Project

New algorithm for modelling land surface subsidence due to the rock mass drainage

Goal: The pursuit of mining operations disturbs the equivalence of groundwaters, which triggers the comprimation process in compressible layers. In underground mines, intermediate impacts, connected with drainage, sum up with so-called direct mining impacts. Total displacements on the land surface are recorded using surveyed field measurement. Drainage changes are often neglected, or even directly skipped, in analyses of surface changes due to their relatively small values (of the dm order) and the long occurrence time.

The whole deformation process in a mountain formation subject to drainage is an extremely complex issue, and its course is conditional upon a number of factors. The presently applied methods for forecasting displacements and deformations following from mountain formation drainage are based on models that require the knowledge of a considerable number of parameters and detailed knowledge of the geological structure of the mountain formation.

In the proposed research has decided to employ a new approach to this issue called Artificial Intelligence.

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.

http://home.agh.edu.pl/~wwitkow/NCN/en/index.html

Date: 20 February 2015 - 19 February 2017

Updates
0 new
0
Recommendations
0 new
0
Followers
0 new
12
Reads
0 new
101

Project log

Ryszard Hejmanowski
added 2 research items
The paper presents a computer program called SubCom v1.0 for determining mathematical model parameters of compaction layers in areas of oil, gas or groundwater extraction. A stochastic model based on the influence function was used to model compaction and subsidence. Estimation of the model parameters was based on solving the inverse problem. Two model parameters were determined: the compaction coefficient Cm of reservoir rocks, and the parameter tgβ, which indirectly describes the mechanical properties of the overburden. The calculations were performed on leveling measurements of land subsidence, as well as on the geometry of the compaction layer and pressure changes in aquifers. The estimation of model parameters allows the prediction of surface deformations due to planned fluid extraction. An algorithm with a graphical user interface was implemented in the Scilab environment. The use of SubCom v1.0 is presented using the case of an underground hard coal mine. Water drainage from rock mass accompanying coal extraction resulted in compaction of the aquifer, which in turn led to additional surface subsidence. As a result, a subsidence trough occurred with a maximum subsidence of 0.56 m.
The presented research aimed to evaluate the spatio-temporal distribution of ground movements caused by groundwater head changes induced by mining. The research was carried out in the area of one of the copper ore and anhydrite mines in Poland. To determine ground movements, classical surveying results and the persistent scatter Satellite Radar Interferometry (PSInSAR) method were applied. The mining operation triggered significant subsidence, reaching 1.4 m in the years 1944-2015. However, subsidence caused by groundwater pumping was about 0.3 m. After mine closure, an ongoing groundwater rebound was observed. Hence, land uplift occurred, reaching no more than 29 mm/y. The main part of the investigation concerned developing a novel method for uplift prediction. Therefore, an attempt was made to comparatively analyze the dynamics of ground movements correlated with the mine life and hydrogeological condition. These analyses allowed the time factor for the modeling of land uplift to be determined. The investigation also revealed that in the next six years, the uplift will reach up to 12 mm/y. The developed methodology could be applied in any post-mining area where groundwater-rebound-related uplift is observed.
Ryszard Hejmanowski
added a research item
Oil and gas development from underground reservoirs disturbs original rock mass balance. The tending of the rock mass to achieve a new, even only temporary balance is manifested in the movements of the ground surface. Movements can affect ground infrastructure like offshore platforms, pipelines and buildings. For increasing the efficiency of the preventive actions a'priori precisely prediction of the subsidence is necessary. By the prediction of surface subsidence changes of pore pressure in time due to exploitation and geometry of the reservoir have to be taken into account. The prediction method based on the influence function of Knothe will be presented in the paper. Some applications according to the oil and natural gas developments will be discussed
Wojciech T Witkowski
added a research item
Based on the previous studies conducted by the authors, a new approach was proposed, namely the tools of artificial intelligence. One of neural networks is a multilayer perceptron network (MLP), which has already found applications in many fields of science. Sequentially, a series of calculations was made for different MLP neural network configuration and the best of them was selected. Mean square error (MSE) and the correlation coefficient R were adopted as the selection criterion for the optimal network. The obtained results were characterized with a considerable dispersion. With an increase in the amount of hidden neurons, the MSE of the network increased while the correlation coefficient R decreased. Similar conclusions were drawn for the network with a small number of hidden neurons. The analysis allowed to select a network composed of 24 neurons as the best one for the issue under question. The obtained final answers of artificial neural network were presented in a histogram as differences between the calculated and expected value.
Ryszard Hejmanowski
added a research item
Hydrocarbon production under certain geological conditions of these deposits can cause surface subsidence and deformation of the terrain surface. Such deformations appear as subsidence troughs of considerable range and the magnitude of the subsidence depending on the total thickness of the reservoir, compaction properties of reservoir and on the number of other factors. In the past there have been widely recognized magnitudes of the subsidence up to 9 meters. The stress zones in the subsidence trough may affect the buildings and surface structures. However there have been well known some cases of destroyed boreholes or pipelinesbelonging to thegas or oil mine.Therefore there is a requirement to analyze the possibility of occurrence unfavorable phenomenon on the ground surface, to monitor surface deformations during production and to protect surface infrastructure located in the range of mining influences.In the paper the issue of surface subsidence caused by hydrocarbon production has been presented. The cause - effect relationship between the compaction of thereservoir rock and the subsidence of surface area has been assumed. The predictionmodel base on the influence functionand on the superposition of elementary influences.For the purpose of building damage protection a newmodel of risk assessment has been developed. Thismodel base on the elements of fuzzy logicallows to incorporate in the analysis the quantitative and qualitative factors that contribute to the risk of building damage. Use of the fuzzy logic made it possible to obtain onevalue which clearlydiscriminate the risk of buildings damage. However, risk analyzes of damage to the large number of buildings has been required additional tools. The spatial analysishas been made by using GIS. The subjects of the paperhave been illustrated with a practical example.
Wojciech T Witkowski
added 4 project references
Wojciech T Witkowski
added a project goal
The pursuit of mining operations disturbs the equivalence of groundwaters, which triggers the comprimation process in compressible layers. In underground mines, intermediate impacts, connected with drainage, sum up with so-called direct mining impacts. Total displacements on the land surface are recorded using surveyed field measurement. Drainage changes are often neglected, or even directly skipped, in analyses of surface changes due to their relatively small values (of the dm order) and the long occurrence time.
The whole deformation process in a mountain formation subject to drainage is an extremely complex issue, and its course is conditional upon a number of factors. The presently applied methods for forecasting displacements and deformations following from mountain formation drainage are based on models that require the knowledge of a considerable number of parameters and detailed knowledge of the geological structure of the mountain formation.
In the proposed research has decided to employ a new approach to this issue called Artificial Intelligence.
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.