Detecting changing geologic conditions with tunnel boring machines by using passive vibration measurements

Abstract

This research explores the detection of changing geological conditions with a Tunnel Boring Machine (TBM) by using passive vibration measurements. In this investigation, three iterations of an acquisition system were deployed at two different field sites. In these deployments, the TBMs were instrumented with accelerometers, and the response of the transducers was paired with the machine's operating data to seek correlations associated with changing geological conditions. To better understand the system's response, both contributions from specific machine components and the dynamic response of the TBM's structure were considered. Additionally, an advanced filtering method was tested to more completely explore the complicated relationship between vibration and geology. Results from this study yield a number of interesting findings. First, that the passive vibration measurements can be used to detect events in successive revolutions of the TBM's cutting head. Second, the structural modes, of the TBM, are not well characterized in the response spectra, because the system's inputs (Torque, Thrust, Advance Rate, etc.) are non-stationary in nature. As a result of the non-stationary inputs, traditional techniques such as operational modal analysis (OMA) were of limited use. Finally, an advanced filtering technique, called principal motion analysis, shows promise as a method for detecting and analyzing whole machine motion and overcoming the limitations of OMA.