| Summary: | A research report submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in partial fulfillment of the requirements for the degree of Master of Science in Engineering.
Johannesburg, October 2017 === The gold reserves in South Africa have been mined for decades, depleting all the easily accessible reserves. In pursuing the deeper reserves South African mining industry has for many years led the development of mining and particularly rock engineering. Various design criteria and tools have been developed and used by South African rock engineers in different mining environments.
It must also be understood that these criteria were developed decades ago in different mining environments compared to where mining is currently taking place. In using these design criteria one needs to look at the relevance of such criteria and question if they are still applicable or if new criteria are required. Scheepers et.al, (2012) reviewed the design criteria used in designing ultra-deep narrow reef stopes in the West Wits and identified that there was no clear correlation between the design criteria used and the seismicity which is the highest FOG risk in ultra-deep mines. They then decided to use modelled elastic closure as design parameter which can be correlated to seismicity.
This report details an investigation into the correlation of the modelled elastic closure to the estimated closure from underground and how modelled closure can be adjusted to better reflect the anticipated closure underground. The investigation was conducted using underground observations and stoping width estimations using installed timber support and numerical modelling results (MAP3D).
Before correlating the modelled closure and the estimated closure, it was critical to understand the basis of the work done by (Scheepers et.al, 2012) in correlating the modelled closure to seismic hazard. McGarr, (1976) introduced the concept of correlating seismic energy to volume changes in stope. However this correlation was on the basis that the closure in the stope is only as a result of seismic failure. This was the basis of work done by (Scheepers et.al, 2012) in correlating volume change due to seismicity (seismic potency) to modelled closure.
It must be understood that (Scheepers et.al, 2012) aim was not for the modelled closure to reflect underground closure, however was to give an indication of the anticipated seismic activity relative to closure. This report further looks at what would the underground closure be relative to the modelled closure which has been used as a design parameter against seismicity.
This report showed no correlation between the 0.27m modelled closure determined by Scheepers et.al, (2012) for Mponeng mine to the estimated closure. Through (Scheepers et.al, 2012) work, it was also shown that the correlation of potency to modelled closure was only in the first 10000m2 of mining a new raise line. Seismic potency is highly dependent on the seismic moment of a seismic event and the larger the event, the larger the seismic potency without any consideration to the mining layout.
The elastic modelled closure was found to be on average only 55.3% of the estimated closure. The MAP3D model only considered the elastic properties of the rock and did not take into account any discontinuities or non-homogeneity in the rock mass, hence the large difference to the measured closure. It is important to note that seismic potency and elastic closure modelled do not take into account critical factors that contribute to both rock mass deformation and seismicity in deep mines. More work is required to gain a better understanding on the correlation of rock mass deformation in ultra-deep mines to seismicity.
Of importance from the research is to acknowledge that the use of modelled elastic closure should always be supported with a good understanding of the actual rock mass behaviour. The elastic properties used in numerical modelling programs could be varied in such a way that the elastic modelling results can closely depict the actual rock mass behaviour in terms of closure. Accurate estimation of closure would be useful in the design of support systems and mining layouts in ensuring the stability of excavations for the required periods.
Closure can be estimated by conducting underground measurements and calculated by running numerical modelling programs. Better correlations between the two results would be possible once the elastic properties used in a model are varied until the results obtained from the model are similar to the underground measurements.
The inclusion of the backfill material into the elastic model has significant influence on the resultant closure. This was shown by varying the stoping width used in the model. In a pure elastic model without backfill the stoping width has no influence on the resultant modelled closure as it is evident in the elastic closure formula by (Malan, 2003) which does not take into account stoping width. Varying the poison’s ration has very little influence of the modelled closure while the adjustments to the young’s modulus has a significant influence to the modelled closure. === MT 2018
|
|---|
