Study of Longwall Powered Support Pressures Under Static and Dynamic Conditions

Introduction Powered supports in longwall mines essentially operate under two conditions: 1. Static (non-working) 2. Dynamic (working) There is a variety of static conditions which may exist in a mine. These may include a temporary stoppage of work for a few minutes or stoppage of work for days because of strikes or lack of demand for coal. Studies on pressure behavior were made during brief stoppages of work (few hours) as well as a long stoppage of work (two weeks). In dynamic conditions, the pressure variation on supports is expected to follow a certain pattern. However, in many cases that does not happen. Pressure-time characteristics of each leg of the supports were studied. This paper discusses the research findings of the studies made on powered support pressure behavior under both static and dynamic conditions. Studies on Pressure Variation in Static Conditions It has been noted (Brenner et a1.,1975) that the pressure distribution on mechanized longwall faces is random in nature and changes with time and position along the face. Figure 1 and Figure 2 show the distribution of support pressures for the front left leg of supports in two different mines. In both the cases it can be seen that the variation of the support pressures along the face is indeed absolutely random. However, the nature of randomness in both the cases is very different. The data in Figure 1 (Mine 1) is much more random than the data in Figure 2 (Mine 2). This can also be seen from the standard deviation of the data (81 26 Pa versus 4000 pa) . By comparing the results of ten other sets of data the same conclusions can be arrived at. Since it was found that in Mine 1, the pressure on support legs was extremely random, batches of ten supports each were taken and the mean pressure plotted. This mean pressure line is shown in Figure 1. It was expected that this would smooth out the variation due to individual props and reveal a better picture of the support pressure distribution. For the case cited, the standard deviation is lowered to 5043 Pa from 8126 Pa which indicates that, on an average, the supports tend to take loads more uniformly, however for individual support legs the load on the supports is very random. The significance of this is that, if the supports are selected on the basis of theoretical load calculations alone (based on the roof block concept or some other concept), only the mean value of the support load required will be known. In the case of a pressure distribution with a high standard deviation, some supports selected on the basis of mean value will fail, while others will be supporting low loads. Hence, for proper selection of support capacity, it is not only necessary to know the average load but also the expected nature of the distribution. The only reason why the supports are supporting the roof in Mine 1 well is the fact that the yield load is above the highest level of load attained. If Mine 2, which has a much higher average load, had the same sort of standard deviation, the loads on the individual legs could easily exceed the support load. For example, if the pressure distribution followed a normal curve and the mean pressure was 36,700 Pa with a standard deviation of 9000 Pa, it is possible that 67% of the supports will be within the pressure range of 36,700 plus or minus 9,000 Pa, or 22,700 and 45,700 Pa. Many of the supports could then exceed the yield load of 40,000 Pa. It is clear then that the mean load is an inadequate criterion for estimating support resistance. The nature of the problem is statistical and such an analysis should therefore be made as far as pressure data are concerned. Statistical Analysis of Pressure Data. In general, in longwall mining and other geo-mechanics studies, it is extremely difficult to specify the exact nature of the distribution of a variable. Under these circumstances, probability