skips and relating them to the spoil volume, the use of a conveyor belt weighing device and the monitoring of excavated material using a laser scanner on the conveyor belt. These more sophisticated techniques are increasingly becoming standard options on TBMs, however there remains skepticism about their actual usefulness as a monitoring tool. Through analysis of the data produced, the merits of belt weighing and the laser scanner are evaluated for two different ground conditions, Marine Clay and Old Alluvium and two different machines, a Hitachi Zosen and a Herrenknecht. The results of the comparisons are promising and suggest that automated spoil monitoring can be used as a means of avoiding large volume losses during tunnelling, and collapse at the surface, through identifying over excavation early in the excavation process and taking appropriate measures to reduce further over excavation. The correlation between methods used show that nearly all the recorded automated spoil volumes fall within 10% of the muck skip counting method with the majority falling within 5%. A closely observed trial over 10 rings provided even better correlation between the laser scanner and muck skip counting with the average differences between the laser scanner and skip counting being 4.5 m3. The study notes a number of anomalies that indicated shortfalls in the systems and suggested refinements are proposed to make the monitoring systems more effective. The study concludes that automated monitoring of spoil will lead to safer tunnelling, but notes that the research is limited to two ground conditions. As the tunnelling has not completed further study is ongoing to refine the systems and provide more definite proof of the benefits of automated spoil monitoring. (A). Reprinted with permission from Elsevier. For the covering abstract see ITRD E124500.