New algorithm for modelling land surface subsidence due to the rock mass drainage
Grant from the National Science Centre
No. 2014/13/N/ST10/02845

The Goal

Artificial Intelligence methods are applied in modelling phenomena in which either the multiplicity of factors or their subjective evaluation prevent formulation of a strict algorithm allowing for unambiguous mathematical description of the phenomenon. The goal of project is development of an algorithm for modelling the drainage displacement surface based on artificial neural networks.

The Methodology

The data subjected to analysis will include current and archival mining and geological data from selected survey regions, among others, levelling data from the benchmark network, data from piezometric holes, boreholes and geological maps. Assessment of the precision and reliability of the acquired information will be necessary in order to maintain the quality of the planned numerical analyses.

The data acquired and initially prepared in this way will be used as an input for selected AI methods, such as MLP, RBF and GRNN networks, the SVM technique with a radial kernel function, recursive networks and hybrid methods.

After the final selection of the network with statistical methods will be devoted to studying the features of the drainage displacement field and development of those changes over time.

The Results

The algorithm developed will open a new research field for studying the impact of hydrogeological changes on mountain formation movements. It will help determine the development of the displacement field over time, in the context of the developing mining operations. In the period of intense drainage-related surface movements connected with mining operations and winding-up of mines, the research will help us obtain better knowledge of the deformation phenomenon in the drainage displacement field.

Project Schedule

--- Stages 0 - Establishment of project website

--- Stages 1 - Gathering data from a few areas of existing underground and open-pit mining facilities

--- Stages 2 - Analysis of the reliability, coherence and precision of the obtained data

--- Stages 3 - Testing the AI methods with respect to their usefulness for describing displacement fields within the drained mountain formation

--- Stages 4 - Comparative analyses of the obtained results using statistical methods

--- Stages 5 - Studying the properties of the displacement field resulting from mining drainage