Chaotic time series prediction of surrounding rock deformation in mining based on grey system theory.

To achieve accurate deformation prediction and assess the temporal stability of surrounding rock in underground mines, this study utilizes the G-P algorithm to determine the optimal embedding dimension and time delay for the reconstructed phase space of the deformation time series. Analyzing the dynamic evolution of the nearest phase point and employing grey prediction theory in the phase space reconstruction, we predict the time series of surrounding rock deformation. Results indicate that the optimal embedding dimension and time delay for the deformation time series are 3. The larger the average distance of the nearest phase point, the more pronounced the chaotic characteristics of perimeter rock deformation, signifying a complex intrinsic dynamical system. The grey theory proves effective in short-term, accurate deformation prediction for perimeter rock. Additionally, the identified stability of perimeter rock based on neighboring phase points and deformation prediction results offers a novel approach for evaluating surrounding rock stability in underground mining processes. [ABSTRACT FROM AUTHOR]