PSI - Issue 32

A.A. Baryakh et al. / Procedia Structural Integrity 32 (2021) 17–25 Baryakh A.A / Structural Integrity Procedia 00 (2021) 000 – 000

18 2

1. Introduction The exploitation ofpotash salt deposits is associated with the need to protect mines from freshwater breakthrough. As a rule, to ensure the safety of the waterblocking pillar that separates the water-bearing horizons from the mined out space in deposits of water-soluble ores, a pillar mining system is used to support the overlying stratum with inter-chamber pillars. Despite the application of protection measures, emergency breakthroughs of freshwater into mines frequently occur in world mining operations, often leading to losses [Shiman,(1992);Prugger et al.,(1991);Baryakh et al., (2013)]. Advanced monitoring tools and computation methods make it possible to identify areas that are potentially dangerous, in terms of discontinuities in waterblocking strata. As a rule, their identification and spatial localisation occurs during periods when it is not possible to implement mountain protection measures (e.g. filling of a mined-out area or creating softening zones, etc.) [Baryakh et al.,(2019)]. Thus, this is the first time that a method for preventing freshwater breakthrough into mines has been proposed and it is at the design stage. The essence of the method is as follows: a hazardous area is isolated by constructing bulkheads overlapping the permanent mine openings, its mined-out space is then filled with relatively saturated brines. Since the freshwater density is lower than the density of brines, even if the continuity of the waterblocking stratum is disrupted, then the freshwater will not flow into the mined-out space, thus, it will not lead to massive dissolution of the salt rocks. At the same time, the brines pumped into the goaf are not completely saturated with NaCl and KCl. They will already be saturated from the process of mine filling. This will cause a certain degradation of the inter-chamber pillars and additional deformations of the undermined mass, which can lead to the formation of water- соnducting cracks and the flow of freshwater into the mine with all the negative consequences that follow. Two possible scenarios were analysed to eliminate a potentially dangerous area. The first scenario considered a planned (controlled) flooding of the site at a flow rate of 600 m 3 /h; the water salination (KCl) was 47.70 g/dm 3 . The second scenario considered the hypothetical possibility of fresh groundwater breakthrough during the planned flooding of the site. The predicted flow rate of the fresh groundwater inflow from the overly saline stratum into the mined-out space was estimated at 1700 m 3 /h, with a water salination (KCl) of 7 g/dm 3 . 2. Estimation of the Degradation of Pillars due to the Dissolution of Salt Rocks Analysis of the dissolution of the inter-chamber pillars was carried out on the basis of the numerical filtration migration model, implemented using the PM8 software package [Langevin et al.,(2013)]. The software was designed to solve a wide range of hydrogeological problems and was aimed at predicting the flow of groundwater in multilayered systems within the assumptions of the continuity of the medium and the laminar flow mode. It allows solving problems in stationary and non-stationary settings and simulates both pressured and non-pressured filtration, as well as the migration of a dissolved substance in the groundwater. At the same time, the numerical filtration model, created on the basis of the ModFlow program, assumes the observance of the condition of medium continuity and cannot adequately describe water movement in an open channel, such as mine workings. In this regard, the similarity of the observed and simulated process of brine movement in the mined- out space was achieved by setting high values of the medium’s permeability. During the calibration of the numerical model, we tested various calculation options with filtration coefficients from 1  10 5 to 1  10 8 m/day. There was no fundamental difference in the flow velocity or in the flooding rate of the site. The fact that there was no relationship can be explained by the fact that both of these variables (with such high filtration coefficients) are almost unambiguously determined by the flow rate of the incoming water at the source (i.e. at the location of the brine injection or the freshwater breakthrough). As a result, in the calculations for the non flooded space of the model area, the initial filtration coefficient was taken to be equal to 1  10 5 m/day. The density effect cannot explicitly be taken into account within the ModFlow software. Therefore, its influence was simulated by a decrease in the filtration coefficients in the currently flooded zones, down to 3  10 3 m/day. The average porosity value was determined for each design block, based on the total void volume of the mined out space. The coefficient of elastic capacity in the already-flooded blocks of the model was set equal to the compressibility coef ficient of water: 5·10 – 10 m – 1 .