Abstract
A rockburst is a loud and spontaneous bursting of rocks inside an excavated tunnel sidewall. Further, its unpredictable occurrence poses a serious threat to tunnel workers. Several studies on rockbursts have been conducted; however, a method for predicting rockbursts has not yet been established. To establish such a method, it is important that the micro seismic activities during a rockburst, such as acoustic emissions (AEs), are well understood. In this study, to better understand the relationship between AE behavior and rock failure, laboratory tests and simulation analyses were conducted. In addition, an AE measurement system was developed and tentatively applied to a tunnel site. The AE behavior with rock failure progress was estimated with the measurements obtained from a triaxial compression test. The test results proved that the AE behavior differed depending on the progress of rock failure. It was observed that just before rock failure, the number of AE occurrences and amplitude of the AE waveforms increased rapidly, and the frequency of the waveform decreased. These AE parameters associated with rock failure were then simulated using a distinct element method. The analysis results were in accordance with the AE behaviors observed through laboratory tests. By applying these modeled AE parameters to the tunnel excavation simulation, a prediction method for rock failure could be established. This measurement system analyzes AE in real time and sends an alarm when the threshold value of each preset AE parameter is exceeded. The system was applied to the tunnel site, where rockbursts were simulated using nonexplosive demolition agents. The system was able to successfully indicate an oncoming rockburst.
The occurrence of rockbursts in tunnels generally depends on the geological and construction conditions of the tunnel; it is therefore necessary to determine and set an appropriate threshold at which rockbursts occur in tunnels. A threshold for the occurrence of rockbursts is generally set based on the results of laboratory tests. It is difficult to set such a threshold because the conditions may vary between the laboratory tests and the tunnel sites. In this study, the goal is to establish a numerical analysis model that can estimate the threshold of rockbursts accurately, while adequately reflecting the geological and construction conditions at each tunnel site. Additionally, the goal is to develop a system that warns tunnel workers of the imminent danger of a rockburst. The test plan of the entire study is shown in Figure 1. In this paper, Steps A1-A2 and B1-B2 are discussed.
