Utility tunnel is increasingly applied for underground space development, and its health monitoring becomes attractively important. The acoustic emission (AE) technology is often used for damage monitoring of small-scale specimens made of concrete or rock in the laboratory. In this study, the AE technology is adopted to instrument a novel prefabricated utility tunnel to provide insights into the leakage behavior of the structure during the sealing test. Filtering is conducted on AE data to eliminate the influence of noise interference, especially for data in the pressurization stage. The traditional ring-down count and AE energy are determined from eigenvalue analysis of AE events. It is found that these conventional calculations can be well related to the pattern of internal water pressure, and the Felicity effect is found in the loading process. Furthermore, time series analysis is carried out for each specific pressurization or pressure-holding stage during the sealing test, in which the b-value (an important mean to study the characteristics of AE signal) is evaluated. Results show that the b-value can also be used as a prediction basis for pre-alarming monitoring before the occurrence of seepage failure for utility tunnel.