Most of the empirical methods are based on the prior knowledge of quantity Smax, the maximum possible subsidence or S, the, maximum subsidence which normally occurs at or near the centre of the mined-out area. The subsidence profile and other associated parameters along any line in the subsidence trough are predicted, in most cases, as a function of either of these quantities (Smax or S) and the distance of the point at which subsidence is to be calculated either from a point vertically above the centre of the excavation or from the point of half subsidence which is near the ribside. This paper describes a method of calculating the value of Smax due to critical or super-critical area, the value of S, occuring at or near the centre of a sub-critical area. The method utilizes an empirical formula developed for Indian coal mines.
Subsidence prediction methods can be broadly classified as empirical and mechanistic. The empirical methods are based on experience gained through numerous field measurements whereas the mechanistic methods are based on the rheology of subsiding materials and their reaction to changing mining geometry. The empirical methods are quick, simple to use and need only such data which are easily available for calculating the subsidence. But these methods are too localised in nature and need measurements and study of the required parameters afresh for their application to a new area. Mechanistic, methods on the other hand, are more fundamental in approach but the real problem remains again of determining and analyzing the physico-mechanical properties of overburden on large scale, which are expensive and many times beyond the means of mine operators.
Empirical method by NCB (1975) makes use of mine geometry only in the prediction of subsidence from nomograms developed by them. Certain influence function methods make use of influence coefficients and sometimes one or two more parameters by way of complementary functions (Sutherland and Munson, 1984).
Subsidence data from 103 mine workings in India, collected and published by Central Mining Research Station, Dhanbad, were used for present studies. The data pertains to 43 bord-and-pillar and 14 longwall workings with caving and 22 bord-and-pillar and 24 longwall workings having sand stowing as goaf support. The data available for studies in these mine workings consist of the following.
The method when applied to 103 cases of mine workings gave the results given in Table 2. Since the magnitude of the movements were small in majority of cases, the assessment of accuracy of prediction as the percentage of measured values may be misleading. For example if the predicted and measured values are 50mm and. 16mm respectively, the discrepancy in terms of percentage of measured value is 203% while as the absolute discrepancy is only 34mm. Therefore the magnitude and not the percentage has been considered for assessing the accuracy. The cases where predicted and measured values agree· within 15 cm were considered as satisfactory.