Yessengossov, Zhangir2022-07-282022-07-282022-04-01http://nur.nu.edu.kz/handle/123456789/6553Seismicity is a crucial factor that should be considered in underground mines, it occurs due to natural earthquakes or due to human induced activity, primary reasons might be excavation of an orebody, construction of haulage roads and tunnels, and mined out zones. From seismicity phenomena our focus will be on voids and fractures that occur because of a mining activity. Mining activity leads to the occurrence of rockbursts, which is a sudden release of energy in the rock mass. Even though this research area was studied for decades, there is no reasonable solution on these issues, since rockbursts defy conventional explanation. One of the proposed solutions is a use of microseismic monitoring systems that helps to forecast rockbursts and mitigate their negative impact on mining activity. Because underground mining environment with voids and fractures constantly changes makes this issue more complex. Moreover, there are some examples of successful implementation of seismic monitoring systems and 3D velocity models. Also, voids occur due to backfilling of mined out zones, where strength and composition of a backfill may degrade with time. So, considering those facts, the research includes laboratory models in form of synthetic sample cubes made up of concrete with different curing times and different cement concentrations, allowing to capture data from different samples. The main idea behind this study is to measure seismic wave velocities by using acoustic emission systems (AE). This system allows us to record seismic wave velocities passing through voids, to know wave propagation patterns inside sample cubes. The result of this research compares seismic wave velocity against sample cube dimensions, presence of holes, and concentration of a backfill. Furthermore, this research proposes utilization of artificial neural networks (ANN) as tool that may enhance rockburst mitigation in the future.enAttribution-NonCommercial-ShareAlike 3.0 United StatesType of access: Gated Accessartificial neural networksANNBachelor's thesis