The 10th International Conference on Energy, Resources, Environment and Sustainable Development (ERESD 2024)

As one of the most serious disasters in deep coal mining, rock bursts can be prevented by destressing boreholes. However, as coal mine is featured by high crustal stress and changeable mechanical properties of surrounding rock, there will be drill tools sticking accidents caused by borehole collapse during pressure relief drilling, disturbing safe production and drilling efficiency. Given the gradual drill tools sticking accident caused by drill cuttings plugging, this paper establishes a sticking prediction model based on long short-term memory (LSTM). Firstly, feature extraction is done on the sticking data to obtain its sticking features. Secondly, feature selection is carried out on the extracted sticking features. Finally, the sticking prediction model is constructed based on LSTM. The experimental results show that the proposed prediction model can live up to the demands for sticking forewarning.

feature extraction; feature selection; long short-term memory; drill tools sticking prediction